RFCC Process Technology Manual

UOP Fluid Catalytic Cracking Process Process Technology Manual

ORPIC Sohar, Oman September 2012

– LIMITED DISTRIBUTION – This material is UOP LLC technical information of a confidential nature for use only by personnel within your organization requiring the information. The material shall not be reproduced in any manner or distributed for any purpose whatsoever except by written permission of UOP LLC and except as authorized under agreements with UOP LLC.

157048 Table of Contents Page 1

FCC PROCESS TECHNOLOGY

TABLE OF CONTENTS I.

INTRODUCTION

II.

PROCESS FLOW Reactor Regenerator Main Column Gas Concentration and Recovery

III.

PROCESS CONTROL Reactor Regenerator Main Column Gas Concentration

IV.

EQUIPMENT Process Equipment and Its Use Metallurgical Corrosion

V.

FLUIDIZED SOLIDS Theory Applications to Fluid Catalytic Cracking

VI.

CATALYST History Modern FCC Catalysts Time and Temperature Effects Poisons Catalyst Management Catalyst Properties and Testing

VII.

PROCESS VARIABLES Reactor and Regenerator Process Variables Feedstock

VIII.

PROCESS CALCULATIONS FCC Flow Corrections and Mass Balance Liquid Product Cutpoint Corrections Reactor and Regenerator Heat Balance FCC Unit Mechanical Summaries Additional FCC Unit Calculations

157048 Table of Contents Page 2

IX.

FEED AND PRODUCT TREATING Feed Treating Product Treating – Reasons and Methods

X.

ANALYTICAL METHODS Minimum Sample Size Typical Sampling Schedule Outline of FCCU Laboratory Methods

XI.

PROCEDURES Refractory Dryout Startup Shutdown Emergencies Catalyst Handling FCC Unit Evaluation

XII.

SAFETY General Additional Safety Precautions for Entering Vessels High Temperature Problems Chemical Hazards

XIII.

ENVIRONMENTAL Emissions Sources and Solutions

157048 Introduction Page 1

Introduction UOP Company History For more than 80 years, UOP has been one of the world’s leading licensors of new and innovative technology. Today, UOP continues in this role with 30 offices on 4 continents and 9 manufacturing facilities worldwide. UOP currently licenses and designs more than 60 different processes and has a total of 5,500 units licensed worldwide. For the last 50 years, the fluid catalytic cracking (FCC) process has been an important and successful part of UOP's licensing activities.

The Early Years UOP was founded in 1914 as the National Hydrocarbon Company on the strength of patent rights developed from the pioneering work of Jesse A. Dubbs, a California inventor. The company was financed by a noted Chicagoan, J. Ogden Armour. In 1915, the company name was changed to Universal Oil Products Company. From the beginning, the goal of the company was to develop and commercialize technology for license to the petroleum refining industry. Under the direction of C. P. (Carbon Petroleum) Dubbs, son of Jesse Dubbs, research and development work continued at the company's small site near Independence, Kansas, where the famous Dubbs Thermal Cracking process was successfully demonstrated in 1919. The then-revolutionary process became the foundation of UOP's rapid growth and its early worldwide recognition by the industry. The early period of growth was ably directed by its president, Hiram J. Halle, and by Dr. Gustav Egloff, one of the world’s leading petroleum chemists. In 1931, UOP established its headquarters in Chicago and its research laboratories in nearby Riverside, Illinois. That same year the ownership of UOP passed to a consortium of its major licensees, led by Shell and Standard Oil of California.


During this stage, the company benefited immensely by the addition to its research staff of Prof. Vladimir Ipatieff, a famous Russian scientist known internationally for his work in high-pressure catalysis. His contributions in catalytic chemistry gave UOP a position of leadership in the development of catalysis as applied to petroleum processing. The first project of Ipatieff and his research team was catalytic polymerization. Other eminent scientists were also attracted to UOP’s research center in Riverside during this period. With the outbreak of World War II, UOP scientists and engineers focused their knowledge and talents on developing new catalytic processes, notably alkylation that helped meet wartime energy requirements, especially for aviation fuel. UOP also cooperated with other companies to develop the FCC process. In 1944, the owners of UOP divested themselves of their holdings in the company, and UOP’s stock was placed in trust. The American Chemical Society was named as the beneficiary. Thus, the Petroleum Research Fund was created with the understanding that income from the trust was to be used for advanced scientific education and fundamental research in the petroleum field. In spite of some financial and legal setbacks suffered by UOP during this period, strong management succeeded in steering the company back to its original course: taking creative research from concept to commercial reality. UOP was recognized as a company employing the world’s most knowledgeable scientific and technical personnel, who understood petroleum refining and the need for improved processing methods and techniques. In 1949, UOP's research staff developed a radically different refining process that used a catalyst containing platinum. Called the Platforming™ process, it revolutionized the art of reforming to produce gasoline with substantially improved octane number. The process was also instrumental in making benzene available in a quality and quantity never before realized on a commercial scale. With the Platforming™ process and other innovative processes, UOP became a vital contributor to the emergence and growth of the petrochemical industry.


In the early 1950s, UOP also began to manufacture its own proprietary catalysts and a variety of refining chemicals at a newly constructed plant in Shreveport, Louisiana. Later UOP built manufacturing plants at McCook, Illinois; Brimsdown, U.K.; and other locations. In 1952, UOP moved its headquarters and engineering activities to Des Plaines, a suburb of Chicago. Soon after, the construction of a new research center at the same location was begun.

The Recent Era In 1959, UOP assumed its fourth different corporate form when it was sold to the public for the first time in its history. As a publicly owned company, UOP entered a new era marked by growth and diversification. The 1960s saw UOP grow from essentially a process-licensing company to a diversified corporation through many acquisitions and mergers with other companies. By 1975, UOP Inc. included more than 20 different divisions involved in such areas as aerospace and automotive technology. During the 1960s and 1970s, UOP's tradition of innovative process development and commercialization continued with the licensing of the first Sorbex™ simulated moving-bed countercurrent adsorption process in 1961 and the introduction of UOP's CCR Platforming™ process early in the 1970s. In 1975, Signal Companies Inc. acquired 50.5% of UOP and in 1978 acquired the remaining 49.5%, making UOP a wholly owned subsidiary of the company. When the Signal Companies merged with Allied Corporation in 1985, UOP Inc. became a subsidiary of Allied-Signal Inc. As the result of reorganizations and restructuring by its parent companies during the 1980s, UOP’s business scope was refocused on the development and licensing of process technology and the marketing of products associated with its licensing activities. Of the 20 different divisions, only the Process Division and UOP Management Services remain in the present UOP.


In 1988, Allied-Signal entered into an agreement with Union Carbide Corporation that resulted in the creation of a unique joint venture company called simply UOP. The new UOP combined the resources of Allied-Signal’s UOP Inc. with the Catalysts, Adsorbents and Process Systems (CAPS) Division of Union Carbide. The joint venture brought together in synergistic union the strong R&D traditions of both companies. The joint venture now contains the new materials R&D of the CAPS Union Carbide researchers and the scale-up and commercialization skills of UOP research. In addition, the joint venture brings together the commercial experience and worldwide marketing presence of both partners. The result is unprecedented growth for UOP and the development of valuable new technologies, products, and services for its customers. Table 1 summarizes some of the historical highlights of UOP as a process technology company.

Table 1 UOP's History 1914

National Hydrocarbon Company formed to hold Jesse Dubbs patents for a process to recover heavy oil from water

1915

Name changed to Universal Oil Products Company -patents for Dubbs cracking process issued

1921

Dubbs continuous cracking process commercialized

1930

Ipatieff joins UOP beginning a wave of new process developments: alkylation, catalytic polymerization, C4 isomerization

1941

FCC technology developed

1949

Platforming™ introduced, many aromatics processes followed

Late 1950s 1961 Early 1970s

Hydrocracking introduced First Sorbex™ unit licensed CCR Platforming™ introduced

1988

UOP merged with the EP&P and CAPS groups of Union Carbide

1995

UOP acquires the Unocal hydroprocessing business


In the last 20 years, UOP has developed and commercialized a variety of new and innovative processes for the refining and petrochemical industry including the Penex™, Molex™, BenSat™, Oleflex™, Ethermax™, Merox™, Styro-Plus™, Alkylene™, Isal™, Isomar™ and Detal™processes. UOP transfers this technology to its clients through its licensing activity. In the technology transfer process, UOP licenses technology; assists in the planning, design, engineering and commissioning of new installations; provides management services and advises on the efficient performance of processing facilities throughout the world. Behind the successful performance record of UOP is a highly qualified and strong team continuously at work on ideas and projects. The scientific disciplines are strongly represented in UOP's team of personnel. UOP has about 4,000 employees worldwide. With a wide array of highly specialized talents, UOP offers its clients the complete capability necessary in meeting the demands of today, and the challenges of the future. UOP licenses or maintains a position of technical expertise for more than 60 different processes in the petroleum and petrochemical industry. Approximately 175 process units are licensed yearly, and to date UOP has licensed more than 5,500 individual process units and provided technical know-how in designs for more than 1,000 additional non-licensed units. UOP presently holds in excess of 9,000 unexpired patents. UOP's worldwide licensing activities are supported by a network of offices and representatives. UOP is centered in Des Plaines, Illinois, and has a district office in Houston. UOP Limited, a 100% UOP owned subsidiary for operations in Europe, Africa and the Middle East, has its main European office in Guildford (near London) and district offices in New Delhi, Jakarta, Jeddah, Beijing and Moscow. UOP Asia Pacific, located in Tokyo, is an affiliate company of UOP for the licensing of UOP processes in Japan and certain other areas in the Far East and Southeast Asia.


UOP has catalyst manufacturing facilities in the United States and in Europe. UOP Asia Pacific operates a catalyst plant in Japan. The international scope of UOP activities is evidenced by the fact that process units have been designed for installation in more than 80 countries around the world. UOP activities related to these installations have ranged from preparation of engineering designs for single process units to extensive planning studies involving market analyses, feasibility and optimization studies, designs for entire grassroots refineries (both process units and offsites), and complete plant commissioning services. The services provided by UOP for these units includes plant design, inspection, commissioning, performance testing, and training of refinery operating personnel. Since 1955, UOP has provided, or is providing, engineering designs for more than 125 grassroots refineries and petrochemical complexes. UOP also provided design specifications for all offsite equipment for many of these installations.

Historical Origins of FCC Technology The advent of the petroleum refining industry can be traced to the rapidly increasing demand for kerosene to fuel kerosene lamps for lighting in the latter half of the 1800s. With the invention of electric lighting and the automobile in the early 1900s, the high value product of petroleum refining shifted from kerosene to gasoline. The increasing demand for gasoline soon outstripped the availability of straight-run gasoline from crude oil distillation. This shortage of gasoline provided the impetus for the development of technologies to increase the gasoline yield from a barrel of crude oil. Table 2 shows a summary of the progression of cracking technology which has led to the FCC process as we know it today.


Table 2 Historical Origins of Fluid Catalytic Cracking 1913 - 1936

Thermal cracking  Burton thermal cracking process (1913)  Dubbs thermal cracking process (1915)  Current use – visbreaking, coking

1936 - 1941

Fixed-bed catalytic cracking  Houdry Process Company (1931) - Multiple reactors -- cyclic process (1937) - Silica-alumina catalyst (acid-activated clay)

1941 - 1955

Moving-bed catalytic cracking  Thermofor catalytic cracking (TCC) developed by Socony-Vacuum (Mobil)  Houdryform catalytic cracking - Continuous process - Macro-catalyst, moving bed

1942 - Present Fluid catalytic cracking (FCC)  Joint development (1938) - Continuous process - Micro-catalyst, fluidized bed

Thermal Cracking The first thermal conversion process was the Burton process first practiced commercially in 1913 by Standard Oil of Indiana. In the original Burton process, oil was exposed batch-wise to high temperature at elevated pressure to achieve thermal conversion to lighter products. Because of the batch nature of the Burton process, commercial units contained a large number of individual cracking stills in order to achieve acceptable daily throughputs.


Following the commercialization of the Burton process, the Dubbs thermal cracking process was developed and patented in 1915 (UOP). The Dubbs process was a continuous process for the thermal conversion of oil to lighter products at elevated temperature and pressure. The Dubbs process was widely used in refineries through the 1920s and into the 1930s. Thermal cracking processes continue to be used in refining today. Examples of currently used thermal processes are visbreaking and various forms of coking.

Fixed-Bed Catalytic Cracking In the mid 1920s, a French mechanical engineer and racecar enthusiast named Eugene Houdry became interested in gasoline quality. After the trial and error screening of hundreds of catalyst formulations, Houdry found that acid-activated clay (silica and alumina) was an effective catalyst for cracking heavy oil to lighter products, particularly high octane gasoline. In 1931, Houdry, in partnership with Socony-Vacuum (now Mobil), founded the Houdry Process Company to develop Houdry's fixed-bed catalytic cracking process. The Houdry catalytic cracking was a cyclic process which typically used four timephased reactors, each of which cycled through a sequence of steps outlined below: 1. 2. 3. 4.

Hot heavy oil is cracked by contact with a fixed bed of catalyst. The reactor is purged to remove hydrocarbon. Coke deposited on the catalyst is burned off with air. The combustion gases are purged from the reactor and the reactor is ready to begin the next cracking cycle.

A number of technical innovations were required to make the Houdry cracking process successful. Among these were the development of automatic valves and the use of control algorithms to control the reaction-regeneration cycles. Many of the innovations associated with the commercialization of the Houdry cracking


process were considered revolutionary in the field of process engineering at the time they were first introduced. The Houdry catalytic cracking process was first commercialized at the Sun-Marcus Hook refinery in 1937. The Houdry process was technically attractive to refiners and by 1940, 14 commercial Houdry units were in operation. Interest in the Houdry process declined after 1941 because of further advances in catalytic cracking technology.

Moving-Bed Catalytic Cracking The next advance in catalytic cracking was the development of a continuous moving-bed cracking process. The Thermofor Catalytic Cracking (TCC) and Houdryform Catalytic Cracking (HCC) processes were developed in parallel in the 1940s and early 1950s. Both processes used a similar concept and had approximately equal success. In the TCC process, the catalyst pellets continuously move through the reactor to the regeneration vessel and are then returned to the reactor. The key to the TCC process was the Thermofor kiln used to regenerate the spent catalyst (the kiln had been originally developed to burn coke off of Fuller’s earth used to filter lube oils). In the TCC process, regenerated catalyst flows by gravity from a surge vessel elevated above the reactor, into the reactor vessel where the catalyst contacts hot oil and the cracking reactions take place. The air environment of the catalyst surge vessel is buffered from the hydrocarbon environment of the reactor by steam injected into the catalyst transfer line. Both the hydrocarbon vapors and catalyst flow down through the reactor to a lower section where the cracked products exit the reactor through separation pipes. The spent catalyst continues to flow by gravity down through a steam stripping zone into the regeneration kiln where coke is burned off the spent catalyst with air. The steam stripping zone also serves to provide a barrier between air in the regenerator and hydrocarbon in the reactor. In early TCC units, the hot regenerated catalyst pellets were mechanically conveyed


back up to the catalyst surge vessel by bucket elevators. Later units employed pneumatic air lift systems to transfer the regenerated catalyst back up to the surge vessel. Socony-Vacuum was the principle developer of the TCC process and the first semi commercial unit started up at the Paulsboro refinery in 1941. The TCC units were licensed and operated by Socony-Vacuum and others from 1941 to about 1955 when the TCC gave way to the more versatile FCC process developed in the during the late 1930s and early 1940s. A few TCC units still continue to operate today.

The FCC Process Early development of the FCC process took place late in the 1930s. A number of motivations were behind the development of the FCC process. Among these were the high fees required to license the Houdry cracking process, the diffusion and heat transfer limitations associated with both the Houdry fixed-bed process and the TCC process (both used large size catalyst pellets), and the increasing demand for high octane aviation gasoline brought on by World War II. Initial FCC process development efforts were led by Standard Oil of New Jersey (now Exxon) in association with two researchers from the Massachusetts Institute of Technology, Warren Lewis and Edwin Gilliland (consultants to Standard-NJ). Lewis and Gilliland had found that under the proper aeration conditions, finely divided solid particles (powders) could flow through pipes and in many respects act similarly to liquids. This was the advent of fluidization. The use of finely divided cracking catalyst offered a means of overcoming the diffusion and heat transfer limitations encountered with the large size catalyst pellets used in the earlier catalytic cracking processes. In 1938, Standard-NJ and some of the other major oil companies, as well as M. W. Kellogg Co. and Universal Oil Products (UOP), formed Catalytic Research Associates (CRA) to jointly develop a fluidized catalytic cracking technology. The first commercial-scale (13,000 BPD) FCC unit, designated the Model I, started up at


Standard-NJ's Baton Rogue refinery in May 1942. Two other Model I FCC’s were designed but were not built as the improved Model II FCC design came very quickly. When Standard-NJ announced the construction and imminent startup of the first FCC Model I, they also announced that Universal Oil Products (UOP) and M. W. Kellogg would be designing and licensing the new FCC technology. In the threeyear period between 1942 and 1945, 34 new FCC units came on stream in the refineries of 20 different oil companies. The installed capacity of these new FCC units was over 500,000 BPD. Thirteen of these units were licensed from UOP. Following the commercialization of the Model I and Model II FCC units within the CRA partnership, the FCC unit design and development diverged with the partner companies largely going their separate ways with regard to future FCC technology development and commercialization.

UOP and Fluid Catalytic Cracking During the 1940s, military requirements resulted in widespread commercialization when UOP designed about 40% of the 34 units that were built and operated. Following this period, UOP was in the forefront with commercialization of the "stacked" FCC unit design which featured a low-pressure reactor stacked directly above a higher pressure regenerator. The stacked design not only met the economic needs of smaller refiners, it was a major step toward shifting the cracking reaction from the dense phase of the catalyst bed to the dilute phase of the riser. In the mid-1950s, UOP introduced the "straight-riser" or side-by-side design. In this unit, the regenerator was located near ground level with the reactor to the side in an elevated position. Regenerated catalyst, fresh feed and recycle were directed to the reactor by means of a long, straight riser located directly below the reactor. Compared to earlier designs, product yields and selectivity were substantially improved. A major breakthrough in catalyst technology occurred in the mid-1960s with the development of the zeolitic catalysts. These catalysts demonstrated vastly superior activity, gasoline selectivity and stability characteristics compared to the amorphous


silica-alumina catalysts then in use. The availability of the zeolitic catalysts served as the basis for most of the process innovations that have developed in recent years. The continuing sequence of advances in both catalyst activity and process design culminated in the most significant concept to date in the field of the FCC process – the achievement of transport-phase cracking entirely in the riser, or all-riser cracking. The key to all-riser cracking is the design of a system that initiates a plug-flow reaction and then stops the cracking reaction at the optimum yield of desired products. UOP commercialized a new design based on this concept in 1971. This design was also applied to existing unit revamps. Commercial results confirmed the expected advantages of the system compared to the older designs. The quick quench design provided the desired high selectivity to gasoline, reduced coke yield, and a reduction of secondary cracking of desired products to lighter, less valuable material. The next major improvement in the FCC technology was the development of catalysts and regenerator systems for the complete internal combustion of carbon monoxide (CO) to carbon dioxide (CO2). In 1973, an existing UOP unit was revamped to include a new combustor concept in regeneration technology to achieve direct conversion of CO within the unit. This was followed by the start-up in 1974 of a new FCC unit specifically designed to incorporate the combustor regenerator technology. This development in regenerator design and operating technique resulted in reduced coke yields, lower CO emissions which satisfy environmental standards and higher circulating catalyst activity that improved product distribution and quality. Table 3 summarizes some of the major achievements in UOP's FCC process technology development and commercialization.


Table 3 Milestones in FCC Technology 1942

UOP begins licensing the FCC Process

1945

13 Units licensed by UOP

1947

UOP commercializes stacked unit design  Economical for small refiners  50 Stacked designs over 10-year period

1950s

UOP commercializes side-by-side design  Straight riser  Better suited for larger units  Riser extension and termination (more reaction in riser)

1973

First complete combustion regenerator

1983

First two-stage regenerator with external dense-phase cooling for highly contaminated resid feed commissioned

1983

First elevated distributors commissioned

1991 - 1995

Newest generations of highly contained riser termination devices commercialized (VDS™ and VSS™)

1994

First Optimix™ feed distributor commissioned

1994

First MSCC™ unit commissioned

2006

First AF™ Packing commissioned

Recent Developments Advances in riser termination devices occurred at a rapid rate in the 1980s to the mid 1990s. Early riser termination devices such as the open Tee resulted in very long residence times for the hydrocarbon products in the reactor vessel. This extended residence time resulted in nonselective thermal cracking and secondary catalytic cracking reactions. Recent improvements have resulted in better containment of the hydrocarbon vapor to the riser and therefore lower post riser residence time. This reduced delta coke and dry gas and improved gasoline selectivity. Early versions of these high containment riser terminations included the vented riser and


SCSS (suspended catalyst solids separation) devices. In 1991, the first VDS™ (vortex disengager stripper) was commissioned. This technology further minimized the post-riser residence time resulting in further improvements in product yields. In 1995, the first VSS™ (vortex separation system) was commissioned. Improvements in feed distribution systems also occurred rapidly in the late 1980s and 1990s. Elevated, radially oriented feed distributors minimize nonselective thermal cracking reactions by providing more uniform feed/catalyst contacting with less back mixing than the earlier wye feed distributors. Acceleration zone technology which pre-accelerates the catalyst into a uniform, moderate density flow pattern for optimum oil penetration and uniform catalyst/oil contacting further improved the performance of the elevated feed distributors. The first UOP elevated feed distributors were commissioned in 1983. Developments in spray nozzle technology resulted in the Optimix™ feed distributor which has a smaller, more uniform oil droplet size and a spray pattern that distributes the oil uniformly over the entire riser area for superior catalyst/oil contacting and performance. The first Optimix™ feed distributor was commissioned in 1994. Since then, the number of refiners using Optimix™ feed distributors has grown to over 80. Resid processing in FCC units began in the mid-1970s. During this same period, reactor temperatures were being increased to maximize gasoline octane. The need for higher conversion, combined with the desire to process residue feeds significantly increased coke yields and ultimately limited the FCC regenerator capacity. The RCC®, or Reduced Crude Conversion, process was developed jointly by UOP and Ashland Oil in the late 1970s to address residue processing. It is an extension of UOP's FCC design experience that incorporates many innovations and modifications from the UOP-Ashland Oil development program. In addition to cold-flow modeling work, a large-scale pilot plant was constructed at Ashland's Catlettsburg, Kentucky refinery. Testing in this 200 BPSD plant examined processing variables and new mechanical designs on a wide range of residual feedstocks. In 1983, Ashland commissioned a 40,000 BPSD RCC unit at the Catlettsburg refinery.


Several major innovations from the pilot plant testing and first commercial design at Ashland have become the foundation of UOP's technical offering for catalytic cracking of residue feedstocks, including the following. • Acceleration zone and feed distribution system • Higher containment riser termination devices for quick disengagement • Two-stage catalyst regeneration • Catalyst cooler Since 1983, eight grass-root RCC units licensed by UOP have been commissioned. In addition, resid feedstocks are being processed in more than 30 existing UOP FCC units. In present times, the distinction between a gasoil FCC unit and a resid FCC unit has blurred to the point where most modern FCC units are capable of processing some level of resid. The term RFCC is used by UOP today to designate a new unit utilizing a 2 stage regenerator designed for the specific intent of processing resid feeds. Table 4 shows a brief summary of resid processing and UOP's activity in the area of resid processing.


Table 4 Resid Processing Milestones 1940s

Resid component added to feed

1950s

Resid processing diminishes

1975

Resid processing regains attractiveness  Market conditions favor increased efficiency in gasoline production  Technology and catalyst advances increase resid processing potential  UOP units begin processing resid/gasoil blends

1976

UOP and Ashland Oil Cooperation  Research and development for reduced crude conversion  Semi-commercial demonstration

1983

First RCC unit commissioned

1984 - 2006

8 New RCC units operating ->30 Units processing resid

Commercial Experience Since commercialization of the FCC process, UOP has licensed more than 210 units, or over 50% of all non-captive installations. More than 140 of these units continue to operate throughout the world. The superior technology and operational reliability built into UOP FCC units are some of the reasons why 58 refineries worldwide have licensed new UOP FCC units since 1980, which is more than all other licensors combined during this period. UOP's commercial activity in the FCC/RCC/MSCC™ processes since 1980 is as follows:     

63 40 330 180 30+

New units licensed New units commissioned Revamps Major revamps Units processing resid with UOP technology


Revamp activity is of equal importance in demonstrating technical expertise. In the period 1980-1998, UOP performed more than 330 unit revamps that encompassed virtually every major section of the FCC unit. This activity is vital to UOP's continuing advances in both process and design engineering. The depth of both grass-root and revamp experience gives UOP great capability to respond to the changing needs of the industry.

FCC Process Description The FCC process converts heavy crude oil fractions into lighter, more valuable hydrocarbon products at high temperature and moderate pressure in the presence of a finely divided silica/alumina based catalyst. In the course of cracking large hydrocarbon molecules into smaller molecules, a non-volatile carbonaceous material, commonly referred to as coke, is deposited on the catalyst. The coke laid down on the catalyst acts to deactivate the catalytic cracking activity of the catalyst by blocking access to the active catalytic sites. In order to regenerate the catalytic activity of the catalyst, the coke deposited on the catalyst is burned off with air in the regenerator vessel. One of important advantages of the FCC process is the ability of the catalyst to flow easily between the reactor and the regenerator when fluidized with an appropriate vapor phase. In FCC units, the vapor phase on the reactor side is vaporized hydrocarbon and steam, while on the regenerator side the fluidization media is air and combustion gasses. In this way, fluidization permits hot regenerated catalyst to contact fresh feed; the hot catalyst vaporizes the liquid feed and catalytically cracks the vaporized feed to form lighter hydrocarbon products. After the gaseous hydrocarbons are separated from the spent catalyst, the hydrocarbon vapor is cooled and then fractionated into the desired product streams. The separated spent catalyst flows via steam fluidization from the reactor to the regenerator vessel where the coke is burned off the catalyst to restore its activity. In the course of burning the coke a large amount of heat is liberated. Most of this heat of combustion is absorbed by the regenerated catalyst and is carried back to reactor by the fluidized regenerated catalyst to supply the heat required to drive the reaction side of the


process. The ability to continuously circulate fluidized catalyst between the reactor and the regenerator allows the FCC unit to operate efficiently as a continuous process. The FCC units are large-scale processes and unit throughputs are typically in the range of about 10,000 to 130,000 barrels per day. This corresponds to catalyst circulation rates of around 7 to 130 tons per minute. The largest commercial FCC unit in operation was designed at 130,000 BPSD, pushed to ~184,000 BPSD, and in 2005 was revamped to a nominal 200,000 BPSD with a catalyst circulation rate in excess of 170 metric tons per minute. These large circulation rates of hot, abrasive catalyst constitute a very significant challenge to the mechanical integrity of the reactor, the regenerator and their associated internal equipment. Thus, mechanical design considerations are critical to the ultimate success of an FCC unit as a prominent refinery process unit. The main features of an FCC unit are:     



Catalytic process Mechanical process Cracks heavy molecules to lighter ones Pressure: 15-45 psig (1-3 kg/cm2g) Temperature: Reactor: 915-1050F (490-565C) Regenerator: 1200-1450F (650-790C) Reaction and regeneration sections intimately linked by heat balance and catalyst circulation

FCC Process Feedstocks FCC units process heavy oil from a variety of variety of refinery flow schemes. Generally, the feed comes from either the refinery crude unit or vacuum unit and constitutes the fraction of the crude boiling in the range of 650 to 1000+°F (350 to 550+°C). There may be additional feed preparation units upstream of the FCC unit such as a hydrotreater or deasphalter. Figure 1 shows a schematic diagram of the possible refinery flows providing feed to an FCC unit.


In addition, the FCC units commonly process heavy fractions from other conversion units as part of the combined FCC feed blend. Examples of these types of streams are coker gasoil and hydrocracker fractionator bottoms. The types of heavy hydrocarbon streams that are commonly charged to an FCC unit are:        

Atmospheric gasoil Vacuum gasoil Atmospheric resid Coker gasoil Demetallized oil Hydroprocessed gasoil Hydroprocessed resid Lube oil extracts


FCC Products The products obtained from the FCC unit are light hydrocarbon gases (C2-) which are normally used within the refinery as fuel gas, light olefins and paraffins (C3’s and C4’s) also referred to as LPG, gasoline, LCO and clarified oil commonly referred to as main column bottoms. In addition, flue gas is generated from the burning of coke in the regenerator. Heat is recovered from the flue gas and is used to make steam and in some cases power is also recovered from the flue gas in the form of electricity via a power recovery expander coupled to a motor/generator. Products produced from an FCC unit are:       

Light gas Light olefins LPG Light paraffins Gasoline Light cycle oil Main Column Bottoms Coke (burned in unit as fuel)



Most of the FCC product streams undergo further processing before leaving the refinery as marketable products. Figure 2 shows typical routes for the FCC product steams going to further processing and ultimately to blending into the refinery product pools.


Figure 2 Typical Use of FCC Products Flue Gas Reactor & Regen.

C3/C4 Paraffins Fuel Gas

Alkylation

LPG Merox Gasoline Gasoline Merox LCO CLO

Alkylate Gasoline Pool

C3/C4 Splitter Main Column & Gas Con

LPG Pool

MTBE Distillate Hydrotreater

Diesel Pool Heavy Fuel Oil Pool

The light liquid products from the FCC process are LPG and gasoline. The LPG from an FCC unit is highly olefinic and has become an increasingly valuable stream for further processing in the present movement toward reformulated gasoline and as petrochemical unit feedstocks. The FCC olefins are an important feedstock for the production of MTBE and alkylate as gasoline blending components and for the production of polypropylene. The FCC gasoline generally has good octane properties (90-95 RON and 80-83 MON) and may make up 30 vol-% or more of the refinery gasoline pool. Some typical characteristics of light FCC products from highconversion operations (VGO Feed, 1.0 wt% sulfur) are:




LPG: 500 - 1500 wppm total sulfur 30 - 40 vol-% C3 olefins 34 - 45 vol-% total C4 olefins



Gasoline: C5 - 380F 90% point (193C 90% point) 92 - 94 RONC 0.1 - 0.2 wt-% sulfur 30 - 40 vol-% olefins 25 - 35 vol-% aromatics 0.5 - 1.0 vol-% benzene

The heavy liquid products from an FCC unit are normally LCO and clarified oil. The LCO product is normally used as a blending component in the diesel pool and/or in the heavy fuel oil pool. It is becoming increasingly common for LCO destined for diesel blending to be hydrotreated first for sulfur reduction. Clarified oil is usually blended off to the heavy fuel oil pool. In some cases, the FCC unit clarified oil is used in coker feed, for asphalt production or sold as feed for carbon black production. Some characteristics of heavy FCC products from high conversion operations (VGO Feed, 1.0 wt% sulfur) are: 

Light cycle oil: 600F 90% point (316C 90% point) 20 - 26 cetane index 1 – 1.5 wt-% sulfur 75 - 80 vol-% aromatics 3 - 3.5 cSt @ 122F (50C)



Clarified slurry oil: 2 - 3 wt-% sulfur 9 - 13 cSt @ 210F (100C)



Source: Middle Eastern light gasoil



Abbreviations and Definitions ABD

average bulk density

ACFM

actual cubic feet per minute

activity

conversion of oil by test catalyst compared to standard reference feed often referred to as MAT activity

adjusted

conversion or yields reported as corrected to standard product cutpoints

afterburning

burning of CO above the dense bed in the dilute phase or flue gas, characterized by temperature increase

AGO

atmospheric gasoil

Al

aluminum

Al2O3

alumina

APS

average particle size

AR

atmospheric column resid

ash

non-combustible particles remaining after burning of a main column bottoms sample

as produced

conversion or yields reported as percent of fresh feed at the actual product rates not adjusted to standard product cut points

ASTM

American Society for Testing and Materials

ßo

Coefficient of thermal expansion at 60°F, (1/°F)

behind in burning

insufficient coke combustion in regenerator, characterized by increased coke production in reactor and dark grey regenerated catalyst (high carbon on regenerated catalyst)

BPD

barrels per day


BS & W

bottoms sediment and water, normally reported in vol-%

C1, C2

methane, ethane, ...

C3=

olefin (propylene)

caustic

sodium hydroxide

CCR

catalyst circulation rate

CFR

combined feed ratio (volume of fresh feed plus recycle, divided by volume of fresh feed)

CN-

cyanide ion

CO2/CO

mole ratio of carbon dioxide to carbon monoxide, indicates degree of partial combustion

cold regenerator

operation in conventional controlled regenerator afterburning mode of regeneration

conversion

measure of the rate of gasoil disappearance (or conversion) from feed to products defined as

COS

carbonyl sulfide

CRC

carbon on regenerated catalyst

CSO

clarified slurry oil

Cu

copper

DA, DS, DG

reactor or regenerator purges using air, steam or gas, respectively

P, DP

pressure drop or pressure difference between two points

dry gas

gas from sponge absorber (usually refers to C2-)

EP

end point of distillation

F-1

research octane number (RON)

F-2

motor octane number (MON)


Fe

iron

Fines

catalyst particles less than 20 microns diameter

Fm

feed metals factor

Gasoline efficiency

ratio of liquid vol-% gasoline to vol-% conversion, indicates selectivity to produce gasoline

GC, GLC

gas chromatography, gas/liquid chromatography

Gb

Fluid gravity at base temperature (60°F)

Gf

fluid gravity at flowing temperature

gpm

gallons per minute

H2/C1

ratio of moles hydrogen to moles of methane

H2S

hydrogen sulfide

HC

hydrocarbon

HCN

heavy cat naphtha product drawn from the side of the main column

HCO

heavy cycle oil

HPS

high pressure separator

H

enthalpy (heat) difference

IBP

initial boiling point of distillation

K (UOP K)

measure of paraffinicity or aromaticity of hydrocarbon

lb/Bbl (#/Bbl)

pounds per barrel

LCO

light cycle oil

LV-%

liquid volume percent

M

prefix for thousand

MC

main column


MCB

main column bottoms product

MON

motor octane number

MW

molecular weight

N (or N2)

nitrogen

Na

sodium

NH3

ammonia

Ni

nickel

NOx

nitrogen oxides

O (or O2)

oxygen

ppm

parts per million

Pf

Pressure at flowing conditions (absolute)

recycle

normally refers to heavy oil from main column which has already passed through the reactor that is returned with the fresh feed to the reactor, this could also refer to light material such as LCO or gasoline; a stream which returns to its source.

RE (or Re2O3)

rare earth (or rare earth oxide)

Rg

regenerator

RON

research octane number

RSH

mercaptan sulfur

RVP

Reid vapor pressure

Rx

reactor

SA

surface area

SCF/Bbl (SCFB)

standard cubic feet per barrel of fresh feed


SCFD

standard cubic feet per day

selectivity

preferential towards specified goal or species

severity

combines different factors to give an overall qualitative measurement of extent or difficulty in cracking and regeneration

Si

silicon

Si2O3

silica

sintering

closure of catalyst pores

SOX

sulfur oxides

spillback

gas recycle, may also refer to liquid recycle

SS

stainless steel, also second stage

Tf

temperature at flowing conditions (absolute)

V

vanadium

VGO

vacuum gasoil

vol-%

volume percent

wt-%

weight percent


UOP P&I Diagram Abbreviations AR

Analysis Recorder

ARC

Analysis Recording Controller

DR

Specific Gravity Recorder

FA

Flow Alarm

FE

Orifice Flange Assembly

FFRC

Flow (ratio) Recording Controller

Fl

Flow Indicator

FIC

Flow Indicator Controller

FIF

Flow Indicator Flow Type

FQI

Flow Meter Displacement Type

FR

Flow Recorder

FRA

Flow Recording Alarm

FRC

Flow Recording Controller

FRCF

Flow Recording Controller Float Type

FRCQI

Flow Recording Controller Integrator

FRQI

Flow Recorder Integrator

FRQIA

Flow Recorder Integrator Alarm

HC

Hand Control

II

Current Indicator

LA

Level Alarm

LC

Level Controller

LG-B

Gage Glass Boiler Type—Visible Length Shown


LG-R

Gage Glass Reflex Type—Visible Length Shown

LG-RLT

Gage Glass Reflex Type Visible Length Shown—Low Temperature

LG-T

Gage Glass Through View Type Visible Length Shown

LG-TK

Gage Glass Through View Type Visible Length Shown—KEL-F

LG-TLT

Gage Glass Through View Type Visible Length Shown—Low Temperature

Ll

Level Indicator

LIA

Level Indicating Alarm

LIC

Level Indicating Controller

LR

Level Recorder

LRA

Level Recording Alarm

LRC

Level Recording Controller

PA

Pressure Alarm

PC

Pressure Controller

PDC

Pressure Differential Controller

PDI

Pressure Differential Indicator

PDIC

Pressure Differential Indicating Controller

PDR

Pressure Differential Recorder

PDRA

Pressure Differential Recording Alarm

PDRC

Pressure Differential Recording Controller

PDRCA

Pressure Differential Recording Controller Alarm


PI

Pressure Indicator

PIA

Pressure Indicating Alarm

PIC

Pressure Indicating Controller

PR

Pressure Recorder

PRA

Pressure Recording Alarm

PRC

Pressure Recording Controller

SI

Speed Indicator

SR

Speed Recorder

TA

Temperature Alarm

TC

Temperature Controller

TDR

Temperature Differential Recorder

TDRA

Temperature Differential Recording Alarm

TDRC

Temperature Differential Recording Controller

TI

Temperature Indicator

TIC

Temperature Indicating Controller

TIX

Temperature Indicator Skin

TR

Temperature Recorder

TRA

Temperature Recording Alarm

TRC

Temperature Recording Controller

TRX

Temperature Recorder Skin

TW

Thermowell

Zl

Valve Position Indicator


When Instruments Are Designated with an Alarm H

Indicates High

HH

Indicates High-High, typically in association with an Emergency Shutdown (ESD) system trip point

L

Indicates Low

LL

Indicates Low-Low, typically in association with an Emergency Shutdown (ESD) system trip point

157048 Process Flow Page 1

PROCESS FLOW INTRODUCTION The modern Fluid Catalytic Cracking unit is a large and complex process for cracking heavy gas oil to lighter hydrocarbons. FCC has largely replaced the old thermal crackers because it is a more efficient process, i.e. more production of valuable products at a lower overall cost by using catalyst and heat instead of simply heat. In its simplest form, the process consists of a reactor, a catalyst regenerator, and product separation. This is shown in Figure 1. Catalyst circulation is continuous, at very large mass flow rates. For this reason, the reactor and regenerator are usually discussed as one section. The product separation is usually divided into its low and high pressure components, i.e. the main column section, and the gas concentration and recovery section.

Figure 1: Fluid Catalytic Cracking Process Flue Gas

Regenerator

Air

Catalyst Transfer Lines

Products Reactor

Raw Oil

Product Separation


Reactor-Regenerator This is the heart of the process, where the heavy feed is cracked. The reaction products range from oil which is heavier than the charge to a light fuel gas. The catalyst is continuously regenerated by burning off the coke deposited during the cracking reaction. This provides a large measure of the heat required for the process.

Main Column The main column cools the reactor vapors and begins the separation process. A heavy naphtha fraction and light and heavy fuel oils (LCO and CLO) come off the tower as products; gasoline and lighter materials leave the top of the tower together and are cooled and separated further into product streams in the gas concentration section.

Gas Concentration and Recovery This section separates the main column overhead into gasoline, liquefied petroleum gas, and fuel gas streams. The composition of each stream is controlled for maximum product value. Figure 2 shows a slightly more detailed schematic of an FCC unit.

Raw Oil

Catalyst Section

MCB HCO

Steam

Light Naphtha

Fuel Gas

Flue Gas

FCC-PF002

LPG

Gas Con Section

Flue Gas Cooler BFW

Heavy Naphtha

LCO

Main Column Section

Power Recovery Section

Figure 2 FCC Block Flow Diagram



PROCESS FLOW DESCRIPTION Reactor-Regenerator The FCC process was developed in the early 1940's. A number of companies participated in the early stages of the work, so most of the early units were virtually identical. The first design, the Model I, was installed at only three refineries and quickly replaced by a more successful Model ll. Thirty-one of these were built, thirteen designed by UOP. Figures 3 and 4 show the configurations of the Model I and the Model II FCC’s, respectively. The Model II units had double slide valves and long standpipes, which were a prime source of operating problems due to loss of catalyst fluidization in the standpipes. The raw oil charge passed through a dense bed of fluidized catalyst in the reactor vessel; however, evidence indicated that a large part of the desired cracking was occurring in the transfer line where the hydrocarbon first contacted the catalyst. The early units used low activity catalysts by today's standards, starting with natural clay and later progressing to amorphous synthetic silica/alumina catalysts. Large amounts of heavy oil were recycled back to the reactor in order to obtain the desired conversion levels of 40-60 vol-%. UOP introduced a major departure from the Model ll design in 1947. The regenerator riser was eliminated and the regeneration air was injected directly into the regenerator dense bed. Single slide valves and the more compact design cut construction costs. An important feature of the design was a long reactor riser, which was a major advantage as FCC technology advanced toward entirely riser cracking. The UOP Stacked FCC design proved to be quite popular and UOP designed about 50 Stacked FCC units. Figure 5 shows the typical arrangement of the UOP Stacked FCC unit.


Figure 3: Model I FCC Cottrell Precipitator

Flue Gas

Catalyst Fines

Products to Main Column

Cyclones

Hoppers Air or Steam

Steam

Standpipes

Regenerator Reactor

Steam Regenerator Riser Water

Catalyst Recycle Cooler

Raw Oil Charge

Reactor Riser

Fired Heater FCC-PF003


Figure 4 Down-flow Model II Catalytic Cracking Unit

Flue Gas

Cottrell Precipitator

Waste Heat Boiler

Multicyclones Regenerator Products to Main Column

Reactor

Steam to Stripping Section Raw Oil Charge MCB Recycle

Air

FCC-PF004


Figure 5 UOP Stacked Fluid Catalytic Cracking Unit To Main Column Cyclone

Flue Gas

Reactor

Orific Chamber Spent Catalyst Stripper

Flue Gas Slide Valve

Stripping Steam

Regenerator Spent Catalyst Slide Valve

Regenerated Catalyst Slide Valve Air Slurry Recycle

Raw Oil Charge

HCO Recycle FCC-PF005


The next advance in reactor-regenerator design was the Side-by-Side unit, shown in Figure 6. This design was better for larger units, where stacking the reactor on top of the regenerator became more expensive. The Side-by-Side layout has also been used for many of the new small units. The straight riser showed less erosion than the curved riser of the Stacked unit. Some of the Side-by-Side units were designed with a reactor dense bed. This bed was eliminated with the advent of zeolite cracking catalysts, and the riser was extended within the reactor to minimize thermal and catalytic cracking by reducing vapor residence time in the vessel. Initially, cyclones were installed on the riser to separate the oil and catalyst, but this was not particularly successful due to poor cyclone performance. The riser cyclones were replaced by a "Tee" shaped termination at the top of the riser. The riser cyclone could then be moved over to allow room for the addition of another cyclone at the reactor outlet, providing two stages of cyclone separation. Later advances in riser termination devices concentrated on maximizing hydrocarbon containment or minimizing the post-riser residence time in the reactor shell where non-selective, thermal and secondary cracking reactions occur. Side-by-Side units have won good acceptance by the industry and over 75 UOP designed Side-by-Side FCC units have been built. Zeolite cracking catalysts were developed in 1963 and gradually accepted by the industry over the next ten years. These catalysts proved to be much more active than amorphous catalysts and were ideally suited for the short contact time riser cracker. Conversion levels rose as high as 80% without requiring excessive reactor temperature. Another significant improvement in FCC reactor technology was the use of elevated feed distributors. The older wye feed distributors injected the raw oil charge into a highly back-mixed catalyst flow that resulted in non-uniform oil/catalyst mixing and excessive light gas and coke formation. In newer systems, multiple, radially oriented feed distributors elevated in the riser inject raw oil more uniformly to maximize selectivity to desired products.


New regenerator designs were also developed over the years. The old perforated plate air distributor was changed to a pipe grid for better air distribution. Two stage cyclones replaced single stage cyclones and reduced catalyst losses. The burning of coke in the old regenerators was not complete, i.e., not all the carbon went to CO2, and the flue gas normally contained 6-10 vol-% CO. The unit ran with no excess oxygen. This prevented afterburning in the cyclones and the resultant heat damage to the cyclones. An extra furnace to generate steam, the CO boiler, was added to utilize heat that would otherwise be lost. All of the excess CO in the flue gas could be burned in the CO boiler, but capital costs were high. The obvious solution to this problem was to burn all of the CO to CO2 in the regenerator, where the catalyst can absorb the heat. Although this could be done in a standard “bubbling bed” regenerator, a new, “high efficiency” type regenerator design proved more efficient. In the high efficiency or combustor style regenerator, shown in Figure 7, the air and catalyst is mixed in a fast fluidized environment in the lower part of the regenerator or combustor. The fluidized catalyst is then carried up the combustor riser to the upper regenerator. The fluidization in the combustor provides excellent air/catalyst mixing and heat transfer to maximize coke burning kinetics. The high efficiency regenerators on stream average less than 100 ppm CO, and less than 40 ppm NOx in the flue gas. This design enables refineries to get greater thermal efficiency from the unit while simultaneously meeting more stringent air quality standards.


Figure 6 UOP Side by Side Fluid Catalytic Cracking Unit Rxn Products to Main Column

Reactor Down-Turned Arm

Flue Gas

Flue Gas Slide Valve

Stripping Steam

Bubbling Bed Regenerator

Spent Catalys Slide Valve

Main Distributo

Air

Regenerated Catalys Slide Valve

Wy Section Raw Oil Feed


Figure 7 Modern UOP Side by Side Fluid Catalytic Cracking Unit With High Efficiency Regenerator, Elevated Feed Distributors and Vortex Separation System Riser Termination


The process flow of the reactor and regenerator section of a typical, modern FCC unit with a high efficiency regenerator can be described as follows: Lift steam and/or light hydrocarbon is injected at the base of the riser or Wye to accelerate the catalyst from towards the elevated feed distributors, which are located about 1/3 the way up the riser. The preheated raw oil charge is pumped through the feed distributors and atomized with the addition of steam then injected into the regenerated catalyst stream. The heat from the catalyst and reduced hydrocarbon partial pressure in the riser both act to help vaporizes the oil. The catalyst, oil and steam travel up the riser to a region of lower pressure in the reactor where the cracked hydrocarbon products are separated from the catalyst in the riser termination device and cyclones before going to the main column for initial product separation. During the cracking reaction, a carbonaceous by-product called coke is deposited on the circulating catalyst. This catalyst (referred to as spent catalyst) drops from the reactor disengager and cyclones into the stripping section where a countercurrent flow of steam is used to remove both interstitial and some adsorbed hydrocarbon vapors. The stripped catalyst flows from the reactor stripper through the spent catalyst standpipe to the regenerator, where the coke is continuously burned off. The catalyst flow through the spent catalyst standpipe is controlled to balance the circulating catalyst flow by maintaining a constant catalyst level in the reactor. In the regenerator, the spent catalyst mixes with air and hot regenerated catalyst from the recirculation catalyst standpipe at the base of the combustor. Here the coke deposited during in the reactor is burned off to reactivate the catalyst and provide heat for the net endothermic cracking reactions. The heat of combustion raises the catalyst temperature in the regenerator to a range of 1200°F-1375°F (648°C-746°C). The catalyst and air flow up the combustor riser and separate at a "Tee" shaped head. The flue gas is further "cleaned" of catalyst in the cyclones in the upper regenerator. The recirculation catalyst standpipe returns some of the hot regenerated catalyst to the combustor either on temperature or density control to provide heat for initiation of the carbon burn. The remainder or the regenerated


catalyst flows down the regenerated catalyst standpipe on reactor temperature control to the riser Wye to complete the cycle. The flue gas exits the regenerator through the flue gas slide valves on pressure control to the regenerator. An orifice chamber located downstream of the slide valves acts to reduce the pressure drop and velocity across valves to minimize mechanical deflection of the body and erosion to the internals. Many units have a power recovery unit in place of the slide valve and orifice chamber to recover electrical energy by letting down the high volume, moderate pressure flue gas across a turbo-expander connected to a motor/generator. Finally the sensible heat energy in the flue gas is recovered through steam generation in either a CO boiler or flue gas cooler depending on the mode of operation in the regenerator. Many units also have an electrostatic precipitator or wet gas scrubber to remove catalyst fines from the flue gas before it is discharged to the atmosphere. The reasons and methods for varying the high efficiency regenerator operation will be discussed in more detail later in the PROCESS VARIABLES section. RFCC Regenerator As a result of the crude oil embargoes and oil price rises of the 1970’s, interest in processing heavier feeds in FCC units grew. However, FCC technology at that time could not handle highly contaminated heavy feeds while maintaining a reasonable degree of conversion. In the mid 1970’s, UOP and Ashland Oil Company embarked on a joint development project to develop catalytic cracking technology capable of processing very heavy, highly contaminated feeds, i.e. feeds with high metals and Conradson carbon contents. The result of this development program was the commercialization of the RCCSM (Reduced Crude Conversion) process at Ashland’s Catlettsburg refinery in 1983. The main feature of the RCC unit is a two stage regenerator equipped with a catalyst cooler to remove heat from the regenerator. The upper or first stage regenerator burns approximately 2/3 of the coke from the catalyst in partial combustion mode to limit the heat of combustion and therefore the temperature of


the catalyst. A portion of the partially regenerated catalyst entering the lower or second stage regenerator flows through the catalyst cooler(s) where heat is removed from the catalyst to generate steam. The cooled catalyst and the remainder of the hot catalyst from the first stage regenerator mix in the second stage regenerator where the coke burning is completed under conditions of complete CO combustion in the presence of excess O2. Carbon is burned off the catalyst to low levels in the second stage regenerator at moderate temperature to maximize catalyst activity. The combustion gases from the second stage regenerator pass into the first stage regenerator where the pre-heated excess O2 improves coke burn kinetics, and is completely consumed. The combined flue gas exits through two stages of cyclones in the first stage regenerator and out through a single common flue gas line. The overall mode of combustion for the two stage regenerator is partial burn with the additional benefit that all of the catalyst returning to the reactor is fully regenerated due to the full burn environment of the second stage regenerator. Figure 8 shows the arrangement of the regenerator of an RFCC unit. The reactor is the same as the modern Side by Side unit shown in Figure 7. Figure 9 shows the process flow for a catalyst cooler. Although catalyst coolers are not a new idea for FCC service, past attempts to employ catalyst coolers on FCC’s have been largely unsuccessful from both mechanical and process points of view. UOP’s catalyst cooler represents an improved design developed and refined to provide both mechanical reliability and a wide range of heat removal flexibility. Heat removal varies with the rate of fluidization air injected to the cooler and the catalyst slide valve opening. The operation of the catalyst cooler is as follows; catalyst enters the cooler shell where the tube bundle is immersed in hot fluidized catalyst. Fluidization air is injected at the bottom of the cooler shell to control the fluidization and heat transfer. Annular bayonet type water tubes are used in the tube bundle. Water entering the bundle flows up through the inner tube, flows out the top of the inner tube and down through the annular space between the inner and outer tubes where heat transfer occurs and water is vaporized to steam. In flow-through style coolers cooled catalyst exits the cooler shell through a standpipe and slide valve and is returned to


the regenerator to allow hot catalyst to enter the top of the cooler to maximize the cooler duty. Back-mix type coolers rely only on fluidization and back-mixing to transfer hot catalyst from the regenerator rather than using catalyst flow through a standpipe. Back-mix coolers have a simpler mechanical configuration but can only remove approximately 70% of the heat transfer capable through a flow-through cooler. A large excess of water is circulated through the tubes where heat transfer generates steam to ensure that the tube walls are always wet and cooled. The steam and water mixture returns from the cooler bundle to a steam drum where the steam and water are separated. Water from the drum is circulated back to the cooler and the saturated steam from the steam drum is routed to the refinery steam system.


Figure 8 UOP RFCC Regenerator Process Flow Flue Gas

Spent Catalyst from Reactor

1st Stage Regenerator

Vent Tubes

Catalyst Cooler

First Stage Air

2nd Stage Regenerator

Water/ Steam Water

Recirculation Catalyst Standpipe Regenerated Catalyst to Reactor Second Stage Air


Figure 9 UOP FCC Catalyst Cooler Process Flow Water and Steam Saturated Steam to Superheater Fluffing Air

Makeup BFW Blowdown

Cooled Catalyst Slide Vlave

Circulating Water

FCC-PF009


Main Column The main column is the first step in the separation and recovery of the cracked hydrocarbon vapors from the FCC reactor. The reaction products enter the column at high temperatures, 900-1022°F (480-550°C). The main column is similar to a crude tower, with two important differences: 1) The vapors must be cooled before fractionation can begin, and 2) a large quantity of lighter gas passes overhead with the gasoline. Figure 10 shows the general process flow for an FCC main column. Large quantities of heavy oil are circulated over a series of disc and doughnut trays to cool the vapors and wash down entrained catalyst. The heat removed by the main column bottoms and the heavy cycle oil is used for feed preheat, steam generation, reboiler heat in the rest of the unit, or some combination of the three. The catalyst washed out of the reactor is concentrated in the main column bottoms stream. Most of the bottoms flow is directed through exchangers for heat removal and returned to the disc and doughnut trays. The return line must be free draining to avoid plugging problems with catalyst fines settling in low points. Some of the cooled bottoms material from the steam generators may be returned directly to the bottom of the tower as quench to reduce the temperature of the liquid and minimize coking and fouling in the bottoms system. Figure 11 shows a typical process flow for the main column bottoms pumparound and product circuit. Many older units used a slurry settler to separate and return catalyst fines to the reactor with a slurry stream off the bottom of the settler. The main column bottoms product comes off the top of the settler and is normally called clarified slurry oil. In reactors with two stages of cyclones and in units with modern riser termination devices, the use of slurry settlers has normally been discontinued. Heavy bottoms product comes directly from the main column bottoms circulating stream, as does any slurry recycle to the reactor.


Figure 10 UOP FCC Main Column To Wet Gas Compressor CW

To Sour Water

To Primary Absorber

1 5 6

Gas Concentration Unit

FI 7

19 21

FI 22

Equalizing Line to/from Feed Drum

Steam

26 27 29

30

Heavy Naphtha Product Gas Concentration Unit

Steam

32

Light Cycle Oil Product

33

Flushing Oil

34 35

Gas Concentration Unit

36

Flushing Oil 37

Torch Oil 38

Reactor Vapors

Steam

BFW

Main Column Bottoms Product CW

Raw Oil from Surge Drum

Raw Oil to Reactor


Figure 11 UOP FCC Main Column Bottoms Process Flow

6 Minimum Spillback

MCB Product Circuit Minimum Flow Valve

3

1

E FRC

Main Column

Rx Overhead

E

MCB Product Pumps

E

E

Tempered Water

Quench

Steam

Steam Main Column Bottoms Circulation Pumps

MCB Product

MCB Steam Generators Water

Raw Oil Water

Circulating Bottoms/ Raw Oil Exchanger

Raw Oil

Net Bottoms/ Raw Oil Exchanger


As previously mentioned, most FCC units with modern riser terminations and reactor cyclones do not require the use of a slurry settler and new units currently being designed do not include settlers. If a refiner has strict specifications on the ash content of the main column bottoms product, then more advanced alternate fines removal equipment is usually employed to reduce the catalyst fines to very low levels. The two most common types of catalyst removal equipment used today are the micromesh filter and the electrostatic separator. Cyclonic separation devices have also been used, but are typically limited to smaller capacity installations. A typical micromesh filter system will have 2 or 3 vessels with up to 100 filter elements in each. When multiple filtration vessels are used, each filtration vessel is sized for 100% of the design flow rate. One vessel is typically in filtration mode while another is in backflush mode to remove the filter cake from the elements. When enough catalyst fines have deposited on the filter elements to increase the pressure drop across the filter to a pre set limit, the vessel is taken off line for back flushing. Once the filter vessel is off line and drained the vessel is filled with backflush liquid, either HCO or LCO, and allowed to soak to help dissolve any heavy aromatic compounds on the elements. The top of the vessel is then pressured up with either fuel gas or nitrogen to provide the driving force for a high velocity back flush. The back flush material is collected in a receiver vessel and pumped back to the reactor riser. A typical process flow for the micromesh filtration system is shown in Figure 12. A typical electrostatic slurry oil filtration unit consists of 4-16 skid mounted cylindrical shells (modules) depending on the volume of filtrate to be process. Each module contains a high voltage cylindrical electrode surrounded by conductive glass beads, with a ground rod located in the center of the module assembly. During the separation cycle, the glass beads are ionized in an electrostatic field. As catalyst particles flow between the beads, they are electrostatically collected on the surface of the beads. Each module is sequentially back-flushed while the remaining modules in the system continue the separation. In the backflush cycle, the electrode is de-energized and the beads are fluidized, resulting in a circulating motion up through the center of a 9-inch annular electrode and down the outside. The circulation up the center annulus and down the walls of the module creates a scrubbing action, to mechanically scrub the beads clean. Mechanically scrubbing


the beads as opposed to solvent soaking as with the micro-mesh filters, raw oil feed, or any compatible oil can be used as the backflush medium to an electrostatic filter. The back flush material is directed back to the reactor riser.

Figure 12 Main Column Bottoms Product Filtration System Backwash Gas (N2 or Fuel Gas)

Clean MCB Product to Storage

Filter #1

Filter #2

Filter #3

N2 or Fuel Gas Vent

Back Flush Liquid (LCO/HCO) MCB Product

Receiver Vessel

Catalyst Backwash to Reactor


There are typically three side-cuts withdrawn from the main column, heavy cycle oil (HCO), light cycle oil (LCO), and heavy naphtha (HCN). The refiner may withdraw all three, only two or one, depending on product needs and tower design. On relatively rare occasions, the main column is designed with a fourth side-cut to discretely fractionate a heavy LCO cut and a light LCO cut. The side-cut streams that go out as product are usually stripped to meet flash-point specifications. Pumparound loops from these side-draws are used to heat balance the main column by exchanging heat with the gas concentration unit reboilers, the raw oil charge or boiler feed water. The heat removed in the bottom and side pumparounds determines the amount of reflux in each section of the tower and must be properly balanced for proper column operation. Gasoline and light gases pass up through the main column and leave as vapors. After being cooled and condensed, unstabilized gasoline is pumped back to the top of the column as reflux to control the top temperature in the column. Figures 13, 14, 15 and 16 show typical process flows for the HCO pumparound, the LCO pumparound, the Heavy Naphtha pumparound and the Main Column Overhead system, respectively.


Figure 13 Main Column HCO Pumparound

E

Main Column

LIC

E

Gas Con Unit Debutanizer Reboiler

HCO Internal Reflux (Pumped)

To MCB/Feed Exchanger Outlet (for startup) Filling Line (from feed pump)

E

Heavy Cycle Oil Circulation Pumps

To Feed Surge Drum (normally no flow) To Pump Flushing Oil Supply Header


Figure 14 Main Column LCO Pumparound and Product

FI FIC LIC Steam

FIC LCO Product

BFW CW Preheater

FIC

LCO to Flushing Oil

Debutanizer Feed Exchanger

FIC Stripper Reboiler Rich Oil from Sponge Absorber Lean Oil to Sponge Absorber FCC-PC403


Figure 15 Main Column Heavy Naphtha Pumparound and Product

Main Column C3/C4 Splitter Reboiler

E

E

Circulating Naphtha/ Debutanizer Feed Exchanger

Steam E FRC

Heavy Naphtha Stripper

LCO Stripper

Reflux (Gravity Flow)

LIC

Heavy Naphtha Circulation Pumps Heavy Naphtha Product Pumps Heavy Naphtha Product Cooler HeavyNaphtha to and from NHT Unit Signal from HCN Hydrotreater

CW FRC E

Hydrotreated Heavy Naphtha Product


Figure 16

Main Column Overhead System

FCC/DS-R00-37


The raw oil feed system is included in the main column section for better process efficiency, i.e. to take advantage of the heat from the main column. Feed enters the unit from storage or directly from upstream processes, such as a vacuum tower or a hydrotreater. The latter scheme is more efficient because the feed will not have to be cooled before storage and then reheated flowing into the FCC. The number and type of exchangers used will depend on cost and process factors that will vary with each refinery. Most newer units do not use fired charge heaters. Fired charge heaters have become unpopular due to the increases in fuel costs, operational safety and impact on overall refinery stack emissions. The feed goes directly to the riser after the raw oil/main column bottoms exchanger. Figures 17 and 18 show typical process flow schemes for the FCC feed preheat system without a fired charge heater and with a fired charge heater, respectively.

Figure 17 Feed Preheat Equalizing Line To/From Main Column

Raw Oil Surge Drum

Raw Oil from Crude Unit/ Hydrotreating

To Reactor LCO Product

MCB Product

Circ. MCB

MCB Recycle HCO Recycle FCC-PF401


Figure 18 Feed Preheat with Fired Heater

Equalizing Line To/From Main Column

Raw Oil Surge Drum

Fired Heater Raw Oil from Storage/ Upstream Unit

MCB Product

To Reactor

Circ. MCB

Fuel Gas

FCC-PF401


Gas Concentration and Recovery This section further separates the main column overhead products into stabilized gasoline, LPG and fuel gas. The normal configuration is shown in Figure 18. Unstabilized gasoline from the main column overhead receiver is pumped to the primary absorber, where it is used to adsorb C3’s and C4’s in the gas stream at much higher pressure than the main column. From here the liquid stream goes to the high pressure receiver (separator), then the stripper column, where H2S and C2- are removed. The gasoline off the bottom of the stripper is pressured to the debutanizer for separation of LPG and gasoline and vapor pressure adjustment of the gasoline. The overhead of the debutanizer is olefins rich LPG which is often further processed, for C3 and C4 separation and propylene recovery. The gas from the main column overhead receiver goes first to the wet gas compressor. From here it is pressured to the HPS, the primary absorber, and finally the sponge absorber. Valuable light products such as LPG are removed in the first of two vessels by absorption into the gasoline. The second vessel is the sponge absorber which uses a LCO pumparound from the main column as a final absorption stage before the gas goes out as fuel. Wash water, typically clean condensate, is injected into the inlet to the wet gas compressor interstage condenser. From the interstage receiver it is pumped to the high pressure receiver then to the main column overhead condensers. This water washes out salt forming and corrosive and species such as H2S, NH3, cyanides and phenols. The wash water flow is shown in Figure 19.

D HPR IR PA SA S WGC

HPR

Wash Water to MC OVHD Receiver

IR

Debutanizer High Pressure Receiver Interstage Receiver Primary Absorber Sponge Absorber Stripper Wet Gas Compressor

Legend:

Wash Water

WGC

Gas from MC Overhead Receiver

Gasoline from MC Overhead Receiver

40

P A

9

1

Rich Lean Oil Oil

S A 36

S

1

Typical Gas Concentration Unit Process Flow

Figure 19:

40

D

1

Stabilized FCC Gasoline

HCO

LPG

Fuel Gas


Main Column

Hydrocarbon Water

Condensate

To Sour Water Stripper WGC 1st Stage

Main Column Receiver

Interstage Drum

Wash Water Flow

Figure 20:

WGC 2nd Stage

High Pressure Receiver


157048 Process Control Page 1

PROCESS CONTROL Reactor-Regenerator Control Systems Most of the control systems in a Fluid Catalytic Cracking unit are similar to those used elsewhere in the refinery. Good control of the catalyst circulation through the reactor and regenerator is critical for stable operation. The catalyst circulation control scheme is shown in Figure 1. This figure shows a side by side unit with a bubbling bed regenerator but the catalyst circulation control between the reactor and regenerator is the same on all FCC and RFCC units. The circulation of hot regenerated catalyst from the regenerator to the reactor is controlled to maintain a constant reactor temperature with the regenerated catalyst slide valve. The circulation of spent catalyst from the reactor to the regenerator is controlled to maintain a constant catalyst level in the reactor with the spent catalyst slide valve. The controls on both the spent catalyst and regenerated catalyst slide valves also include a low differential pressure override. If the differential pressure across either slide valve drops to a very low or negative value the override will close the slide valve. This minimizes the possibility of reverse flow in the standpipes, either air entering the reactor or hydrocarbon entering the regenerator, which are hazardous situations. Figure 1 shows signals from the reactor temperature controller and level controllers going to low level selectors (LSS). The low signal selectors also receive signals from the differential pressure controllers on the corresponding slide valve. If the differential pressure across the slide valve is greater than the override setpoint, typically 2 psi (0.14 kg/cm2) the LSS will select the process variable (level or temperature) to control the slide valve. If the slide valve pressure drop falls below the override setpoint the LSS will send that output to the slide valve which will start closing. The low differential pressure override controllers should always be in automatic.


Figure 1 Catalyst Circulation Controls Products to Main Column

TIC PDIC

Reactor

Riser Termination Device

HSS

> LIC

Flue Gas

PIC

Regenerator

LSS

LI

< PDIC

XI (Density)

Spent Catalyst Slide Valve LSS

< PDIC

Raw Oil

Air

Regenerated Catalyst Slide Valve Lift Gas/ Steam

FCC-PC001


For steady control of the catalyst circulation between the reactor and regenerator the differential pressure across the slide valves must be constant. To ensure steady slidy valve DP’s the differential pressure between the reactor and regenerator is controlled with the double disc flue gas slide valves on the outlet of the regenerator. In addition to the slide valves an orifice chamber is also used to take approximately 2/3 of the total flue gas system pressure drop to minimize erosion in the flue gas slide valves. A typical flue gas system without power recovery is shown in Figure 2. The reactor pressure is not controlled directly and floats on the main column overhead pressure. The reactor-regenerator differential pressure controller allows the regenerator pressure to change along with the reactor pressure. Pressure control for units with power recovery is discussed later.

Figure 2 Regenerator Pressure Control/ Flue Gas System HSS >

PIC

PDIC

Signal from Reactor Pressure Tap

Flue Gas

Flue Gas Slide Valves

Orifice Chamber Steam

CO Boiler Air Air

Electrostatic Precipitator

Water

FCC-PC002


Reactor Control (Figure 3) Reactor temperature is controlled by the flow of hot regenerated catalyst as described above. The temperature controller may be located in the reactor vapor line or in the upper vapor space of the reactor vessel depending on the type of riser termination device. The reactor pressure is not directly controlled. Reactor pressure floats on the main column overhead pressure. Thus, the reactor pressure is the sum of the main column overhead receiver pressure plus the pressure drop through the main column and MC overhead condensers plus the pressure drop through the reactor cyclones and reactor vapor line. The reactor catalyst level is controlled by the flow of spent catalyst to the regenerator as described above. Modern reactor designs include a wide range level indicator and a more accurate, narrow range level controller. Also, the density in the spent catalyst stripper is measured to allow compensation of the level indication for actual catalyst density. The level is typically controlled off of the wide range LIC because the signal has less noise but a switch is included in newer unit designs to allow use of the more accurate narrow range LIC if desired. Steam is used to atomize the feed in the elevated Optimix feed distributors (Figure 4). The amount of steam used determines the both the pressure drop and extent of atomization as well as the velocity out of the distributor tip and penetration into the catalyst. During normal operation the steam and oil enter the nozzle separately and are mixed internally near the tip. The steam flow is normally set at design rates, typically1-2 wt% of the design charge rate. During operation at turndown additional steam may be used to maintain the optimal pressure drop across the nozzles (typically 50-75 psig) to ensure adequate atomization is maintained. An alternative that allows maintaining near design pressure drop and atomization at turndown with minimal additional steam is to mix a small amount of steam directly with the oil before it enters the nozzle. This results in a large increase in pressure drop with minimal increase in velocity out of the tip. The oil and steam flows to each nozzle have restriction orifices with pressure drop indicators and globe valves to ensure that the flows to each nozzle are equal.


Figure 3 Reactor Control and Instrumentation PDI TIC LIC

LIC

X

SWITCH

LI (Density Compensated)

XI (Density) XI (Density)

Prestripping Steam

PDIC

< LSS

Stripping Steam

MPS

Fluffing Steam

Spent Catalyst LSS PDIC

Regenerated Catalyst

<

Atomizing Steam For Feed/Steam controls see Figure 4 Raw Oil

FIC Lift Gas FIC

Lift Steam

MPS FIC Start-up/Emergency Steam

FCC-PC003


The feed bypass system is also shown in Figure 4. When a situation requiring a quick shutdown is encountered, a control board mounted switch is activated which trips a solenoid valve controlling the pneumatic signals to the feed bypass valves, causing these valves to move to their failure positions, i.e. the valve in the line to the riser closes and the valve in the bypass line to the main column opens. Normally, the next course of action is to open the emergency steam to the riser to either maintain catalyst circulation if the regenerated catalyst slide valve remains open or to clear the riser of catalyst if the regenerated catalyst slide valve is closed. On units with elevated feed distributors, another important operating variable affecting the yield pattern is the lift zone velocity. The lift zone velocity is a function of the lift steam and/or lift gas flow rates which are controlled on straight flow control. As the lift zone velocity is increased the catalyst density at the point of the feed injection decreases allowing better penetration of the atomized oil droplets into the catalyst. The optimum lift zone velocity is typically in the range of 10 – 15 ft/sec (3 – 4.5 m/sec). Either lift gas, lift steam or a combination of both may be used to achieve the optimum lift zone velocity. Stripping steam flows are also controlled on straight flow control. The optimal total stripping steam rate is typically 1.7 – 2.5 lb/M-lb catalyst circulation but this can vary significantly with depending on the unit design. The stripping steam rate should be changed when any process conditions are changed that result in a significant change in the catalyst circulation rate. The stripping steam rate should also be tested occasionally to ensure that the optimum steam rate is used.


Figure 4 Feed/Atomizing Steam Control

FI

FI FI

To other nozzles

PI

Steam FO

Local PI must be readable from valve PI

Local FI must be readable from valve

FI FI

To other nozzles

Local FI must be readable from valve To other nozzles

NORMALLY NO FLOW HEADER PURGE

HS

Feed Bypass Switch

Raw Oil to Main Column

S Instrument Air

FO Vent

FC

Raw Oil from Preheat

FCC-PC004


Bubbling Bed Regenerator Control In full combustion units without a catalyst cooler the regenerator temperature is not directly controlled and is a function of a number of process variables. In simple terms, the regenerator temperature is a function of delta coke, i.e. the wt% coke on spent catalyst entering the regenerator minus the wt% coke on regenerated catalyst leaving the regenerator. The concept of delta coke is discussed in more detail later in the section of this book covering Process Variables. The regenerator may be operated to burn the coke on catalyst completely to CO2 (complete combustion mode) or may be operated so that some of the coke is burned to CO (partial combustion mode). In units that operate in partial combustion mode the CO2/CO ratio of the flue gas may be controlled to adjust the heat of combustion and therefore adjust the regenerator temperature. The CO2/CO ratio is controlled primarily with the amount of combustion air entering the regenerator. Partial combustion operation is discussed in more detail in the Process Variables Section. In units with a catalyst cooler, the regenerator temperature may be controlled by controlling the amount of steam generated in the cooler. The controls of the catalyst cooler are discussed in more detail later in this chapter. The regenerator catalyst inventory serves as the surge capacity for catalyst in the system and there is no control instrumentation on the regenerator catalyst level. A level indicator is provided to monitor the regenerator catalyst level. The regenerator catalyst level changes with catalyst additions, withdrawals and losses. In most units the level is controlled by intermittent withdrawals of equilibrium catalyst. It is important that the level be maintained above the terminations of the cyclone diplegs and below the level that would cause the catalyst in the cyclone diplegs to back up into the cyclone dustbowl.


The air rate to the regenerator is controlled either to maintain a minimum of excess oxygen in the flue gas (typically 2%) for full combustion operation or to control the CO2/CO ratio and therefore the heat of combustion and regenerator temperatures in partial combustion operation. High Efficiency Regenerator The instrumentation and controls for a FCC unit with a high efficiency, combustor style regenerator is shown in Figure 5. The pressure, regenerated catalyst temperature and combustion air rate for the high efficiency regenerator are the same as the bubbling bed regenerator described above. The difference between the high efficiency regenerator and a conventional "bubbling bed " regenerator is that the regenerator is divided into two sections. The lower section is called the combustor and is where the spent catalyst and air mix and coke combustion occurs. The combustor operates in the fast fluidized regime of fluidization. All the catalyst entering the combustor is transported up the combustor riser into the upper regenerator where the regenerated catalyst disengages from the flue gas. The upper regenerator holds the cyclones, provides volume for the regenerated catalyst to disengage from the flue gas and provides the surge capacity for catalyst in the system. An important feature of the high efficiency regenerator is the recirculating catalyst standpipe and slide valve. The recirculation of hot regenerated catalyst from the upper regenerator to the combustor is important in controlling the coke combustion rate. By controlling the amount of catalyst recirculated, the following parameters are controlled in the combustor: the pre-combustion mixing temperature, the catalyst density, catalyst flux and catalyst residence time. This, in turn, allows the coke combustion rate and catalyst regeneration to be optimized. The recirculating catalyst slide valve is controlled through a low signal selector and a slide valve PDIC, similarly to the other slide valves. On early designs, this slide valve position was set on hand control. In current designs, the recirculation slide valve position is controlled by a temperature or density controller located in the upper section of the combustor. A switch is used to select the input signal to the recirculation slide valve low signal selector.


Figure 5 High Efficiency Regenerator Controls and Instrumentation >

TI's (1 Each Cyclone)

PIC

PDIC Signal from Reactor Pressure Tap

LI

XI (Density)

Recirculating Catalyst Slide Valve

FIC Fluffing Air Regenerated Catalyst to Reactor

PDIC TIC LSS

XIC (Density) Spent Catalyst from Reactor

<

SWITCH

FCC-PC005


In a high efficiency regenerator the dense bed catalyst level in the upper regenerator provides the surge volume for the unit and is not controlled directly except by catalyst additions and withdrawals. A small flow of fluffing air to the upper regenerator on straight flow control is required to ensure proper fluidization and flow into the regenerated and recirculation catalyst standpipes.

RFCC Two Stage Regenerator Control The regenerator control systems for the RFCC unit with a 2 stage regenerator are shown in Figure 6. The reactor control systems are identical to those described above for the conventional reactor-regenerator. The principal difference is that the coke combustion is completed in 2 stages. The upper regenerator, or 1st stage, is a bubbling bed regenerator operating in partial combustion mode without excess oxygen. The catalyst exiting this stage still contains a significant amount of coke which is burned off in the 2nd stage operating in full combustion mode with excess oxygen. This allows the benefits of both partial combustion (lower regenerated catalyst temperature) and full combustion (very low carbon on regenerated catalyst for maximum activity). In the RFCC the catalyst level in the second stage is controlled by the flow of catalyst from the first stage to the second stage through the recirculation catalyst standpipe and slide valve. This slide valve has a differential pressure override which closes the slide valve if the pressure drop across the valve drops to low as described above for the other slide valves. The level in the first stage regenerator is the surge volume for the unit and is typically controlled by periodic withdrawals of equilibrium catalyst. Since RFCC units are designed for heavily contaminated feed stocks one or more catalyst coolers are included in the design and are used to control the temperature of the regenerated catalyst circulated back to the reactor. The total air rate which controls


the CO2/CO ratio in the flue gas and therefore the heat of combustion is also used to adjust the temperature of the regenerated catalyst.

Figure 6 RFCC Regenerator Controls and Instrumentation

LI Spent Catalyst from Reactor XI

Side View

Vent Tubes PDIC

PDIC LIC

First Stage Air LSS

LSS

<

< TIC

Recirculation Catalyst Standpipe

Cooled Catalyst Standpipe

Regenerated Catalyst to Reactor Second Stage Air

Second Stage Air

FCC-PF006


Air Blower Control The control system for the main air blower depends on whether the blower is an axial or centrifugal machine and whether it is turbine or motor driven. The most common configuration built today is a turbine driven axial blower because these are generally more efficient, but the choice is unit specific. Figure 7 shows a conventional control scheme for a turbine driven axial blower. Flow is measured by the venturi in the discharge line and is controlled by varying the speed of the turbine. Alternatively, the air rate may be controlled by varying the stator vane position for a fixed speed, axial blower. A variable vent line, called a snort, is located on the air blower discharge and is used to prevent air blower surge. An anti-surge system control system constantly monitors the flow through the blower and the discharge pressure and compares the operating condition to the surge line on the blower curve. If conditions close to the surge line are detected the anti surge-controller opens the snort valve to increase the flow to atmosphere and reduce the discharge pressure of the blower. Modern anti-surge control systems available from specialized vendors continuously monitor a number of process variables and calculate deviation from surge to allow operation closer to the surge line while providing better protection for the equipment. The process variables monitored by the modern anti-surge controllers are shown in Figure 7. Occasionally, on older partial combustion units, an additional, smaller vent is used to fine tune the air flow to the regenerator. This may be tied into a differential temperature controller (DTC) which controls the temperature difference between the regenerator dense and dilute phases to limit or control afterburn in partial combustion units. On RFCC units control valves are used on the air lines to the first and second stages of the regenerator (Figure 8). Typically the air to the second stage is set at ~25-30% of the total air flow. In some units an additional control loop is included to minimize the discharge pressure and thereby minimize the energy consumed by controlling the axial air blower stator vane position. This control loop minimizes the blower discharge pressure until one of the 2 flow control valves on the discharge is nearly wide open.


Figure 7 Main Air Blower Control Vent to Atmosphere

Silencer Snort Valve

Low Flow Shutdown

FO

FIC (Temp, Pressure Corrected)

Instrument Air

XIC (Anti-Surge)

S

Steam

Vent

ST SIC Suction Filter Housing

T TT PT

Air Cylinder FT

TT PT

TT PT

FT

Air to Regenerator

Special Check Valve

FCC-PC007

Figure 8 RFCC Main Air Blower Control (Anti Surge and Special Check Valve Details Not Shown) Vent to Atmosphere

Silencer

FIC (Press, Temp Corrected)

TT PT

ZIC

ZT To First Stage Regenerator

XIC (Anti-Surge) Steam

Suction Filter Housing

FO

Special Check Valves

FIC (Press, Temp Corrected)

T TT PT

ZT To Second Stage Regenerator

Stator Vanes

FCC-PC008


The check valve in the air blower discharge line isolates the blower from the regenerator if the blower trips. This minimizes the possibility of hot catalyst backing up into the blower and minimizes the volume of air in the piping if surging occurs. The closing action is assisted by a spring loaded air cylinder, which operates when the air flow falls below a certain limit. When the flow drops below the low limit a solenoid valve deenergizers and vents the air from the cylinder allowing the spring to move a cam which bumps the check valve to assist it in closing. An air line from the catalyst hoppers or plant air is used to provide plant air to clear catalyst from the discharge line if the blower is down. It can also be used to supply warm blower air to the catalyst hoppers.

Power Recovery Controls A typical process flow and control scheme for the flue gas system on an FCC with a power recovery unit is shown in Figure 9. On units with a power recovery turbine, butterfly valves are used in the flue gas line instead of slide valves for pressure control. The butterfly valves operate on a single, split range controller which first opens the valve on the line to the expander inlet then opens the butterfly valve on the bypass around the expander if the capacity of the expander is exceeded. The bypass valve may also be controlled to limit the speed of the expander or the pressure in the expander. In the past, flue gas power recovery systems were designed with regenerator pressure held steady to minimize fluctuations to the power recovery expander-motor/generatorair blower train. With this control strategy, a regenerator pressure controller output signal was used to control the power recovery butterfly valve positions. In this control scheme, the regenerator pressure is fixed and the reactor-regenerator differential pressure is allowed to float within reasonable limits. Recently, the control system for flue gas power recovery pressure control has been modified so that the reactor-regenerator differential pressure is controlled with a differential pressure controller as in conventional flue gas systems. The objective of this change in pressure control strategy is to insure that the reactor-regenerator pressure balance and catalyst circulation are maintained at steady conditions. Now the output signals of the reactor-regenerator PDIC and the regenerator PIC are fed to a high signal


selector (HSS). The high signal selector directs the appropriate control signal (normally the PDIC) to set the positions of the expander inlet and bypass butterfly valves via a split range signal. The expander turbine is discussed further in the section covering Equipment. Modern power recovery controls, while controlling in the same manner, are more complicated than discussed here. Power recovery controls and anti-surge controllers are provided by specialized vendors. The additional functions performed by these controllers are beyond the scope of this manual. On units with power recovery, a steam turbine may be used for startup of the air blower. The turbine can also provide auxiliary power if necessary. The motor/generator imports or exports power to maintain a constant speed on the power recovery train. If more energy is being supplied to the expander and the turbine than is required by the blower there will be surplus electricity generated and exported. If the blower needs more power than the expander is providing then electricity will be consumed to hold the train at normal speed. The flue gas leaving the regenerator flows through a series of small cyclonic devices in the third stage separator for catalyst fines removal to minimize erosion in the expander. The underflow catalyst stream from the third stage separator, carried by a small gas flow, bypasses the expander and typically leaves the system with the main flue gas stream downstream of the expander turbine. The underflow from the bottom of the separator is controlled by a restriction orifice called the critical flow nozzle. A flue gas quench system is also included on units with power recovery. If the flue gas temperature exceeds the design temperature of the expander a flow of cooling steam is started to the regenerator plenum through a split range controller. If the steam fails to cool the flue gas sufficiently a flow of cooling water, typically BFW, is started to the plenum. Most power recovery vendors also include an emergency steam quench at the inlet of the expander for additional protection.


Figure 9 Power Recovery Controls Quench Connection See Detail Below

TIC > PIC

HSS

Split Range

Flue Gas

Orifice Chamber

PDIC

Butterfly Valves

Third Stage Separator

Flue Gas Cooler

Critical Flow Orifice

Motor/ Air Blower Generator M

Electrostatic Precipitator

Steam T

Steam Turbine

Expander Air

Regenerator Plenum Chamber Quench Detail HighTemperature Signal Opens This Control Valve First

TIC

Split Range

LPS

Atomizing Steam FC

Purge Steam

FI BFW

Signal From Regenerator Temperature Transmitter

To Other Nozzles FC

RO

Concentric Sleeve Purge

Stuffing Box

Plenum Shell

Retracta ble Tip

FCC-PC009


Catalyst Cooler Controls Catalyst coolers provide FCC operating flexibility, permitting direct control over the regenerated catalyst temperature. The regenerated catalyst temperature is a major variable impacting cracking reactions since it sets both the catalyst to oil ratio and determines the temperature of the catalyst surface at its first contact with the oil feed. Both of these variables are important in determining the overall conversion and yield pattern in the FCC. Figure 10 shows a schematic of a flow-through catalyst cooler installed on a high efficiency regenerator with the associated cooler fluidization air, water circulation, steam drum and control instrumentation. As already discussed, hot catalyst from the upper regenerator flows through the cooler shell, around the water tubes of the inserted tube bundle and out of the cooler through a cooled catalyst standpipe and slide valve into the combustor. The fluidization air lance system delivers fluidization air to the cooler shell to maintain catalyst fluidization and mixing in the shell and to ensure that catalyst flows smoothly through the cooler and out through the standpipe. The combination of mixing and net catalyst flux through the cooler provide the driving force for heat transfer by maintaining contact of hot catalyst with the tube wall. The cooler duty is therefore controlled by controlling the amount of fluidizing air and the flow through the cooled catalyst standpipe. Minimum and maximum fluidizing air rates are typically specified to ensure that the air lances do not plug with catalyst and that high velocities in the cooler do not cause erosion on the tube surface. In back mix type coolers without a standpipe and slide valve the cooler duty is controlled only with the fluidizing air. To protect the tubes from thermal damage and oxidation, a large excess of water is circulated through the tubes to ensure that the tube walls remain wetted. The ratio of water circulated to steam generated is typically in the range of 20:1 to 25:1 lbs water circulation per pound of steam generated. In the most recent designs the water circulation rate is determined by the pump curve and no control valve is provided. In earlier designs water circulation is controlled by a control valve. A low flow signal from the flow indicator or controller activates a spare water circulation pump autostart. A low low flow signal from the water circulation flow indicator or a low low steam drum level


effectively causes the cooler to shutdown by closing the cooled catalyst slide valve and closing the fluidization air control valve. The mixture of steam and water exiting the cooler tube bundle flows to the steam drum where the steam and water are separated. The saturated steam flow from the drum is metered and flows out to the refinery steam system. Normally the saturated steam leaving the steam drum is superheated in some type of downstream steam superheater. The water returned to the steam drum is recirculated back to the cooler tube bundle. A small continuous blowdown flow of water is removed from the drum to control accumulation of impurities in the circulating water. The outputs of the steam product flow meter and the steam drum level indicator are summed and control the flow of boiler feed water (BFW) makeup entering the steam drum.


Figure 10 Catalyst Cooler Controls Steam and Water

Low Low Flow or Low Low Level Fluffing Air Trip

Steam

Low Low Level (2/3 Voting)

LG

LT

I

FI

LI PI

VENT

FIC

S

I

LT

TI

DE

PG

STEAM SEPARATOR

Air

LIC

LT

TI

I

FI

Low Low Flow or Low Low Level Slide Valve Trip

Intermittent Blowdown

Continuous Blowdown

Low/Low Low Flow (2/3 Voting)

FIC

I

FI

I

Low Flow Pump Auto Start FT

Makeup BFW VENT S DE

To Motor Control Circuit

To Motor Control Circuit

Steam

FO

T

M

M

FCC-PC010 I

Interlock System


Emergency Interlock Systems For many years automated interlock systems that removed feed from the FCC were not used because it was assumed that a well trained operator would make a better decision regarding stopping feed to the unit than any logic system looking at only a limited number of inputs. Also, since many of the inputs into the interlock system relied on pressure taps around the reactor and regenerator which were prone to plugging, the threat of spurious trips was too great. Recently, however, many refiners understand the value of a properly designed interlock decision to automatically remove feed from the FCC and place it in a safe condition. Also, in many refineries the turnover of operations personnel has increased so that many FCC operators have limited experience. There have been a number of incidents where operators tried to keep the unit running during upset conditions warranting shutting down the unit only to greatly increase the mechanical damage done to the unit. Instead of a shutdown lasting only a few hours an extended shutdown resulted. UOP now includes an emergency interlock system on all new units and major revamps. The purpose of this system is to move the unit to a safe condition during an abnormal event while permitting a safe, fast restart of the unit once the problem is resolved. This system automatically performs the steps necessary to accomplish these goals that were once left to the operator. The system is also designed to minimize the risk of spurious trips or shutdowns resulting from false indications when no abnormal condition existed. The emergency interlock system monitors the air flow rate, regenerated and spent catalyst slide valve pressure drops and valve positions, feed flow rate, reactor temperature, and reactor stripper level to determine and verify the existence of an abnormal event warranting a shutdown of the FCC. Once the abnormal event is detected and verified by 2 out of 3 voting systems or a by a combination of abnormal process readings the following actions are automatically initiated:


    

The feed is bypassed to the main column Raw oil flow rate is reduced (now going to the main column) The spent and regenerated catalyst slide valves are closed Steam is increased to the riser All related controllers are placed in manual as required

Torch Oil Nozzle Control Figure 11 below shows a typical torch oil arrangement for an FCC regenerator. The torch oil nozzles provide a means of injecting heavy oil, usually raw oil feed or HCO, into the regenerator when extra heat is needed, e.g. during startup. Details of the nozzle will be described in the section covering equipment. Control of the torch oil flow and the torch oil atomization steam is often by hand control on older units. New unit designs usually use a flow controller with a split range output to the torch oil flow and the torch oil atomization steam to ensure that the atomizing steam is commissioned before the oil. The torch oil assembly is provided with a continuous steam purge to the annular space around the torch oil nozzle to keep it clear of catalyst. This purge steam flow is regulated to less than 50 lb/hr (25 kg/hr) steam with a restriction orifice. In addition, a small flow of steam is sent to the nozzle tip through a restriction orifice when torch oil is not in use to help cool the nozzle and keep it from plugging with catalyst.


Figure 11 Torch Oil Control Atomizing Steam

PI FC

Steam

Annular Sleeve Purge Steam

STRAINERS

Split Range (Steam Valve Opens First)

Manifold Purge Steam

FI MUST BE READABLE FROM CONTROL VALVE

PI

RO

FI

FC

FIC

Regenerator Shell and Lining

PI MUST BE READABLE FROM CONTROL VALVE

I

Stuffing Box

S

Instrument Air

FC VENT

DE

Emergency Interlock Unit Shutdown Logic

From Raw Oil Pumps

From HCO Oil Pumps FCC-PC011


Direct Fired Air Heater (DFAH) Figure 12 shows the control system of a modern direct fired air heater, present on all FCC units between the main air blower and the combustion air inlet to the regenerator. The air heater is used for refractory curing and dry-out following repair or renewal of regenerator linings as well as during normal startup to heat the regenerator catalyst inventory. The DFAH outlet temperature is controlled by the fuel gas rate. A high temperature shutdown trips the fuel gas to the heater to prevent damage to the air grid in the regenerator. Modern air heater controls trip the fuel gas on a flame out signal from a flame sensor or on low air flow rate. Separate shutoff valves are provided so that the tight shut off of only the fuel gas control valve is not relied upon for emergency trip conditions.

Figure 12 Direct Fired Air Heater Control To Regenerator

Air Ignitor Start Flame Sensor

High Temperature Shutdown I

PDI Must Be Readable From Damper

I

TIC

Damper

Ignitor

Sight Port

Burner

Pilot/ Ignitor

Sight Port

Pilot Gas PI

I

Sight Port (sighting flame)

PI

Sight Port (sighting opp. wall

PI Pressure Control Valve PI

PI

PIC

S S

Interlock Shuts Down the Air Heater on:

S

VENT

DE

FC

INSTRUMENT AIR

Fuel Gas

  

Low Air Flow High Outlet Temperature Loss of Flame Detection

FCC-PC012


Catalyst Addition Controls Continuous additions of fresh catalyst are required to maintain the activity of the catalyst in the reactor and regenerator and to replace catalyst fines lost from the unit through the cyclones. Regular additions of small batches of catalyst results in more stable operating conditions and yields than larger batches added less frequently. The typical UOP catalyst addition system uses a volume pot and automated valve sequence to add a constant volume of catalyst at timed intervals. This system is shown in Figure 13. Specialized valves, designed to close on the catalyst transfer lines when they are full of catalyst, are required to ensure reliable service. Other systems are available including weight addition systems which include a load cell to add weighed batches of catalyst on a regular interval.


Figure 13 Catalyst Addition Control Fresh Catalsyt Hopper INSTRUMENT AIR

I

I

VENT

S

S

FC

Instrument Air

Plant Air Vent

Plant Air FC

RO

Volume Pot

RO

I

S

Instrument Air

Plant Air Vent

FC

FI FI Plant Air

Sight Flow Indicator

To Regenerator

FCC-PC013


Main Column Control The main column is the first step in the product separation sequence. The superheated reactor vapors need to be cooled so that fractionation can be conducted. In large measure, operation of the main column becomes an exercise in controlled heat removal coupled with sufficient liquid-vapor contacting to effect the desired degree of fractionation into desired product streams, typically main column bottoms (MCB), light cycle oil (LCO), heavy naphtha (HCN), unstabilized gasoline and wet gas. MCB, LCO and HCN are drawn as products directly from the main column, although on many FCC units HCN is not removed from the unit as a product stream. On some units, Heavy Cycle Oil (HCO) is drawn as a product from the main column between the MCB and LCO products. Main column sidecut products are often steam stripped in sidecut strippers for flash point adjustment. The unstabilized gasoline and wet gas are further separated in the gas concentration section. The arrangement and integration of heat exchange from an FCC main column varies from refinery to refinery based on the specific requirements and economics of a given installation. The following discussion describes typical heat exchange circuits used on an FCC unit. Figure 14 shows a simplified FCC main column schematic.


Figure 14 Main Column Overview Wash Water

WGC Spillback OVHD Vapor to WGC

CW

Net OVHD Liquid to Gas Con

HCN from Gas Con

Sour Water

LCO from Gas Con Steam HCN Product

HCO from Gas Con Steam

LCO to Flushing Oil LCO Product

HCN to Gas Con LCO to Gas Con HCO to Gas Con Reactor Product Vapor Steam Raw Oil to Reactor

BFW

MCB Product

FCC-PC400


Main Column Bottoms Pumparound Circuit The main column bottoms circulation system (Figure 15) is designed to desuperheat the reactor vapors, condense the bottoms product and scrub entrained catalyst fines from the reactor product vapor. Main column bottoms (MCB), also commonly called slurry oil, is removed from the bottom of the main column and typically pumped to a circulating bottoms/raw oil exchanger and one or more steam generators. MCB may also be used to provide heat to reboilers. The reactor vapors are desuperheated by contact with a large stream of cooled slurry oil on the disk and donut trays. The bottoms flow over the disc and donut trays also washes catalyst fines out of the reactor vapors. The total bottoms circulation rate over the disc and donut trays is normally set at 150% to 200% of the feed rate or 6 gpm per ft2 of column area (15 m3/hr per m2 of column area), whichever is greater. A minimum spillback valve is provided to maintain this minimum flow during turndown operation. This ensures enough circulating MCB is returned to the main column to adequately wet the disc and donut trays, thereby cleansing the reactor vapors of catalyst fines and preventing coke formation on the disk and donut trays due to insufficient liquid flow over the trays. Typically the bottoms temperature is maintained at 670-700°F (354-370°C) to minimize coking and fouling in the slurry circuit. The bottoms temperature is controlled by the LCO product draw rate (or HCO product draw rate if HCO is withdrawn as a product). This flow is adjusted manually by the operators. If the bottoms temperature is too high the LCO product rate is reduced to drop more LCO to the bottom of the column which lowers the bubble point of the MCB product. Alternatively, if a higher LCO endpoint is desired than can be achieved while controlling the bottoms temperature in this manner, a stream of cooled bottoms from the steam generator (quench) may be returned directly to the bottom of the column to sub cool the liquid there. In this manner the temperature in the bottom of the column is no longer composition dependent and the LCO/MCB cutpoint may be varied independently of the bottoms temperature.


Figure 15 Main Column Bottoms and HCO Flow and Control

FIC

FIC

LIC

HCN Stripper Reboiler

Debutanizer Reboiler

Torch Oil Pump Flushing Oil

MCB Quench

Disc and Donut Minimum Flow

FIC

Reactor Product Vapor

Steam FI

FIC

LIC

Raw 

FIC

BFW FIC FIC

MCB Product CW

Raw

FCC-PC401


During normal operations the heat input from the circulating main column bottoms to reboilers and preheat exchangers will be set by product and process considerations. This is true of the heat removed in the other pump around loops as well. Heat input to the steam generators is generally the only variable available to the operator for adjustment of the column heat balance and reflux rates. Increasing the bottoms flow through the steam generators will cause more heat to be removed. This will reduce the amount of hot vapor traffic up the column and eventually will reduce the overhead reflux rate and the heat removed in the overhead system. Each exchanger in the main column bottoms pumparound circuit is designed for oil containing catalyst particles. Main column bottoms flows through the exchanger on the tube side and the velocity should be kept between 3.75 ft/s and 7.0 ft/s (1.14 m/s and 2.13 m/s) for straight tubes and between 3.75 ft/s and 5.75 ft/s (1.14 m/s and 1.75 m/s) for U-tubes. Below the minimum velocity, catalyst can accumulate on the tube walls and slowly plug the tube while greatly reducing heat transfer. If the velocity is above 7 ft/s (2.13 m/s), there is a risk of erosion on the tube walls. If the circulating MCB is very low in ash content (<0.15 wt-%) the risk of erosion is greatly reduced. The rates required to meet these velocities must be calculated for each exchanger before startup. Bottoms product is withdrawn on straight flow control adjusted manually by the operator to maintain a steady bottoms level in the main column. The product is cooled, typically in cooling water exchangers before being sent to tank. In many cases the bottoms product is also exchanged with the raw oil. Removal of catalyst fines from the bottoms product by filtration, settlers or hydroclones is an option which may be included depending on the end use of the bottoms product. If a slurry settler is used to reduce catalyst fines in the bottoms product, the settler inlet flow is drawn upstream of any bottoms product coolers. Hot main column bottoms enters the settler tangentially and the swirling motion imparted distributes the heavy oil evenly as it moves up to the outlet. Settler superficial velocity is limited to 30 BPD/ft2 (50 m3/d/m2) or less in the settler so that the fines can settle to the bottom of the vessel. Clarified oil product (CLO) leaves from the top of the settler. Diluent (cooled main column bottoms, raw oil or HCO) is added at the bottom of the settler to carry the


fines back to the riser. The diluent addition rate is always equal to or slightly less than the slurry flow to the riser and the settler feed rate is normally equal to or greater than the clarified product from the from the top of the settler so that net flow is downward carrying the settled fines out of the settler. Figure 16 shows the slurry settler flow and control. Slurry settlers are not as common today because of the improvement in reactor riser terminations and cyclones. Filtration of the slurry product is becoming more common either to produce carbon black feed stock (which typically requires less than 100 wppm solids), to minimize downstream problems in fired heaters or to minimize tank cleaning cost. Figure 17 shows a typical flow and control for a bottoms filtration unit.

Figure 16 Slurry Settler Flow and Control FIC Clarified Slurry Oil Product

PSV

FIC

Return to Main Column

Main Column Bottoms

Slurry Settler FI

FIC

Diluent (HCO or Raw Oil) FIC

Sewer Slurry to Riser

FCC-PC402


Figure 17 Main Column Bottoms Filter Flow and Control PIC

Backwash Gas (N2 or Fuel Gas)

FIC

Clean MCB Product to Storage Bypass (normally closed)

PDI LS

LS

Filter #1

PDI

PDI LS

Filter #2

Filter #3

Split Range

N2 or Fuel Gas PIC

Solvent Fluid (LCO/HCO)

Vent

Receiver Vessel

MCB Product

LIC

FIC

FI

FIC

Catalyst Backwash to Reactor

Valve sequencing controlled by programmable logic controller(PLC) Signals to/from PLC not shown for clarity FCC-PC412


HCO Pumparound Circuit HCO is withdrawn from the column and commonly used to provide heat to the raw oil preheat exchanger and the debutanizer reboiler as shown in Figure 15. Occasionally, HCO is also used to generate steam and reboil side cut strippers such as a heavy naphtha stripper. HCO flow to each of these exchangers is regulated by a flow controller. Each controller is set by the operator to achieve desired product or process specifications. If a steam generator is used, then the HCO pumparound heat removal may be varied to some extent independently of process or product requirements. The HCO reflux to the column is typically pumped back to the column on flow control reset by the HCO draw tray level. A decrease in heat removal from the HCO circuit will require an increase in heat removal elsewhere in the column. For example, if less heat is removed in the debutanizer reboiler then fewer vapors will be condensed in this part of the column increasing the amount of vapor rising up the column. If the excess heat is not removed in another pumparound circuit the overhead duty and reflux to the top of the main column will increase to remove this heat. HCO product (if any) can be removed through a stripper for flash point adjustment. The amount of HCO product produced will depend on reactor operating conditions, feed quality and catalyst type. The plant operator will make adjustments in HCO, LCO and naphtha draw off rates to maintain target cut points or end points for these products. If HCO is taken as a product it would typically be cooled against raw oil in a charge preheat exchanger and then sent through a water cooled product cooler on flow control. The HCO circuit also provides torch oil to the regeneration section for start-up and emergencies and flush oil to the main column bottoms pumps.


LCO Pumparound and Product Circuit Circulating LCO normally provides heat to the gas concentration unit stripper reboiler and the debutanizer feed exchanger as shown in Figure 18. Flow to each exchanger is regulated by a flow controller which is set according to process requirements. A stream of LCO, after heat exchange at the stripper reboiler, is sent to the sponge absorber as “lean” oil to adsorb light gasoline range components from the light gas in the gas concentration section. The “rich” oil from the sponge absorber is returned to the main column with the cooled LCO circulation stream, providing internal reflux for the main column LCO section. On many units this reflux steam is circulated from a total trap tray by gravity through a flow meter to provide additional column process information. Most units contain an LCO stripper for LCO product flash point adjustment. Depending upon the plant arrangement, light ends in the LCO can be removed either by steam stripping or by refluxing the liquid using heat from the circulating HCO. The LCO product may exchange heat with the raw oil feed stream before being cooled with water in a product cooler and being sent to storage. Flow to storage is set by a flow controller. The amount of LCO sent to storage is typically varied to control the temperature in the bottoms of the main column as described previously. Alternatively the LCO product rate may be adjusted to maintain a constant temperature at the draw tray and therefore maintain a constant endpoint. Stripped LCO is available for use as emergency quench to the riser, instrument flush and pump gland seal oil.


Figure 18 LCO Flow and Control

FI FIC LIC Steam

FIC LCO Product

BFW CW Preheater

FIC

LCO to Flushing Oil

Debutanizer Feed Exchanger

FIC Stripper Reboiler Rich Oil fromSponge Absorber Lean Oil to Sponge Absorber

FCC-PC403


Heavy Naphtha Pumparound and Product Circuit A heavy naphtha (HCN) product is typically withdrawn through a side cut stripper column in order to remove light ends for vapor pressure adjustment. The product may be either steam stripped or reboiled with HCO. The naphtha stripper bottoms product is cooled with water in a product cooler and then sent to storage or treating on flow control. The HCN product flow rate is adjusted by the operator to maintain a constant draw temperature and therefore endpoint at the draw tray. The portion of the naphtha draw from the main column which is not stripped as product is pumped and used for heat exchange with raw oil feed, the C3/C4 splitter reboiler, water or other low temperature streams and returned to the main column as internal reflux for the main column naphtha section. Figure 19 shows the flow and control for the naphtha pumparound circuit.

Figure 19 HCN Flow and Control

FI FIC LIC HCO FIC HCN Product CW

FIC

C3/C4 Splitter Reboiler

FCC-PC404


Main Column Overhead System Reactor product vapors contain large quantities of light gas and gasoline vapors which pass through the entire main column as gases. A portion of these products are condensed in the overhead condenser, the trim condenser and separated in the main column overhead receiver. A quantity of the condensed hydrocarbon liquid (unstabilized gasoline) is pumped to the main column as reflux (Figure 20). Reflux to the main column controls the overhead vapor temperature. This temperature determines the endpoint of the debutanized gasoline product from the gas concentration unit. The reflux also heat balances the column. If heat removal from one section is changed the overhead reflux rate will respond in the opposite direction to maintain a constant top temperature. For example, if more heat is required in the feed to the debutanizer, LCO circulation will be increased to provide the heat. The increased heat removal from the LCO circulation will condense more vapors rising up the column. Less heat will reach the top of the column. The top temperature controller will then reduce the reflux flow to maintain the column top temperature, thus heat balancing the column. The unstabilized gasoline liquid in the receiver not used as reflux is pumped to the primary absorber in the gas concentration unit on receiver level control. Gas flows to the suction drum of the wet gas compressor in the gas concentration unit. Water from the overhead receiver water boot is pumped to the waste water treating unit. The main column pressure, and therefore the reactor pressure, is controlled at the overhead receiver by the amount of gas removed through the wet gas compressor. This control must be very steady because swings in this pressure will affect the pressure balance between the reactor and regenerator and therefore will affect the catalyst circulation. The wet gas compressor control system depends on the type of driver and the compressor design. Regardless of the primary pressure control system, a backup control system is present in case the wet gas compressor trips or can not handle all of the gas produced in the reactor. This is shown in Figure 20. In this case an overpressure control valve, typically


set at 0.3-0.5 psi (0.02-0.035 kg/cm2) above the setpoint for the primary pressure controller will open venting some of the gas to flare or a low pressure fuel gas system. This control system allows the unit to continue operating during a short duration compressor failure and is useful during startup before sufficient gas is produced in the reactor to run the compressor.

Figure 20 Main Column Overhead Flow and Control Wet Gas Compressor First Stage Spillback Signal to Wet Gas Compressor Controls

Cooling Water FC

PSV Wash Water

PRC

PIC

To Flare Header To Wet Gas Compressor Suction Drum

LIC

TIC

LIC

FIC

FIC

Net Overhead Liquid to High Pressure Receiver FI

Sour Water FCC-PC405


Wet Gas Compressor Control Steam Turbine Driven Wet Gas Compressor Control The compression of the wet gas from the main column overhead receiver to the gas concentration section is done in two stages, with external cooling and liquid knockout between the stages. A variable speed steam driven centrifugal compressor is shown in Figure 21. The control system is set up to hold pressures steady and to prevent the gas compressor from surging, which can cause serious damage to the wet gas compressor. The pressure control signal from the main column overhead receiver has a split range, to the low signal selector controlling the spillback, and to the governor on the steam turbine. The second stage spillback to the first stage compressor discharge is controlled by a low signal selector which is fed by the interstage suction drum pressure signal and the second stage flow signal. The low signal selector will ignore one of the two inputs, whichever is higher, and transmit the lower output signal intact. The system works as follows: 1.

Normal flow - at lower charge rates, the main column overhead receiver signal will control the first stage spillback to hold receiver pressure constant. The second stage spillback will be controlled by the anti-surge controller to satisfy the minimum flow requirements of the surge curve. The turbine will be running at minimum governor speed. As the charge rate is increased, there is a greater net flow of gas, so the two spillbacks are gradually closed by the controllers. When the spillbacks reach the fully closed position, the output signal from the overhead receiver pressure controller will be high enough to begin increasing the compressor speed. The speed will then be continuously varied to remove the gas from the main column and hold the pressure constant. Throughout this, the flow controllers will have no effect as long as there is sufficient gas flow to keep the machine out of surge.


2.

Abnormally low flow - there would be two main reasons for low gas flow. The first would be insufficient gas from the main column, such as when the feed was suddenly cut out. The second reason would be an abnormally high second stage discharge pressure, caused by problems in the gas concentration unit. Whatever the cause, when the flow falls below the point set by the operator on the FRC, the output signal falls, which would then be transmitted by the low signal selector. This opens the spillback valves to recycle gas back to the suction, allowing the gas to be moved through the machine in sufficient quantities to keep it out of a surge condition. When normal conditions are restored, the flow controller output will rise, and the low signal selector will return the spillbacks to pressure control. If the low flow is caused by high pressure in the gas concentration unit, opening the first stage spillback will overpressure the main column. This condition will open the overpressure control valve to flare, holding the receiver pressure at the overpressure set point.

3.

Abnormally high flow - the main cause for very high flow rates would be operating the FCC unit in such a fashion that the compressor could not handle the gas flow. The machine would be at maximum governor speed with both spillbacks closed. Pressure in the main column would rise, opening the overpressure control valve to flare.

Modern anti-surge and pressure control systems take advantage of faster transmitters and computer technology to allow monitoring of multiple variables and rapid calculation of corrected flow rates to allow more efficient operation closer to the surge line with less spillback. These inputs to the anti-surge controllers are shown in Figure 21.


Figure 21 Wet Gas Compressor Control Signal from Main Column Overhead PRC

From Main Column Overhead Receiver

FI

FI

Antisurge Controller

Antisurge Controller

To High Pressure Receiver

Antisurge Signal Opens Control Valve TT

Nirogen Purge

Nirogen Purge

FO

FO Steam

TT

TT

TI SI

SC

PT

PT

PT

Antisurge Signal Opens Control Valve

TT

PT

T

Speed Sensor

First Stage Spillback to Main Coulumn Overhead

To Surface Condenser

First Stage

Second Stage

FCC-PC407

Steam Turbine, Variable Speed


Motor Driven Wet Gas Compressor Control This case would be for a constant speed centrifugal compressor. The instruments, controls, and actions are essentially the same as discussed above, except that the governor speed control is replaced by a suction butterfly valve to throttle the gas flow to the compressor. This valve normally stays partially closed, against a limiting stop. As more flow is required, after the spillbacks have closed, the butterfly valve will open allowing more gas to move through the compressor. The action on low gas flow is to move gas through the compressor spillback valves in the same manner as described above for the variable speed machine. Reciprocating Wet Gas Compressor Control Older FCC units occasionally used reciprocating wet gas compressors. These are rarely used today because of the better efficiency of the centrifugal compressors. Reciprocating compressors are normally constant speed machines. At normal flows and pressures, the spillbacks open or close to hold the overhead receiver and interstage suction drum pressures at the point set by the operator. Abnormally low flow rates, such as when the charge is cut, will cause a drop in pressure, opening the spillbacks. High second stage discharge pressure will have the same effect. Reciprocating machines may also have clearance pockets and suction valve unloaders which can be varied to control flow, but this will depend on the individual machine. Variations There are other control schemes which are in use today, depending on the individual refiner's needs. Some plants have been designed with only one spillback, from second stage discharge to the main column overhead receiver. This does not allow the same freedom of action as that offered by two spillback valves. Excess pressure drop through one part of the system caused by such things as exchanger fouling, is one reason why it is better to have the extra freedom of two spillbacks.


Main Column and Gas Concentration Section Water Wash The overhead stream from the main column contains a number of contaminants which can cause corrosion, plugging and fouling. These contaminants include ammonia, sulfides, cyanides, chlorides and phenols. A wash water stream is used to remove the contaminants. The wash water should be clean, preferably steam condensate, to prevent adding more problems such as salts or dissolved oxygen to the system. Wash water is effective in removing most of these impurities because most of them are ionic or polar species which tend to be readily soluble in water. Figure 22 shows the preferred arrangement of the water wash system for an FCC fractionation and gas concentration section. Clean water is pumped into the first stage discharge of the wet gas compressor at the inlet to the interstage cooler. The water from the interstage receiver is pumped out on level control to the wet gas compressor discharge at the inlet of the high pressure cooler. Water collected in the high pressure receiver water boot is pressured on water boot level control to the inlets of the main column overhead condenser and trim condenser. Sour water is then pumped to disposal from the main column receiver water boot on receiver water boot level control. The recommended water wash rate is 6.5 to 7.0 vol% of feed or about 2 gpm per 1000 BPD feed (approximately 1.15 liters/min of water per 1 m3/hr of fresh feed rate). The drain water from both the overhead receiver and the high pressure receiver should be checked regularly for pH. The main column overhead receiver water should be in the range of pH 8.0 to 8.5. The high pressure receiver drain water should also be slightly alkaline, pH 7.5 – 8.0.


Figure 22 Wash Water Control Main Column OVHD Vapors FI

CW

Main Column OVHD Rec. CW

Liquid from Primary Absorber Vapor from Stripper To Primary Absorber

Sour Water

High Pressure Receiver M First Stage

Second Stage

Wet Gas Compressor

LIC

To Stripper

To Primary Absorber

LIC

FIC

Condensate Water Break Drum

Water Flow Indicated by Bold Lines

FCC-PC409


Feed Preheat Train There are two basic raw oil preheat schemes used in FCC units. One uses a fired heater to supply some or all of the feed preheat duty. Fired feed preheaters are expensive to build and operate by today's standards so fired preheaters are mainly in use on older FCC units. The other feed preheat scheme uses extensive heat integration to provide feed preheat duty. Most new FCC units use this heat exchange scheme and do not employ fired feed preheaters. Figures 23 and 24 show FCC feed preheat schemes with and without fired heaters, respectively. In both preheat trains, the raw oil flows to a feed surge drum directly from the crude unit, vacuum unit, hydrotreating or from storage. The surge drum pressure rides on main column pressure through a vent line connected to the lower section of the main column just above the HCO draw. The surge drum level is normally controlled by controlling the flow of one of the feed sources into the unit. The charge is pumped from the surge drum on flow control through heat exchangers before reaching the feed distributor at the reactor riser. The only significant difference between the two schemes is the presence of the fired heater. Both schemes use a TRC to control the raw oil outlet temperature from the circulating main column bottom/raw oil exchanger by bypassing some of the raw oil around the exchanger. The flow of main column bottoms through the exchanger is varied to keep the temperature controller in a good operating range. With the fired heater, each heater pass is also controlled by a TRC.


Figure 23 Feed Preheat System Equalizing Line To/From Main Column FIC

LIC

Raw Oil Surge Drum

FIC

High Temperature Closes Control Valve

Split Range

TIC To Reactor

FIC LSS

< Split Range FIC

Raw Oil from Raw Oil from Storage Crude Unit/ Level Control Hydrotreating Signal from Upstream Unit

LCO MCB Product Product Low Temperature Closes Control Valve

High Level Closes this Valve First

Circ. MCB MCB Recycle HCO Recycle

FCC-PC411


Figure 24 Feed Preheat with Fired Heater Equalizing Line To/From Main Column

Raw Oil Surge Drum

LIC

FIC

TIC

TIC Fired Heater

Raw Oil from Storage/ Upstream Unit

MCB Product

To Reactor

Circ. MCB

Fuel Gas

FCC-PC410


Gas Concentration Unit Control A typical flow scheme for the overall gas concentration unit is shown in Figure 25. It is convenient to discuss the controls in 2 sections - the absorber section and the fractionation section. The absorber section removes LPG and light material from the gases and recycles them back to the high pressure receiver. The fractionation section strips C2 and lighter material as well as H2S from the LPG and gasoline and recycles them back to the high pressure receiver. The fractionation section also separates the LPG and gasoline. Because the absorbed and stripped material are recycled back to the high pressure receiver good control of the gas concentration unit requires a good balance of stripping and absorption. If excessive stripping is occurring then excessive absorption will also be required to achieve reasonable C3 and C4 recoveries from the fuel gas. This is at a minimum a waste of utilities and in severe cases can result in increasing recycle flows, or “snowballing”, until one or more of the columns flood.


Figure 25 Gas Concentration Unit Flow Lift Gas to Reactor

Lean Gas KO Drum

Unstabilized Gasoline from Main Column OVHD

Wet Gas Compressor Discharge

Sponge Gas to Treating

CW

CW

Primary Absorber

Sponge Absorber

Rich Oil to Main Column CW


Stripper Wash Water to Main Column OVHD Receiver

LCO from Main Column

Wash Water from Interstage Receiver

LPG to Treating

Debutanizer

Circ. HCO

Stabilized Gasoline to Treating CW

FCC-PC703


Absorber Section - Primary Absorber LPG and heavier hydrocarbons are recovered from the light gas (also called sponge gas or fuel gas) in this section of the gas concentration unit shown in Figure 26. Gas from the high pressure receiver enters the primary absorber below the bottom tray and is contacted by counter current flow with unstabilized gasoline on level control from the FCC main column overhead receiver. Many units also recycle stabilized gasoline (debutanizer bottoms) on flow control to the top tray to increase recovery of the C3 and C4 hydrocarbons. Unstabilized gasoline is typically fed to the 6th tray from the top when recycle gasoline is used. Heat is removed from the column to maximize adsorption efficiency with two pumparound inter-coolers, one about one third the way down the column and another about two thirds of the way down from the top. The rich oil flows from the bottom of the column to the high pressure condenser on level control cascaded to a flow controller. The overhead gas stream flows into the sponge absorber. Absorber Section - Sponge Absorber The sponge absorber is typically a packed column which serves to recover nearly all the remaining C5 and C6 material and some C3 and C4 hydrocarbons. Circulating light cycle oil on flow control from the stripper reboiler is heat exchanged with the rich oil and then cooled before being pumped to the top of the absorber as “lean” oil. The gas from the primary absorber flows upward from the bottom of the column. Rich oil leaves the bottom of the column on level control. The rich oil is heated in the rich oil/lean oil exchanger before returning to the main column with the circulating LCO stream. Lean sponge gas leaving the top of the sponge absorber is cooled and any condensed liquid drops out in the lean gas knockout drum. The liquid is periodically drained to the LCO return line. Gas from the lean gas knockout drum is normally sent on pressure control to a fuel gas amine treater before being sent to the refinery fuel gas system. In FCC or RCC units using lift gas technology, a portion of the lean sponge gas (before amine treating) is recycled to the reactor riser on flow control for use as lift gas. The sponge gas pressure


controller sets the pressure for the knockout drum, the sponge absorber, the primary absorber, the high pressure receiver and the stripper column. The pressure of each vessel depends upon the pressure drop between the vessel and the lean gas knockout drum.

Figure 26 Gas Concentration Unit – Absorber Section Lift Gas to Reactor

Lean Gas KO Drum

Unstabilized Gasoline from Main Column OVHD

Wet Gas Compressor Discharge

Sponge Gas to Treating

CW

CW

Primary Absorber

Sponge Absorber

Rich Oil to Main Column CW


Stripper Wash Water to Main Column OVHD Receiver

LCO from Main Column

Wash Water from Interstage Receiver

LPG to Treating

Debutanizer

Circ. HCO

Stabilized Gasoline to Treating CW

FCC-PC703


Fractionation Section - Stripper Column Removal of light hydrocarbons from the LPG and gasoline is accomplished in the stripper column (see Figure 27). Liquid from the high pressure receiver is heated with debutanized gasoline before entering the top of the stripper. Heat to the stripper is provided from the debutanized gasoline reboiler and the LCO reboiler. The LCO flow to the reboiler is controlled on a cascade loop to maintain a constant vapor rate from the top of the column. A bypass around the debutanizer bottoms to the lower reboiler is provided in case the demand for LCO to the upper reboiler is less than the required flow to the sponge absorber. The stripping rate is varied to reject all of the C2 and lighter material and in some cases to control the amount of H2S going to the debutanizer and therefore into the LPG product. Stripper overhead vapors return to the high pressure condenser. Liquid leaves the bottom of the stripper on level control and is pressured to the debutanizer column feed exchanger. Fractionation Section - Debutanizer Column Gasoline vapor pressure adjustment and separation of the C5 and heavier components from the LPG is achieved in the debutanizer column. Circulating LCO provides heat to the debutanizer feed before the feed enters the column at tray 20. LPG leaves the top of the column as a gas and is cooled in the condenser before collecting in the overhead receiver. The pressure in the column is controlled by controlling the amount of vapor entering the condenser. A hot vapor bypass around the condenser is used to control the pressure drop between the column and the receiver to ensure that the opening of the pressure control valve stays in a good operating range. If the hot vapor bypass valve is fully open then the temperature, and therefore the pressure, in the receiver is too low indicating that less cooling is required at the condenser and that either the fan speed should be reduced or one or more fans should be shut off. Likewise, if the hot vapor bypass valve is fully closed then the fan speed needs to be increased or additional fans turned on. Liquid from the overhead receiver is returned to the column top as reflux to control the column temperature at tray 6. Net overhead liquid is pumped to the LPG treating unit to control the level in the receiver.


Heat to the debutanizer reboiler is typically supplied by circulating HCO. The flow of HCO is adjusted by the operator to adjust the reflux rate in the column and the quality of fractionation. The temperature controller is adjusted to control the RVP (Reid Vapor Pressure) of the gasoline. Stabilized gasoline flows from the column bottom to supply heat to the stripper reboiler and then to the stripper feed exchanger. Gasoline from the stripper feed exchanger can be split into two streams. Recycle gasoline is cooled and pumped to the primary absorber on flow control. Net debutanizer bottoms (i.e., gasoline) is cooled and pumped to the treating unit. The net gasoline flow controller is reset by the debutanizer bottoms level controller to maintain the column bottom level.


Figure 27 Gas Concentration Unit – Fractionation Section PIC

Vapor to HPR Debutanizer

LIC

FIC Stripper

PDIC

1

1

7

TIC 18 19

LPG to Treating

19 20

Circulating LCO To/ From Main Column

36

35 36

TIC FIC

40

FIC LIC Circulating HCO To/From Main Column

Liquid from HPR Split Range

FI

FIC Stabilized Gasoline to Primary Absorber CW

FIC Stabilized Gasoline to Treating

FCC-PC700

157048-1 Equipment Page 1

EQUIPMENT INTRODUCTION The Fluid Catalytic Cracking unit is exposed to severe temperature, erosion and corrosion effects. The equipment has been designed to withstand these conditions for an acceptable mechanical life, but the life can be drastically shortened by abuse or poor operations. Proper control and operation will avoid the unnecessary problems which lead to premature failure. The FCC unit should be inspected every turnaround. The inspection can be done by a UOP inspector, the refinery inspection department, or both. Adequate record keeping of these inspections is necessary to develop the unit history which will aid the refiner to judge equipment life, determine potential problems and evaluate the effect of different metallurgy and process conditions related to the equipment. A good inspection will include the following: 1)

An evaluation of the equipment which has experienced erosion and/or corrosion

2)

A list recommending minor repair work

3)

An assessment of mechanical problems caused by the operating conditions of the previous run

4)

A list of spare parts required for the next turnaround

Erosion and corrosion are not always obvious. A thorough inspection of the equipment will reveal the extent of any damage and help determine the cause and effect relationship. Also, because no equipment lasts forever, the inspection will help the refiner determine when equipment must be replaced so advance orders can be made.


Minor repair work can prevent small problems from expanding to larger problems. It is difficult to predict the amount of minor work which will be required. The desired length of the next run will determine the extent of the repairs.

PREPARATION for INSPECTION Preparation for inspection should begin well before the shutdown. Proper scheduling of inspection and maintenance will avoid delays and wasted time. As soon as an inspector finds a problem, parts can be ordered and work can be scheduled so the completion of the turnaround will not be delayed. The normal shutdown procedure will remove most of the catalyst from the unit and cool the vessels to 200-250°F (90°-120°C). The manways should be opened on both the reactor and regenerator to air cool the vessels. Vacuum connections are provided to remove any remaining catalyst. The vacuuming operation can begin while the manways are being opened, if there is sufficient manpower. When the vacuuming is complete, water washing can be started. This will remove the dust and fines which would hinder a complete inspection. A simple water spray is usually sufficient, with the water draining out of the reactor and regenerator at the bottom of each vessel. The water will not cause any problems with the vessel internals, even stainless steel. A high-pressure blast could obviously damage the refractory; common sense is required. Clean, potable water with less than 50 ppm chlorides should be used. Excess water should not be allowed to stand on the equipment for long periods of time. An air hose can be used to blow away these puddles. There are a few areas that are difficult to drain, such as the regenerator plenum chamber. These can be cleaned with a heavy-duty vacuum cleaner. Because there are different reactor and regenerator designs, the exact cleaning method should be determined by the refiner. The main column and gas concentration section should also be cleaned for inspection after the normal shutdown procedures have been completed. Hydrocarbon and sour water should be pumped or pressured out of the unit. Gas and vapors are removed by steaming out the vessels. Any remaining material can


be removed as necessary by water washing the equipment. Do not allow any water in the wet gas compressor. Normal refinery safety practices should be followed for toxic vapors, explosivity, oxygen content of vessels, etc. The refinery safety engineer should follow the turnaround carefully.

AIR BLOWER The main blower of an FCC unit supplies large quantities of air to the regenerator. Advances in rotating machinery technology have led to the replacement of the old positive displacement reciprocating air compressor with centrifugal and axial machines. Blowers can be driven by steam or gas turbines, electric motors, or flue gas turbines, usually referred to as power recovery expanders. Depending on the mode of operation and other factors such as feed quality, the FCC unit needs 1014.5 pounds of air per pound of coke, which is approximately 2000-3000 SCF/bbl (330-500 Nm3air/m3FF). Air is filtered through a screened suction housing that should be designed for noise abatement. Compressed air leaves the blower at about 300-450°F (150-230°C) and 30-60 psia (2.1-4.2 kg/cm2(a)). CENTRIFUGAL MACHINE Air flow rates on a centrifugal machine are controlled by varying the speed of rotation, throttling the suction, or venting off excess air. Four to six stages are common, with labyrinth seals used to prevent leakage between stages. These machines normally use forced lubrication systems for the bearings and may be equipped with temperature and vibration probes for early detection of mechanical problems.


AXIAL MACHINE An axial blower, shown in Figure 1, uses rotating blades to move the air. The air flows through the machine in a straight line, each successive stage adding pressure energy much like a propeller blade. The air flow rate is most often controlled by varying the shaft speed. This can be accomplished either through a turbine driver, or by including a Variable Speed Drive (VSD) unit on a motor. If the machine is designed for constant speed, other means of flow control must be provided. One option is to snort (vent) excess air to atmosphere. However, for normal control this would require power to compress air that is vented back to atmosphere, and is simply not energy efficient. Practical options include the use of small variable pitch blades known as “stators” on the blower housing. The variable pitch stators redirect the air flow into the path of the rotating elements. When this redirection is at a steeper angle, more air is transferred. These machines use forced lubrication systems and are normally equipped with temperature and vibration probes. Another option to control air flow from a fixed speed blower is to include a suction throttle valve. This mode of operation is very common in older machines, but as with a snort valve, it is not an energy efficient system. The suction throttle valves on motor driven air blowers can be eliminated through the installation of a VSD unit onto the motor. Axial compressors are generally more efficient at larger capacities than centrifugal machines. They are smaller and lighter than an equivalent size centrifugal unit. Choice of machine depends on the individual refiner, but axial blowers are more common for larger units. Specific operation of these large machines is too complex to describe in this manual. Individual manufacturer's instructions should be followed for each unit.


Figure 1 Axial Compressor

BLOWER PROBLEMS The problems encountered with an FCC blower can be divided into two groups; operational and mechanical. Examples of mechanical problems are vibration, shaft displacement and noise. These may result from manufacturing defects, construction mistakes such as misalignment of the driver and blower, and routine wear of the bearings. In most cases, however, mechanical problems are caused by operational difficulties. An example is surge, which occurs when the air blower is not able to produce enough head to overcome downstream resistance. A centrifugal compressor curve, shown in Figure 2, gives a typical head-flow relationship, while Figure 3 shows a curve for an axial machine. At a fixed speed the compressor will follow the curve. Flow decreases with increasing head, similar to a pump. Unlike a pump, however, the characteristic curve begins to turn down toward the zero capacity region after reaching a peak in


pressure. This peak is called the stability limit or surge point. The machine will thus produce less head at the decreased capacity. The pressure downstream of the machine is higher and the flow reverses through the blower. When the downstream pressure is relieved, normal flow is restored. Resistance quickly builds again and the machine surges. The condition can be recognized by a characteristic cycling sound and a rapid rising and falling of the flow between normal and zero. When the gas moves back into the machine, the rotor tends to stall, which can cause serious damage as the machine is subjected to large unusual forces and increased gas temperature across each surge cycle. Surge cycles can result in damage to the thrust or axial bearings and labyrinth seals. In severe cases the rotor itself may crack or rub against the casing. On an axial blower the blades may break. To prevent this backflow condition, many blowers are fitted with an anti-surge controller. The controller essentially compares pressure rise and flow against a programmed operating map of the blower. The operating map includes a calibrated surge line. As the machine approaches the surge line a blow-off valve, commonly referred to as a “snort valve”, is opened slowly. Opening the snort increases flow through the machine and prevents surging. Surge is generally considered more dangerous to axial blowers than to centrifugal, but should be avoided for either. Choke, or stonewalling, is a low pressure, high flow condition where the gas velocity approaches the speed of sound. Dangerous vibrations result and can cause the rotor to crack. This condition may also be seen on the characteristic curve at low head. The line becomes almost vertical as the capacity increases and air velocity approaches the sonic value. This condition is infrequent but care must be taken to avoid it. Choke is more of a problem for axial than centrifugal machines. Rotating stall is a somewhat rare phenomenon, indicated by an inability to build pressure while the flow is more or less normal. It results when air moves around the axial rotor, rather than through it. The best cure is to back down to starting conditions and restart increasing the flow again. The air blower suction line should be inspected for cleanliness. The suction hood or housing should be cleaned. Except for vibration from the blower, this section of the plant is not subject to any unusual stresses, and normally lasts for many years.


Inspection and maintenance on the blower and associated systems should follow the manufacturer's recommendations. Replacement of the bearings, seals, lubricating oil, etc. is required at certain intervals. The air filters on the suction housing should be inspected during normal turnarounds and replaced or repaired as required. Performance (capacity) of the blower can diminish over time due to fouling of the blades. Foulants may include dust and chemical contaminants drawn into the machine through the filter housing. Episodes of blade fouling have been associated with sudden and heavy rains, washing contaminants off of the suction filters and into the blower. Performance decline of up to 10% of rated flow have been associated with blade fouling. Most vendors offer a surfactant or solvent injection system that can be added to the suction line of the machine to help remove blade foulants from the blower. The air blower discharge line is not subject to corrosion or metal loss. All air snort valves should be checked by the instrument department; few problems are ever encountered with the air snort valves. In the event of an emergency trip, many modern machines require that they be rotated while cooling down. If this procedure is not followed when required, serious rotor deflection can result. Excessive rotor deflection can result in serious mechanical damage to the compressor, requiring a major overhaul of the machine.


Figure 2 Centrifugal Air Blower Performance Curve

Figure 3 Axial Air Blower Performance Curve


POWER RECOVERY The power consumed by the air blower is a significant part of an FCC unit's operating expense. Utility costs continue to rise, only partially offset by more efficient motors and turbines. Recovery of the energy in the hot flue gas from the regenerator can increase the overall efficiency of the unit. This was first done with steam generation systems. The flue gas exchanges heat with a circulating water stream. A 30,000 BPD (195 m3/hr) FCCU without a CO boiler can produce 40-70 Mlb/hr (18-32 t/hr) of 600 psig (42 kg/cm2) steam. Heat recovery in this scheme is somewhat limited by a minimum allowable flue gas temperature. Sulfur oxides and water vapor in the stack gas can cause corrosion of the equipment if they condense in the flue gas duct. The temperature at which the condensation occurs is known as the acid gas condensation point, which shifts depending on the concentration and distribution between the different oxides. The acid gas condensation point is typically in the range of 400-600°F (200-315°C), although it may be higher for some units. The maximum temperature limit of the flue gas is typically a function of metallurgical design limits for downstream equipment, which may include an electro-static precipitator, flue gas scrubber, and/or stack. The major disadvantage of a straight steam generation energy recovery scheme is that no power is recovered from gas pressure, normally 10-40 psi (0.7-2.81 kg/cm2) above atmospheric pressure at the regenerator outlet. Another approach to recovering energy from the flue gas was tried in 1950. This was a turbo expander, driven directly by hot flue gas. Initial results were unsatisfactory; after only 750 hours of operation catalyst fines in the flue gas had substantially eroded away the turbine blades and casing. The fines problem was solved by placing an additional catalyst separator, known as a Third Stage Separator (TSS) outside of the regenerator. In the TSS, flue gas moves through a large number of small cyclone assemblies in which the catalyst is centrifugally separated from the flowing gas stream. To remove the separated catalyst fines from the TSS, a small amount of gas, typically 3% of the regenerator flue gas, is used to pneumatically sweep the catalyst fines


out the bottom of the vessel. The clean flue gas is then directed to the inlet of the power recovery expander. UOP has been designing power recovery systems since 1973. Between 1973 and 2004, UOP licensed 33 TSS’s with 22 placed into operation. The original units were designed by UOP under license from Shell. Over the years, UOP improved upon the original design by implementing several modifications. Even with these modifications incorporated into the base design, very little had actually changed in the overall design of the TSS in 25 years. These TSS designs still suffered from the limitations imposed by radial flow gas distribution and reverse flow in the cyclone elements. In 1996, UOP launched a development program to design and offer a smaller, more economic, high efficiency TSS that could not only be utilized in power recovery installations, but also be a viable alternative to electrostatic precipitators and wet gas scrubbers for environmental applications. The cold flow modeling (CFM) test program extended over 2 years, during which both dimensional variables and process flow variables were studied. Based on a thorough understanding of cyclone theory, and drawing on other sources of cyclone expertise, the UOP program investigated the contribution of many variables on catalyst separation efficiency. These variables included:      

Cyclone diameter and geometry Inlet velocity Length to diameter ratio Outlet velocity Catalyst loading Gas distribution

Over 200 individual tests were conducted on single and multiple cyclone models to determine the highest efficiency and highest capacity design cyclone. The tests were conducted with commercial FCC catalyst fines. Computational fluid dynamic (CFD) computer modeling was used to validate and benchmark the CFM work, and to quickly investigate potential improvements and guide the physical modeling program.


The development work culminated in the new UOP TSS design (see Figure 9). The most significant improvement in the design is that the UOP TSS utilizes axial flow for catalyst/gas separation. The flue gas flow is maintained essentially in one direction - in the top and out the bottom of the unit. Axial flow distribution minimizes the potential for solids re-entrainment resulting from gas flow direction change and resultant eddy current formation. The older style TSS utilizes radial flow distribution in which the flue gas is distributed from the centerline of the TSS, radially outward between the two tube-sheets. As such, the inner tubes see a higher gas and dust loading than the outer tubes. The mal-distribution of flue gas and fines inherent in this design results in varying efficiency across the older style TSS. The new UOP TSS is about 40% smaller than other TSS offerings for the same capacity; making it less expensive to fabricate, easier to install, and better suited where plot space is a premium. The first UOP TSS was commercialized in April 2002. Performance testing on the unit was performed twice in 2002, following the unit startup in April and again in December. The initial test showed that the UOP TSS discharged between 36-50 mg/Nm3 of particulates, depending on flue gas rate. The NSPS compliance testing resulted in a particulate matter emission of 0.6 lbs/1000 lbs of coke burn, only 67% of that allowed by NSPS standards. This performance showed that the UOP TSS could not only provide power recovery expander erosion, but could also be used as in the refiners particulate emission control strategy, by replacing more traditional, costly, and hazardous means (electrostatic precipitators and wet gas scrubbers) of controlling particulates exiting the flue gas stack. A comparison of the older style TSS and newer style TSS is shown in Figure 4. Both vessels are carbon steel vessel with 4" (100 mm) of refractory lining and stainless steel internals. The cold-wall construction is more effective on both erosion and cost basis than the early hot-wall stainless steel separators. A coarse screen, or grate, covers the flue gas outlet entrance to trap large chunks of refractory or other debris. The overall efficiency of the separator depends on the efficiency of the regenerator cyclones and the quantity of catalyst fines being generated in the reactorregenerator system. The separator should remove >70-90% of the particles for high


and low regenerator cyclone efficiency, respectively. Most of the fines which pass through the separator are smaller than ten (10) microns. These small particles do not cause much erosion to the expander blades but the smallest particles can deposit on the expander blades and casing, causing vibration problems. The pressure drop across the expander is on the order of 10-30 psi (0.7-2.1 kg/cm2), with a temperature drop of 200°-250°F (110°-140°C). After driving the turbine, the flue gas goes to a steam generator for further energy recovery. The majority of the catalyst is removed from the flue gas with the underflow from the third stage separator which is typically routed back into the flue gas downstream of the expander. If required, an electrostatic precipitator or flue gas scrubber may be placed downstream of the steam generator to remove any remaining catalyst fines before the flue gas is exhausted to atmosphere. Alternatively the underflow may be filtered to achieve ~99.99% removal of the catalyst fines, or routed to a 4th stage cyclone separator to achieve ~60-90% removal of the catalyst fines from the underflow stream, depending on local environmental restrictions. The power recovery train usually consists of five parts; the expander turbine, motor/generator, air blower, and a steam turbine, and is commonly referred to as a “5-Body Train”, see Figure 5. In this arrangement the expander turbine is coupled to the main air blower shaft to directly supplement the power requirement of the blower. The 5-body train requires a steam turbine or motor to get it started; in some cases only one of them is provided. The expander, shown in Figures 6 and 7, is a single stage machine because of the low pressures involved. The gas to the expander is accelerated over a parabolic nose cone. Pressure energy is converted to velocity energy, and the high velocity gas drives the turbine. Expander turbines designed in the past were generally limited to an inlet temperature of 1200-1250°F (650-675°C) to prevent heat damage. This generation of expanders however, still required quench injection systems in the regenerator plenum chamber to protect the expanders in the event of a regenerator temperature


excursion. Quench systems essentially dumped steam and water into the regenerator plenum to cool the flue gas. While this provided for thermal protection of the expander, the increased steam and water in the flue gas often resulted in “sticky” catalyst that agglomerated in cement-like deposits that increased blade fouling on the expander. Newer expander turbines normally have a design temperature in excess of 1375°F (750°C) and do not require a quench control system. For units with a power recovery system, butterfly valves in the flue gas line control the differential pressure between the reactor and regenerator. The PDIC sends a signal to the large butterfly valve which is located at the inlet to the expander. A smaller butterfly valve will allow flue gas to bypass the expander when the large butterfly valve is fully open because of an excessive flue gas rate or when the expander is off line. This prevents over pressuring the regenerator. In the traditional five piece power recovery train, the motor/generator is usually a constant speed induction type machine that provides extra power to the blower shaft when needed. If the expander produces more energy than is required by the blower, the machine will act as a generator and feed power into the electrical grid. This acts as a braking mechanism and provides some over-speed protection for the machine.


Figure 4 Third Stage Separator Old Style TSS

New Style TSS

11' 6" OD 48 Tubes

29'

70 Tubes

23'

19' 3" OD

Figure 5 5-Body Power Recovery Train Flue Gas Exhaust Flue Gas Inlet

Air to Air In Regenerator Inlet Guide Vanes

Steam Turbine

Main Air Blower Expander Steam Inlet

Gear Box

Motor / Generator

Exhaust Steam Outlet

Electrical Connection to Power Grid


Figure 6 Flue Gas Expander

Because power recovery trains are generally fitted on larger units, the blowers used are of the higher efficiency axial type. The blower is a constant speed machine in most cases, especially if used with an induction type motor/generator. Varying the angle of the stator blades in the blower is the most economical control scheme. Because the motor/generator has a large startup electrical power requirement, a steam turbine may be used to bring the train up to speed. The turbine will normally provide 50-75% of the power needed for the blower. Once the train is close to design speed, the motor can be started without using excessive amounts of electricity. This in turn decreases the size of the transformers and switch-gears needed. When the expander is running the turbine is allowed to freewheel, or may


be used to provide extra power. If there is no steam turbine, the expander turbine may be used to "bootstrap" the train up to speed. In unit revamp situations where the main air blower is not to be replaced, the power recovery train can be reduced to 3 parts; the expander, gear reducer and generator. This configuration is known as a “Gen Set” power recovery installation. In this configuration, the power recovery system is completely isolatable from the remainder of the FCC unit. The electrical power generation from the system is routed directly into the refinery power grid. The net power recovery capable through a Gen Set system is lower than a 5-body train due to efficiency losses in the switch gear, and motor. However, the capital expenditure of Gen Set systems is lower, and they can be completely isolated from the remainder of the FCC unit should there be any equipment problems with the power recovery system.


Figure 7 Flue Gas Expander Outlet

Blades Rotor

Bearing Coupling

Inlet

Shaft

FCC-E001


A typical power recovery expander performance map is illustrated in Figure 8. The power recovery system achieved commercial success in 1963. More than 30 units are in operation or under construction. It has proven to be a valuable tool in increasing efficiency and decreasing costs for Fluid Catalytic Cracking units.

Percent Expander Horsepower

Figure 8 FCC Power Recovery 1300 ºF (704 ºC)

1600 ºF (871 ºC)

1200 ºF (649 ºC) 1100 ºF (593 ºC) 1000 ºF (538 ºC) 44.7 psia (3.14 kg/cm2a)

40 psia (2.81 kg/cm2a) 36 psia (2.53 kg/cm2a) 32 psia (2.25 kg/cm2a)

28 psia (1.97 kg/cm2a) 26 psia (1.83 kg/cm2a)

Percent Flue Gas Mass Flow Rate


CHECK VALVE The check valve is installed on the discharge of the air blower. It prevents backflow of air or catalyst, which could cause serious damage to the blower. Fluidized catalyst will easily flow back through the air heater if pressure is lost. If the blower starts to surge, the large volume of the regenerator must be isolated from the blower. Figures 9-11 show the check valve and its associated equipment.

Figure 9 Blower Discharge Check Valve


Figure 10 Blower Discharge Check Valve - Side View

The special check valve is a swing style check valve with a spring loaded air cylinder that provides spring force assisted closing. An oil filled dash pot provides a damping action on opening. Construction is of heavy wall steel to resist temperature and pressure stresses, with 11-13% Cr or stainless trim. The stainless steel shaft is supported by hardened stainless steel bushings, with graphoil packing used to prevent leakage. Older designs have incorporated asbestos packing that may need to be addressed with the appropriate abatement procedures. The shaft is connected to the dashpot and to a lever arm that has counterweights that support 75% of the disc weight. These weights minimize the pressure drop through the valve, but should never hold the disc open when there is no air flow. The lever arm is usually cut to the proper length in the factory, and the weights set in the field.


The air cylinder consists of a small piston opposed by a spring. Under normal conditions, air is supplied to the piston, which moves up to compress the spring. The piston rod moves freely between two small lever arms attached to the check valve shaft. As the piston rod rises, the valve is free to open. Air flow from the blower forces the valve open. The air to the piston is supplied through a three-way valve that will vent off cylinder pressure when actuated. For older units, a shutdown of the blower would signal the valve to cut off the air supply. For the new units that have venturi meters on the blower discharge line, a low flow signal will signal the valve to cut off the air supply. A shutdown on the blower itself, low blowerregenerator differential pressure, or low air flow can all be configured to cut off the air supply to the piston and activate the special check valve. Instrument air failure will also vent off pressure from the cylinder. Upon initial venting, the spring provides a sharp thump to the valve shaft to help free the disc in the event that it has become slightly stuck in the open position. The force of the spring is not enough to close against the normal operating air flow from the main air blower. As such, a spurious activation of the special check valve, i.e., a loss of instrument signal to the solenoid valve, would result in a higher pressure drop through the check, but would not force a unit shutdown. After initial actuation, with no air pressure to oppose it, the spring provides a constant load on the valve shaft. As the piston rod comes down, it pulls on the lever arms, which exert a closing force on the shaft. This provides a starting boost to close the check valve and will bring the valve closer to the seat before the air flow to the regenerator actually stops. Following a solenoid trip, the three-way valve must be manually reset in the field. This functionality is included in the system design to help ensure that movement of the check valve disc is controlled and stable, rather than a sporadic situation that would result if the air cylinder was pressured and depressured in a random fashion during an upset. Typical turnaround maintenance on the special check valve includes maintenance of the air cylinder, refilling the dashpot oil, repacking of the stuffing box hinge


assembly, replacement of the accordion type protective piston rod cover. After the check has been reassembled, the flapper should move freely and close normally under its own weight. Proper seating of the check disc should be confirmed internally with visual inspection. If the main air line is too small to facilitate internal visual inspection, the approximate position of the disc can be verified by the position of the counter-weight arm.

DASHPOT The dashpot provides a resistance to a sudden opening of the check valve. It has a loose fitting piston that rides in an oil filled cylinder. The valve in the bypass line restricts oil flow from the top of the dashpot to the bottom as the piston moves up. This restriction prevents the check valve from opening too quickly. As the piston moves down on the check valve closure, the valve in the bypass on the dashpot opens wide to allow rapid closing of the check valve. Some snubbing action remains to prevent excessive slamming. The dashpot should be filled with a light lubricating oil such as SAE 10W. It is also important to provide a "volume leg" in the oil piping to account for the volume of the piston shaft in the closed position. The setting on the oil dash pot needs to be verified on initial installation and subsequent turnarounds by opening the disc and allowing it to fall closed. Proper setpoint of the oil valve should result in a smooth controlled closure of the disc with no substantial impact on the valve seat. If the check valve is thrust open under low air flow conditions, it may fly up too far, and then slam back onto the seat. This can cause damage to the valve seating surfaces. As a result, the check valve should be examined for any unusual wear, such as impact erosion on the seat during the turnaround.


DIRECT FIRED AIR HEATER The direct fired air heater (Figure 12) is a carbon steel, internally insulated vessel that heats the air to the regenerator primarily during startup. A pilot burner, as well as one main burner (two - on rare occasion), are used to fire fuel gas, LPG, or oil on heater discharge temperature control. The pilot for the burner is lit by a high energy electrical electrode igniter. Air is directed to the burner by a large damper, controlled externally with a hand crank. Sight ports are provided for flame observation. Air purges to the sight ports keep the glass cool and can be used to blow catalyst out of the ports if it backs up from the regenerator. A block valve is provided to shut off and isolate the ports when the burner is either being ignited, or not being used. On heaters that are mounted vertically, directly beneath the regenerator, there are several stainless steel baffles spaced approximately three inches apart at the outlet. The baffles prevent flame impingement on the air grid in the regenerator, which could cause extensive damage to the grid. The baffles should be routinely inspected during each turnaround. The extent of the direct fired air heater repair work required during a turnaround will depend on the hours and type of use. Over-firing and burner misalignment are the two main causes of refractory spalling and reduced equipment life. On low Coke operating units, the regenerator temperature can be cool enough to adversely affect regenerator performance. On occasion, some refiners supplement the regenerator temperature by Auxiliary firing of the air heater during normal operation. While this has proven effective for some refiners, care must be taken not to exceed maximum recommended exit velocities on the main air distributor. Operating with distributor jet velocities too high can result in excessive erosion to the main air distributor as well as excessive catalyst fines generation in the unit. The fuel source to the DFAH needs to be maintained within the fuel specifications outlined by the vendor. Improper fuel/air/burner combinations can result in severe mechanical damage to both the DFAH and the regenerator internals; i.e., accidental injection of liquefied LPG through a fuel gas burner.


Figure 12 Direct Fired Air Heater

Air Splitting Damper with Limit Stops

Air Inlet

Air Outlet

TI's (2 Required)

Pilot/Ignitor Assembly

Air Purge

Air Purge

Main Gas Sight Port (2 Required Must Sight Pilot and Main Burner)

Air Purge

Air Purge and Blast Connection

Baffle

4" (100 mm) Vibrocast Insulating Refractory

Sight Port Sighting Burner

Sight Port Sighting Opposite Wall

4" Vacuum Cleanout Connection Manway

FCC-E002


AIR GRID The original design for a conventional regenerator air distribution system was a perforated plate. Air entered the base of the regenerator and then passed through a large number of small holes in a large metal plate. In the late 1970's and early 1980's, the processing of more contaminated feedstocks increased air demand and regenerator temperatures. Increased incidents of erosion, full CO combustion, and the introduction of new regenerator designs lead to significant changes in the design of the air grid. The bubbling-bed regenerators feature a high catalyst entry point into the dense bed relative to the air grid. Two basic types of air grid designs are used by UOP in this style of regenerator: pipe grid and mushroom grid. The high-efficiency regenerators have a low catalyst entry point relative to the air grid. On this design, UOP uses a pipe grid. The mushroom grid with extension arms is used in situations where a standpipe inlet is below the air grid. A dome grid was used previously, but an exit in the grid was needed to transfer catalyst to the regenerator standpipe. The mushroom air grid with extension arms distributes air through jets located in the dome and arms. A side view of the grid is shown in Figure 13. Figure 14 shows a plan of the mushroom grid with arms. The mushroom grid is constructed with 1" (25mm) lining on the dome and ¾" (19 mm) lining on the extension arms. The lining minimizes the thermal stresses on the grid wall resulting from the temperature differential between the inlet air and regenerator.


Figure 13 Mushroom Air Grid with Extension Arms


Figure 14 Plan View of Mushroom Air Grid with Arms

Plugged Nozzle Open Nozzle


The dome air grid distributes air through jets located on its dome. A side view of the grid is shown in Figure 15.

Figure 15 Dome Air Distributor


Some dome-type air distributors have experienced jet erosion. UOP believes that the erosion on this type of air grid is due to catalyst from the spent and recirculating catalyst standpipes impacting directly onto the surface of the distributor. When the catalyst impacts the air grid in this fashion, the catalyst can be forced into the jets resulting in erosion as the catalyst is blown out of the jets. Erosion can also occur on the external portion of the jet from the catalyst impact. Several methods of combating this type of erosion have been developed. One method is to install extended catalyst deflectors, which distribute the catalyst over a wide cross-sectional area of the regenerator to minimize the localized impact of catalyst onto the grid. Another modification is to cover the surface of the air grid with an abrasion-resistant lining so that the outlet of the jets is flush with the abrasionresistant lining. The dual-diameter jets have also been replaced with singlediameter, higher velocity jets. Because of mechanical reliability the pipe grid is the most commonly designed type of air grid today. The pipe grid distributes the air through two to four large laterals into a number of small branches. A side view of the grid is shown in Figure 16. Figure 17 shows a plan of the pipe air grid. Modern air grids use a dual diameter jet (Figure 18) with a restriction orifice at the inlet. This allows a higher pressure drop for better air distribution while minimizing the velocity out of the jet for minimum catalyst attrition. Air grids are designed for a total pressure drop between 0.8 and 1.2 psi (0.06-0.085 kg/cm2). The pressure drop must be maintained above 0.5 psi (0.035 kg/cm2) to achieve even distribution and should be below 1.5 psi (1.05 kg/cm2) to minimize main air blower discharge pressure. Many of the pipe grid and supports designed for partial combustion units were low alloy, such as 5% chrome. The higher temperatures encountered in a most modern full CO burning units require 304 SS.


Pipe-grids designed by UOP in the early 1970's had experienced cracking at certain butt welded joints, but otherwise worked well. UOP has modified the pipe-grid design and the advantages of this style of air grid are as follows: 1)

by using a 90° elbow (instead of a 45° lateral arm), thermal stress in the air grid is greatly reduced because the elbow has greater flexibility

2)

the 90° elbow is attached to the header arms and main hub with extruded connections which moves the welds away from the highly stressed junction

3)

the branch arms pass through the header arms which increases the strength of these joints

4)

external abrasion resistant lining protects against erosion and also provides a smooth thermal gradient

Figure 16 Pipe Air Grid


Figure 17 Plan of Pipe Air Grid

Figure 18 Dual Diameter Jet Detail


A pipe grid is subject to severe catalyst erosion on the upper surface of the branches. For this reason, older units had stainless steel retaining dams, Figure 19, welded on top of each lateral. These dams held an insulating layer of catalyst which alleviated both erosion and possible heat damage problems. Currently, UOP designs the air distributor so the entire surface of the branches have abrasion resistant lining. The lining provides both the erosion resistance and the thermal barrier required.

Figure 19 Coffer Dam Stiffener

Branch


The jets on the pipe air grid and on the arms of the mushroom grid point down to avoid channeling the catalyst bed, as seen in Figure 18. It should be noted that some of the jets are normally plugged. This is done for three reasons: 1.

To prevent a blast of air from impinging directly on an air pipe or the regenerator wall.

2.

To increase the pressure drop to the proper level if the grid has too many holes (future design case).

3.

To properly distribute air across the full regenerator cross section to compensate for spent catalyst maldistribution.

The pressure drop across the air grid can be calculated with the equation: P =

2.238 * W2 P * V2 = Cd2 * Ah2 *  2200 * T

where: P P T V

= = = =

W = Cd = Ah =

Grid pressure drop, psi Air pressure to grid, psia Air temperature to grid, ° R Velocity of air through jets: flow rate of air/total cross sectional area of all open jets, ft/sec Air Flow, lb/sec Orifice Coefficient Orifice Cross Sectional Flow Area, in2



Flowing Air Density, lb/ft3

=


REGENERATOR The regenerator is an internally lined carbon steel vessel. The lining, called refractory, is a concrete type material which is gunned onto reinforcing support anchors. This lining is necessary to protect the metal wall of the vessel from the high temperatures at which the regenerator operates and should keep the outer shell of the regenerator below 650°F (343°C) at all times. The refractory is applied over stainless steel hexmesh or steerhorn anchors. Different grades and depths are applied depending on the service. In general, four - five inches (100 - 125mm) is used in the regenerator when insulation is of primary importance. Abrasion resistant refractory lining are used on all internal surfaces in the regenerator to protect the base metal from the erosive environment. ¾ - 1 inch (19 25mm) of lining is typically used and is anchored by stainless steel hex mesh anchors. This refractory is much harder and denser than the insulating refractory so that it is provides more erosion protection but does not offer the same insulating properties. Instrument connections are inserted through the refractory. Thermowells, which are used to measure catalyst or gas temperatures, are hard surfaced with a cobaltchrome stellite hard surfacing to protect them from the erosive conditions. Pressure taps (and pressure taps used as level indicators) are protected by steam, gas, or air purges. These are discussed later in this section. The purges provide a buffer between the catalyst bearing gas in the regenerator and the small instrument taps which can easily plug. Figure 20 shows a conventional (bubbling bed) regenerator in detail. The air flows from the grid up through a dense bed of catalyst where the carbon is burned off. The catalyst enters the vessel from the spent catalyst standpipe, at the end of which is a deflector baffle to distribute the catalyst evenly over the bed, not straight down to the outlet. The design shown does not have an internal catalyst hopper, which is a large diameter cone above the regenerated catalyst standpipe. The higher density catalyst in the cone provides extra head pressure in the standpipe if needed by a particular unit.


Combustion gases, excess air, and catalyst particles traveling from the dense phase to the dilute phase are separated using two-stage cyclones. Flue gas leaving the cyclones enters a plenum chamber at the top of the regenerator. The hot gases travel through the double-disc slide valves, which are set to regulate the reactorregenerator differential pressure. The flue gas then travels through the orifice chamber, where its pressure is dropped through a series of perforated plates. Finally, the energy of the flue gas is recovered in a CO boiler (or steam generator for a full combustion unit) where the CO is burned along with auxiliary fuel gas and air to generate steam (or simply cooled to generate steam in a full combustion unit).


Figure 20 Conventional Regenerator THERMOCOUPLES (1each cyclone) CYCLONE SUPPORTS REGENERATOR PLENUM

REFRACTORY LINING

MANWAYS

LEVEL AND PRESSURE TAPS TI’s FIRST STAGE CYCLONES REFRACTORY LINING

SECOND STAGE CYCLONES

T

EXTERNAL LINING TRICKLE VALVES TORCH OIL

OPEN PRIMARY CYCLONE DIPLEG TERMINATIONS

SPENT CATALYST DEFLECTOR

SPENT CATALYST STANDPIPE LEVEL AND DENSITY PRESSURE TAPS

MANWAY

AIR DISTRIBUTOR

CATALYST WITHDRAWAL

REGENERATED CATALYST STANDPIPE

FCC-E003


TWO STAGE REGENERATOR The two-stage regenerator (RFCC) is used for units with heavily contaminated resid feed stocks. The coke deposited on the catalyst is burned off by air distributed through a grid mounted at the bottom of each stage. The spent catalyst enters the upper, first stage regenerator which operates in partial combustion mode to minimize the heat of combustion. Approximately 70% of the coke on the spent catalyst is burned off in this stage. The catalyst is then transferred to the lower, second stage regenerator through the recirculation catalyst standpipe. The second stage operates in full combustion mode with excess oxygen to completely remove the remaining carbon from the catalyst. This combination provides the heat balance advantages of partial combustion operation with the advantage of low carbon, high activity regenerated catalyst. In the second stage regenerator, air is typically distributed through pipe grid distributors although dome air distributors can also be used. In the upper regenerator, the air distributor is typically the mushroom-and-arm type. Arms are radially arranged around a central dome. The skirt which holds the upper air distributor in place physically separates the two stages. Vent tubes in the skirt allow the transport of combustion gasses and excess oxygen (with some catalyst) from the second stage to the first stage. A two-stage regenerator is shown in detail in Figure 21. The recent RFCC design uses multiple pipe air grids in the first stage regenerator entering through the cone rather than a single dome grid entering through the second stage. This is a mechanically simpler design which eliminates the need for a complex expansion joint on the air line. This also allows individual control to the air grids in each section of the first stage. This configuration is shown in Figure 22.


Figure 21 Two Stage Regenerator Flue Gas

Primary Cyclone

Secondary Cyclone

Mushroom Grid Distributor

Spent Catalyst

First Stage Regenerator Spent Catalyst Distributor “Ski-Jump”

Recirculation Catalyst Standpipe

Catalyst Cooler

Second Stage Regenerator (Side View)

Vent Tubes First Stage Air Inlet

Second Stage Regenerator

Cooled Catalyst Standpipe

Regen Standpipe Hopper

Regenerated Catalyst

Pipe Air Grid Distributor 2nd Stage Air


Figure 22 Updated RFCC First Stage Air Grid Design

1st Stage Regenerator

First Stage Air In

2nd Stage Regenerator

FCC-E004


RFCC units will also have one or more catalyst coolers. The catalyst cooler slide valve position is set by the signal from the second stage regenerator temperature controller or the slide valve differential pressure controller over-ride via a low signal selector. The recirculating catalyst slide valve position is set in a similar manner, by signals from the lower regenerator level controller or the slide valve differential pressure controller through a low signal selector. Each slide valve packing is steam purged like the regenerated catalyst slide valve. In the first stage the flue gas and catalyst are separated by two-stage cyclones. The catalyst falls down the cyclone diplegs which are submerged in the catalyst bed to provide a seal against gas passing up the diplegs. The primary cyclone diplegs are typically open ended pipes with a splash plate and the secondary cyclones have trickle valves. The catalyst flows into the annular zone around the mushroom air distributor and into the recirculation catalyst standpipe and catalyst cooler(s). Air is directed through nozzles on the underside of each arm of the upper air distributor to help maintain proper fluidization of catalyst in this area.


HIGH EFFICIENCY REGENERATOR The high efficiency regenerator approximates a plug flow burning profile for the catalyst as opposed to the back-mix regime of the standard bubbling bed design. Because the high efficiency design burns essentially all of the CO in the regenerator, there is no need for a CO Boiler and there is typically very little afterburning. The plug flow burning profile also results in much lower NOx emissions than a bubbling bed. The spent catalyst from the reactor mixes with the blower air and roughly an equal amount of recirculating regenerated catalyst at the bottom of the regenerator (combustor). The recirculating flow of catalyst is necessary because the spent catalyst at 925°-1025°F (495°-550°C) is not hot enough to initiate and complete burning in a reasonably sized vessel. The mixing takes place in the lower part of the regenerator, called the combustor or in a mixing riser, depending on the design. Figures 23 and 24 show the two designs. These were developed to obtain the best regeneration, with vessel cost and maintenance considered. The coke burns off the catalyst as it travels up the combustor riser with the air. There is a rough separation at the top of the riser through a "tee" shaped outlet. The flue gas goes up to a twostage cyclone system and out to energy recovery. The catalyst is returned to a dense phase. From here the flow splits, part of the catalyst going to the base of the reactor riser, and the rest back to the combustor. The high efficiency regenerator has most of the same fittings as the conventional bubbling bed design. The torch oil nozzles, instrument connections, and catalyst loading lines are positioned differently because of the different regenerator configuration, but function in the same manner as those in a conventional unit. If the unit is equipped with a flue gas power recovery system, there will be spray nozzles in the plenum chamber for emergency cooling if the temperature at the inlet to the expander exceeds the safe limit of the machine.


Figure 23 UOP Fluid Catalytic Cracking Process High Efficiency Regenerator System

Cyclones

Upper Regenerator Combustor Riser

Combustor

Regenerated Catalyst Fluffing Air Ring Spent Catalyst

Mixing Zone Air Grid

Spent Catalyst Distributor (Ski Jump)


Figure 24 High Efficiency Regenerator with External Mixing

Upper Regenerator

Combustor Riser

Combustor Recirculation Catalyst Standpipe

Regenerated Catalyst Catalyst/Air Distributor Spent Catalyst Lift Riser

Air


COMBUSTOR CONE Figure 25 illustrates the changes in regenerator cone design that have taken place since the mid-1980s. UOP has developed a new cone detail (see Figure 26) that eliminates the radial and axial constraint imposed by the hard refractory lining and cold regenerator shell. The resulting stress is dissipated by incorporating a flexible soft pack ceramic lining that permits both radial and circumferential expansion. Additionally, the new cone detail minimizes thermal stresses due to radial expansion by providing a flexible skirt section that connects to the regenerator wall. The key feature of the new internal combustor cone design is the air space incorporated between the cone skirt and the regenerator shell. The air space is required to provide the optimal heat transfer medium between the cone and the regenerator shell. By using this air space, it has been proven that the thermal stresses in the cone are less than the stresses compared to other internal cone designs that have experienced deformation. In order to maintain this air space, it is mandatory to keep the air gap free of catalyst. The catalyst seal device achieves this objective (see Figure 27). Some characteristics of the catalyst seal are as follows: 1) 2) 3)

the seal is not tight; the gap operates at regenerator pressure the seal design supports the weight of the catalyst within the regenerator the seals allows thermal expansion of the cone


Figure 25 Combustor Cone Modifications

UOP 1906H-7 UOP 3110-4


Figure 26 Detail of Combustor Cone Design

Fiberfrax Moist Pak-D Abrasion Resistant Lining

Refractory Lining

Catalyst Seal Device (Figure 26)

Ceramic Fiber Blanket Insulation

Retaining Ring Air Space


Figure 27 Detail of Catalyst Seal Device Ceramic Fiber (2 layers)

3.5” Sch. 10S Pips

4” (Cold Position) 2” (Hot Position)


COMBUSTOR RISER ARMS Several old-style high-efficiency regenerators have experienced excessive deformation of the regenerator riser arms. The rectangular openings in the bottom of the arms have "ovaled" and sagged. The arms have rotated at the leading edge of the opening and have sagged 1 ft. to 2 ft. (300-600 mm) below the guides on the regenerator wall. The primary cause of the deformation is attributed to high-temperature creep relaxation. During the late 1970's, the original circular openings in the arms were enlarged to rectangular openings to improve catalyst separation efficiency. This larger opening significantly decreases the inherent bending strength of the arm. Several modifications to existing units have been implemented. The first method is to install stiffeners to the arm to reinforce the opening. The second method is to install a tension member from the riser to prevent excessive deformation. The new-style riser arms have a modified geometry. The arms have been designed as oblong members to improve strength. Additionally, a greater number of shorter, smaller diameter arms are used (Figure 28). The shorter arm reduces the bending stress resulting from weight. The increase in the number of arms improves the flow characteristics in the regenerator by more uniformly discharging the catalyst across the regenerator's cross-sectional area.


Figure 28 New Style Combustor Riser Arms


FLUFFING AIR DISTRIBUTOR To improve fluidization in the upper regenerator and flow into the standpipes a fluffing air ring (Figure 29) is installed in the cone. An alternate source of fluffing air other than the main air blower is now specified for the upper air distributor. On designs where the fluffing air was provided by the main air blower, circumstances could arise where insufficient P was available for good fluidization of the catalyst.

Figure 29 Fluffing Air Distributor

STANDPIPES

ABRASION RESISTANT LINING

ABRASION RESISTANT RESTRICTION ORIFICE

LINING PIPE

3/4 " XX-STRONG PIPE DISTRIBUTOR JET


TORCH OIL NOZZLES The torch oil nozzles shown in Figure 30 provide a means of injecting heavy oil into the regenerator when extra heat is needed (e.g. during the startup). The oil is sprayed into the regenerator with atomizing steam through a special nozzle made of tungsten alloy to withstand the high temperature. These nozzles may be retracted through a packing gland if they need to be cleaned. Steam is continuously injected through a 1/8" (3mm) restriction orifice to the annular space around the nozzle to keep the area clear of catalyst, which could pack up and prevent retraction of the nozzle. Steam is also continuously injected through the nozzle tip to keep it cool and prevent plugging with catalyst. Excess steam may contribute to erosion and catalyst breakup. The nozzle should be marked so that after cleaning it can be returned to its proper position - recessed ¼" (6 mm) back from the refractory face. This position allows the oil spray to miss the regenerator wall, yet protects the nozzle. During a turnaround, the various steam and torch oil nozzles should be inspected. If there has been too much purge steam around the barrel there may be erosion problems in the refractory surrounding it. The nozzles should be checked for wear and cleanliness. If the nozzle was not recessed the proper ¼" (6mm), there will probably be severe metal loss. Refractory damage may indicate the nozzle was recessed too much. Nozzle positions should be checked when they are replaced.


Figure 30 Torch Oil Nozzle

SPRAY NOZZLES Spray nozzles are used to inject water through an atomizing nozzle to cool off the flue gas in the plenum chamber if the unit has power recovery. The spray water should be clean, such as steam condensate. Contaminants such as sodium will cause problems by deactivating the catalyst or contributing to its breakup. Mechanically, the spray nozzles are similar to the torch oil nozzles.

CATALYST COOLER The ability to control and vary the amount of heat removed from the regenerator creates an additional degree of freedom by moderating the regenerator temperature as a limiting constraint. The catalyst cooler provides a variable heat sink, which


allows the refiner to vary the catalyst/oil ratio, reactor temperature, and feed temperature independently of one another. The catalyst cooler tube bundle is inserted into a refractory lined shell off the side or bottom head of the regenerator. The tubes of this exchanger are the bayonet type. The boiler feed water enters the cooler through the inner tubes and the mixture of water and steam exits the cooler through the annulus between the inner and outer tubes. The outer tubes are 3 inch (75mm) O.D. made from 1¼ Cr, ½ Mo seamless tube material. The inner tubes are 1-3/8 inch (35mm) O.D. made from carbon steel seamless tubing. The stainless steel fluidizing air lances distribute air into the cooler near the bottom of the tubes. The air creates turbulence and increases heat transfer coefficient as the bubbles travel upward. The backmixing created by the bubbles also brings hot catalyst into the cooler from the regenerator. The air is delivered to a common manifold supplying all the lances through a flow controller. The lances contain a restriction orifice, located near the piping header at the top of each lance, to help distribute the air uniformly over the cross sectional area of the cooler. The countercurrent fluidizing air improves heat transfer by creating turbulence and mixing in the region of contact between the hot catalyst and the tubes. A differential pressure transmitter, with taps located above and below the cooler, gives a direct indication of the density of the fluidized catalyst at various conditions of catalyst flow and air injection. Mechanical reliability is achieved by locating the cooler in the dense phase of the regenerator. In the dense phase, the heat transfer coefficient is higher which permits lower catalyst and fluidization air velocities. Lower velocities minimize erosion within the cooler. In addition, the cooler tubes are located in the vertical plane. This feature generates a uniform heat transfer coefficient over the entire tube surface thereby preventing uneven surface temperatures which cause localized stress. Catalyst coolers have been designed and built to fit virtually every regenerator configuration, including single-stage bubbling beds, high-efficiency combustors, and two-stage regenerators. Three basic styles of UOP catalyst coolers are currently available:


Flow-through catalyst cooler – The catalyst flows downward into the cooler shell and exits into the cooled catalyst standpipe near the bottom of the tube bundle. The standpipe transports the cooled catalyst through a slide valve and expansion joint into the combustor on a high efficiency regenerator or to the second stage regenerator on an RFCC. In a single stage bubbling bed regenerator the catalyst can be lifted back into the regenerator with air through a lift riser. Both the catalyst flow through the cooler and the fluffing air rate are used to control the cooler duty. Backmix catalyst cooler – This style contains no catalyst exit standpipe. Hot catalyst enters the cooler by backmixing as a result of fluidization air injected near the bottom of the tube bundle. The major advantage of this cooler design is that no slide valve, expansion joint, or standpipe is required. This configuration also permits the cooler to be lower to the ground if elevation is a limiting constraint. The duty of a back mix cooler is ~60% of an equal sized flow through cooler and is controlled only with the fluffing air. Hybrid catalyst cooler – The combination of flow-through and backmix operation constitutes the hybrid catalyst cooler. In a hybrid, the catalyst exits into a standpipe located at the midsection of the tube bundle (instead of at the bottom as in flowthrough coolers). In the hybrid cooler, the upper portion of the bundle operates in the flow-through mode, and the bundle length below the catalyst outlet operates in the backmix mode. This configuration achieves somewhat less heat-removal capacity than a full flow-through cooler but still transfers cooled catalyst down to the lower portion of the regenerator. The catalyst cooler steam generation circuit includes the cooler, steam drum, and circulation pumps. Boiler feedwater is pumped to the bottom head, enters the inner tubes, then flows down through the annulus between the inner and outer tubes where it absorbs heat to generate steam. The steam-water mixture leaves the catalyst cooler to be separated in the steam drum. Makeup boiler feed water is delivered to the steam drum through a flow controller which is cascaded to signals from the drum level and steam generation flow transmitters. Steam flows from the drum through a stop check non-return valve and a superheater (either part of the flue gas cooler or a fired heater) before entering the refinery steam header.


Figures 31-35 show some examples of catalyst coolers and catalyst coolerregenerator configurations that have been constructed.

Figure 31 Flow Through Catalyst Cooler UOP Catalyst Cooler General Arrangement


Figure 32 UOP Backmix Catalyst Cooler


Figure 33 Examples of Bubbling Bed Regenerators with Catalyst Coolers

Flow Through

Backmixed

Aeration Air

Aeration Air

Lift Riser Cooled Catalyst Standpipe

Water and Steam

Water and Steam

Water

Water Air

Lift Air Distributor


Figure 34 Examples of High Efficiency Regenerators with Catalyst Coolers

Cone Mounted Flow Through

Side Mounted Hybrid

Cone Mounted Backmix


Figure 35 RCC Regenerator with Catalyst Coolers Example of an RCC Regenerator with Multiple Catalyst Coolers

Fluffing Air

Fluffing Air Air

Backmix Catalyst Cooler

Flow-Through Catalyst Cooler

Air

UOP 2119-27


Operating the catalyst cooler with sufficient water circulation to ensure that the tube walls are always wet so that they can not overheat and limiting the fluffing air so that the tubes are not subjected to erosion are critical in ensuring mechanical integrity and long life of the cooler. In new units the flow controller regulating the water flow to the cooler has been eliminated so the water flow is set by the pump curve. The spare pumps are instrumented to start automatically on low water flow. If the water flow is not recovered by the auto start the cooler is shutdown by closing the cooled catalyst slide valve and the fluffing air control valve. The fluffing air rate should never exceed a flow that will result in a superficial velocity in the shell of more than 1 ft/sec (0.3 m/sec). In recent years, as catalyst cooler reliability has improved, UOP has shifted from a reactive to a proactive approach to catalyst cooler design. Several design modifications have been developed recently. Tie Rod Support Previous designs had the tie rods, which support the eggcrate bracing, protrude through a hole in the upper tubesheet. The tie rod was welded to the tubesheet on the steam (bottom) side. While this method was acceptable under normal operating circumstances, the tie rod could potentially push through the tubesheet if tie rod growth was restricted due to an obstruction or thermal binding. This would create a path for the steam to enter into the catalyst side of the cooler, having the same effect as a tube leak. The current design eliminates the hole through the tubesheet, using instead a cup into which the tie rod is inserted (see Figure 36). This cup is countersunk and welded into the catalyst (top) side of the tubesheet. The tie rod is welded around the rim of the cup above the refractory face providing easy access for bundle maintenance or refurbishment. This design eliminates the possibility of the tie rod creating a steam leak.


Figure 36 Modifications to Tie Rod Supports

Tie Rods

Bracing Bars

Egg-Crate Support Bracing

Refractory Lining

Old Method

Upper Tubesheet

New Method


Tie Rod / Eggcrate Connection Several recent catalyst cooler inspections have revealed tie rod deformation, usually more severe in the upper section between the eggcrate supports. While this by itself had not caused any operational problems, it suggests that the fit between the eggcrates and the outer tubes is extremely tight. Thermal expansion of the eggcrates causes binding on the outer carbon steel tubes which are water cooled. The hot stainless steel tie rods expand a greater amount and do not have adequate strength to move the eggcrates as they are intended. Their thermal growth being restricted, the tie rods buckle. As a result, the eggcrates are no longer welded directly to the tie rods which allow them to expand independently of each other (see Figure 36). The eggcrates are supported by minimal friction on the tubes themselves. The hot tie rods are free to expand within the eggcrates. As a precaution, large washers are welded to the tie rods a few inches above and below the eggcrates to limit any unexpected movement of the eggcrates either during operation or bundle handling. Flat Tube Caps The fluffing air headers and lances are supported by arms that are welded near the top of the air lances (see Figure 37). Originally, each arm was welded to a stainless steel pipe support stool that was welded to the outer tube hemispherical tube cap. Later, the arm to support stool weld was removed in order to allow for a greater degree of flexibility for the fluffing air headers and lances. In the current design, the support stools have been replaced with flat tube caps (see Figures 37 and 38). These new tube caps are machined bar stock that are rounded on the inside and flat on the top. The removal of the pipe support stools has several advantages : • • • •

A weld to a pressurized tube cap is no longer necessary. A postweld heat treatment step has been eliminated. Tube alignment is easier. A more uniform surface on which the air lances can rest has been created.

The small additional cost of the flat tube caps is recovered in assembly time.


Figure 37 Modifications to Outer Tube Caps Aeration Pipe Header Aeration Lance Lance Support Arms Pipe Support Plate Pipe Support Stool

Old Method

Flat Tube Cap

Outer Tubes

New Method


Figure 38 Flat Tube Cap

Inner Tube Wall Thickness The wall thickness of the inner tube has been increased on most recent catalyst cooler designs. The resulting smaller diameter increases the inlet water pressure drop which ensures a uniform distribution of water to all of the tubes and guarantees the tube cap and inside wall of the outer tube are fully wet. Eight-Foot Cooler As FCC and RFCC units become larger and are required to process very heavier resid, the heat removal demand increases. In anticipation of this, UOP now offers an eight foot diameter cooler (96" ID of shell). This cooler can provide about 40% more duty than the typical seven-foot design. This is particularly beneficial when the


required duty is slightly higher than the duty of a seven foot catalyst cooler, making a single eight foot cooler a less expensive alternative to two smaller coolers. Update: Debris Screens Since 1991, debris screens have been installed in all existing units. Debris screens, which are installed at the entrance to the cooler, have become standard supply for all coolers mounted on the lower regenerator cone or head. The screens prevent large refractory pieces or other loose debris from accumulating at the bottom of the tube bundle where the debris can restrict or divert the flow of air from the air lance, potentially causing catalyst impingement on a tube and an eventual tube leak. Aside from some minor improvements in the anchoring method, the screens have held up well in operation and have performed their function. There have not been any tube leaks caused by accumulated debris in coolers with debris screens. Update: Air Lance Pressure Testing Because the internal air piping and lances are not subject to code requirements, UOP implemented a required shop pressure test of the completed aeration assembly in 1991. Since that time, there have been no reported air leaks or weld failures in any operating coolers that have received this testing. Prior to the shop testing, air leaks had occurred in at least five units, three of which lead to tube leaks.


CYCLONES The flue gas leaving the dense bed will entrain some of the catalyst particles with it. Some of these settle back to the bed; others are carried higher. Measurements of the entrained particles taken at increasing heights above the bed show a gradual decrease in the amount of fines entrained. At some point above the bed, the concentration of particles entrained levels out. This is called the Transport Disengaging Height, or TDH. Early FCC units used a number of small tubes, called multi-clones to remove particles. These were not particularly effective and were difficult to maintain. The cyclone design shown in Figure 39 was the next step. The shaveoff was intended to increase efficiency, but also proved difficult to maintain. The cyclones used in the modern FCC unit use a fairly simple principle to remove most of the particles. See Figure 40. The catalyst bearing gas enters a cylinder through a tangential opening. The catalyst is 500-1000 times as heavy as the gas, and is subjected to forces several hundred times that of gravity as the gas swirls around the cylinder. The larger particles are removed through centrifugal forces which force the particle outward to collide with the wall. The collisions slow down the particle so that they fall by gravity into the dust hopper and are returned to the vessel through the diplegs. The viscous drag forces of the gas tend to carry some of the catalyst particles with it. Generally only the smaller particles are light enough to stay with the gas, because the inertial and centrifugal forces acting on them are small. The catalyst separated from the gas stream swirls downward due to the force of gravity. The chamber below the entrance of the cyclone tapers downward and tends to keep the catalyst against the wall which is away from the cleaner area at the center core (where the gas disengages and moves up). There is more disengaging area in the hopper, which feeds catalyst to the dipleg; this disengaging area also decreases the amount of erosion which could be created by the vortex of the catalyst particles.


Figure 39 Cyclone-Catalyst Fines Collector with Shaveoff Catalyst-Free Gas Outlet

Gas Out Catalyst Shave-Off

Bypass “Catalyst Shave-Off” Bypass Catalyst-Laden Gas Inlet Re-Entry Opening

Note: Because of the High Maintenance Required on the Catalyst Shave-Off (Caused by Erosion) many Refiners are Choosing to remove this Device

Catalyst-Laden Gas Inlet Stream PatternLower Portion Stream PatternUpper Portion (Principally Finer Particles) Disengaging Hopper Catalyst Outlet Dip Pipe UOP 2119-28


Figure 40 Cyclone-Catalyst Fines Collector

There is a pressure drop between the cyclone inlet and dipleg outlet. If the catalyst is to be returned to the bed, the dipleg must hold a head of catalyst sufficient to overcome this differential. If the catalyst in the dipleg stops flowing, the gas will simply carry the catalyst out the top. If the dipleg is submerged in the bed, then the pressure at the bottom will increase and catalyst will be forced out as the level in the leg rises. The required head will determine the dipleg length. There are two general types of dipleg termination devices – the trickle valve and counterweighted flapper valve. The trickle valve in Figure 41 is generally used on submerged diplegs. The valve is simply a flat plate held closed by gravity and external pressure until the catalyst head in the dipleg is sufficient to open it. The leg dumps, and the trickle valve swings shut. The counterweighted flapper valve in


Figure 42 is another method, with a better seal than the trickle valve. The counterweight holds the flapper closed until the catalyst head is sufficient to open the flapper. UOP has updated the design of the counterweighted flapper valve to improve its mechanical reliability and maximize the unit onstream efficiency. The major revisions to the design are: 1.

Addition of a stop bar to prevent the valve from opening more than 45° (see Figure 42).

2.

Addition of a stiffener bar on the fixed portion of the pivot mechanism to minimize warpage or movement over the course of repeated thermal cycles.

3.

Minimizing the tolerance between the concentric bushings to 3 mm plus 1.6 mm minus 0 mm (1/8" plus 0.0625" minus 0") to allow for thermal growth (see Figure 43).

4.

Requirement that the hard surfacing used on the pin and bushings be crack free. There is some concern that surface cracking may contribute to roughness and thus restrict smooth motion of the hinge mechanism. Alternative hard surfacing such as Waspalloy, Wallex 50 and Triten should be considered as options for hard surfacing of the pin and bushings.

5.

Increasing the amount of counter weight used (Table 1).


Figure 41 Trickle Valve

Hinge

Flapper Plate

Stop

3-5º From Vertical


Figure 42 Counterweighted Flapper Valve

STIFFENER

LUG DETAIL

1

1

Closed Position

Open Position


3mm

+0.0625" -0.000" +1.6mm -000mm

1 1/2 " PLUS/MIN 1/32 "ID

1/8 "

(BOTH SIDES) (BOTH SIDES)

38mm PLUS/MIN.8mmID

Figure 43 Bushing Detail

3/8 "(10mm)

7/8 " (22mm)

C

OF BUSHING

(INCLUDES 1/8 "(3mm) HARD SURFACING)

LUG ON DIPLEG

1/8 "

(3mm)

3"(75mm) DIA

HARD

1/8 "(3mm)

BUSHING

SURFACING

CONCENTRIC

DIA

2 9/32 "(58mm)

C

C

OF HOLE

OF BUSHING

7/8 "

3/8 "(10mm) (INCLUDES 1/8 "(3mm)

C

OF BUSHING

HARD SURFACING)

C

3"(75mm) DIA

HARD

(22mm)

1/8 "(3mm)

DIA

2 9/32 "(58mm)

BUSHING

SURFACING

ECCENTRIC

38mm PLUS/MIN.8mmID

1 1/2 " PLUS/MIN1/32 "ID

ECCENTRIC BUSHING

CONCENTRIC BUSHING

OF BUSHING AND HOLE


TABLE 1 FLAPPER VALVE COUNTERWEIGHT Dipleg Diameter 6” (150 mm) 8" (200mm) 10" (250mm) 12" (300mm) 14" (350mm) 16" (400mm) 18" (450mm) 20" (500mm) 22" (550mm) 24" (600mm)

Counterweight Lbs / Kg 1.9 3.3 5.4 7.8 9.6 12.8 16.4 20.5 25.0 30.0

0.9 1.5 2.4 3.6 4.4 5.8 7.4 9.3 11.4 13.6

Two important factors in cyclone efficiency are the velocity of the gas and the size distribution of the particles. In general, the higher the velocity, the higher the efficiency. It should be remembered, however, that higher cyclone velocities may mean higher vessel velocities, with more catalyst carried up to the cyclones. If these are larger particles, most of them will be collected, but in extreme cases, the cyclones could be overloaded and the larger particles lost. Higher velocities also means higher rates of erosion and catalyst attrition. The solids distribution is not under immediate control, except by minimizing attrition of the catalyst by avoiding areas of high velocity. In general, the small fines, less than 20 microns, will be lost, with most of the larger size particles collected. The reactor cyclones generally do a better job because here the coke on the catalyst fines makes them larger and easier to collect. The design of the cyclone is the important factor in its efficiency as well as the resistance to erosion; it is normally handled by the cyclone manufacturers. However, in the early 1980's, erosion in the cyclones and cyclone diplegs increased noticeably. Units were beginning to process larger quantities of heavier and more contaminated feeds. These feeds produce more coke, which increases air demand and regenerator temperatures. The net result is increased erosion as a result of an


increase in catalyst loading and cyclone velocities. Competition among vendors was another factor leading to an increase in instances of erosion. In some cases, vendors supplied cyclones with key geometric ratios, such as L/D, minimized for cost reasons. To address these problems, UOP began to specify cyclone geometric relationships and velocity criteria. The purpose was to ensure that all cyclone vendors bid on the same basis and that the cyclones supplied met the process and mechanical requirements set by UOP. The requirements set forth in the "Mechanical Considerations In FCC Design" paper presented at the 1996 UOP FCC Symposium have significantly reduced erosion problems. These requirements are as follows:

Single-Stage Reactor Cyclone Criteria      

 

Inlet velocity shall not exceed 65 ft/sec (19.8 m/sec). Outlet velocity shall not exceed 100 ft/sec (30.5 m/sec). Ratio of barrel area to inlet area shall be 5.5 minimum. Ratio of the main cone outlet diameter to the barrel diameter shall be 0.4 minimum. Ratio of cyclone height, measured from the roof of the cyclone to the outlet of the dust hopper cone, to its barrel diameter shall be 5.0. Projected apex point of the main cone shall terminate at a minimum distance of 0.3 x barrel diameter above the outlet of the dust hopper cone (0.5 x barrel diameter for eccentric diplegs). Ratio of the dust hopper diameter to the main cone outlet diameter shall be 1.5 minimum. Ratio of the dust hopper cone height to the barrel diameter shall be 0.45 minimum.


Regenerator Cyclone Criteria First Stage   

Inlet velocity shall not exceed 65 ft/sec (19.8 m/sec). Outlet velocity shall not exceed 80 ft/sec (24.4 m/sec). Ratio of barrel area to inlet area shall be 3.7 minimum.

Second Stage   

Inlet velocity shall not exceed 75 ft/sec (22.9 m/sec). Outlet velocity shall not exceed 120 ft/sec (39.6 m/sec). Ratio of barrel area to inlet area shall be 4.3 minimum.

Both Stages     

Ratio of the main cone outlet diameter to the barrel diameter shall be 0.4 minimum. Ratio of cyclone height, measured from the roof of the cyclone to the outlet of the dust hopper cone, to its barrel diameter shall be 5.0. Projected apex point of the main cone shall terminate at a minimum distance of 0.3 x barrel diameter above the outlet of the dust hopper cone. Ratio of the dust hopper diameter to the main cone outlet diameter shall be 1.5 minimum. Ratio of the dust hopper cone height to the barrel diameter shall be 0.45 minimum.

The reactor cyclones are normally a low alloy steel, such as 1¼ Cr, ½ Mo. Some older partial CO combustion regenerator cyclones used 5 Cr or 12 Cr, with 1¼ Cr diplegs sometimes. For the higher temperatures of modern, complete CO combustion, Type 304 stainless steel (18 Cr, 8 Ni) is used. Abrasion resistant lining is used on all cyclones to protect them from catalyst erosion. The inlet horn, the gas outlet pipe, barrel, disengaging hopper, and part of the dipleg will be lined. The


extent of the dipleg lining will depend on accessibility and prior experience; erosion in one area usually leads to a lining at the next turnaround. The exterior of the diplegs must be protected with an abrasion lining where they are subject to catalyst impingement, such as from the spent catalyst inlet or from another dipleg discharge.

ORIFICE CHAMBER Orifice chambers (Figure 44) are used on FCC units when the regenerator pressure is controlled by flue gas slide valves. The chamber is a cylindrical vessel with a series of perforated grid plates (Figure 45). These plates hold a backpressure downstream of the slide valves. By reducing the pressure drop across the valves, their operating life is greatly extended because there is no sudden acceleration of the catalyst bearing gas stream. The flow through the orifice chamber may be upflow or downflow depending on downstream equipment. There are no moving parts, so no adjustments can be made on stream. UOP designs the orifice chamber as follows: 

 

the flue gas slide valve is designed for approximately one-third of the pressure drop and the orifice chamber is designed for the remaining two-thirds of the pressure drop. enough grids are installed to limit the pressure drop across each individual grid while maintaining a specified velocity across each hole. the distance between the top grid and the flue gas slide valve is kept at a maximum. In revamp situations, this distance may be short and lining will be required for the first grid only. Additionally, the installation of a shroud at the inlet has proven to be beneficial in reducing erosion.

Given the UOP design philosophy of using cold wall construction whenever possible, UOP has developed a cold wall orifice chamber. As a result, the chamber walls are no longer prone to buckling, cracking, bulging and other phenomena associated with high temperature stainless steel design. Because of the high inlet temperature of the orifice chamber, the grids must be designed for stainless steel metallurgy (if a waste heat boiler is installed upstream, then carbon steel or low chrome can be used). The grids are supported by a


cylindrical skirt section; typically two grids are supported from one skirt section. The cylindrical skirt is supported by a tapered skirt section which is then welded to the chamber wall. The tapered skirt requires a certain length to accommodate the thermal expansion of the stainless steel grid relative to the cold wall shell. A gap is also provided between the skirt and the refractory lining to allow for this expansion. Blanket insulation is provided behind the tapered skirt to allow for the thermal deflection. The cold wall orifice chamber significantly reduces the amount of thermal growth of the flue gas line (as compared to the hot wall design). This reduction will reduce or eliminate the amount of expansion joints required. Given the change in thermal movements, as well as the additional weight (due to the refractory), the entire flue gas system should be reviewed for both support and flexibility for revamps. Inspection UOP has developed two types of orifice chambers to accommodate various revamp inspection requirements. The first option allows external manway access to each pair of top skirt grids (due to the skirt support scheme, external access to all grids is not possible). To accommodate the external manways, extra tangent length for the orifice chamber is required. In the event there are space limitations, a shorter version has been developed. Inspection access for this design requires the use of internal manways on each grid.


Figure 44 Cold Wall Orifice Chamber Inlet Shrould FLOW Internal Manways


Figure 45 Orifice Chamber Details C GRID

RING NO 5

RING NO 4

RING NO 3

RING NO 2

RING NO 1

INTERNAL MANWAY

AIR SPACE

HIGH DENSITY REFRACTORY LINING RING NO 5 RING NO 4 RING NO 3

STAGGER HOLES WHERE POSSIBLE

CERAMIC FIBER BLANKET INSULATION LINING TRANSITION

RING NO 2 RADIUS RING NO 1

C GRID

ABRASION RESISTANT LINING HIGH DENSITY REFRACTORY LINING

+ - GAP AIR


ELECTROSTATIC PRECIPITATORS Electrostatic precipitators (Figure 46) are used to remove catalyst fines from the flue gas. The equipment consists of a large rectangular steel shell which contains a number of wires called discharge electrodes, and a number of flat collecting plates. The wires and plates are hung vertically, in alternating rows. A high electrical potential is put on the discharge electrode wires, and the plates are electrically grounded. A corona discharge surrounds the discharge electrode because of the high potential. Gas ions formed by this corona move rapidly towards the collecting electrode. When an ion strikes a catalyst fine, it becomes charged. The electrical field between the electrodes causes the particle to deposit on the collecting electrode. As more particles collect, a layer of fines builds up on the plate. A mechanical rapper periodically strikes the frame of the collecting plate, and the particles are knocked off. The fines tend to agglomerate into larger particles which fall into dust hoppers under the electrodes. There is no physical change in the particles, simply an agglomeration which makes them large enough for gravitational forces to exceed the carrying force exerted by the gas. Precipitators usually operate at pressures slightly above atmospheric. Inlet temperatures range from 400-800°F (205-425°C). The temperature is especially important if the flue gas has acidic components such as SOx or NOx. These may condense and combine with water to form corrosive agents. The precipitator shell may be insulated to hold the metal temperature high enough to prevent condensation.


Figure 46

Electrostatic Precipitator Insulator Discharge Electrode Rapper

Collecting Surface Rapper Transformer Rectifier

Gas Out

Collecting Surface Gas Flow In Hopper

UOP 2119-34

Discharge Electrode

Catalyst Fines Out


Collection efficiencies range from 70-99%, with 90-95% more common. The factors affecting the efficiency are: 1. 2. 3. 4. 5. 6.

Effective voltage and other electrical factors Resistivity of fines Size distribution Gas flow rate Collector plate area Rapping cycles

As the effective voltage increases, the efficiency increases. Too high of a voltage can lead to arcing, which will damage the electrodes. Arcing can also occur if the weights holding the discharge electrode wires fall off and allow the wires to swing over towards the collecting plate. The precipitator should be designed with the proper number of power supplies and control equipment to prevent excess sparking. The resistivity of the fines is approximately constant. It may be decreased for better collection efficiency by adding small (<20 ppm) amounts of ammonia, but this is usually not required. Spray water is also used in some cases. The size distribution will be determined by cyclone performance in the regenerator. Large solids rates to the precipitator, even with high efficiencies, may still lead to emission problems, so it is better to have lower loadings to the unit. Larger particles are easier to collect, because the smaller ones are more easily carried with the gas. Typical gas flow rates are 5-6 ft/sec (1.5-1.8 M/s). Lower flow rates will give greater efficiency. A higher collecting area will also increase efficiency. The final factor, the rapping cycle, is normally designed for one strike from each rapper every 1-3 minutes. Greater power and frequency of rapping will increase efficiency, but must be balanced against cost. Both collecting and discharge electrodes are fitted with rappers, most of them positioned on the collectors.


CATALYST STORAGE HOPPER The FCC unit is normally built with two or three catalyst hoppers (Figure 47). One is for fresh catalyst, the other is for equilibrium catalyst. RFCC units or FCC units with high metals feed stocks may have a third hopper for storing low metals equilibrium catalyst. The vessels are normally carbon steel, constructed to withstand a full vacuum, although this should be checked before the first use. Many units are built with automatic gauging devices, but these are not always accurate because the float tends to sink in the catalyst. For accurate measurement of the catalyst level, the hopper should be fluffed with air from the bottom and the catalyst allowed to settle. A hand gauge can then be used to read the tons of catalyst from a chart of inventory as a function of hopper ullage. Exact settled densities for a given type of catalyst may be determined by measuring level change after a weighed amount of catalyst has been loaded or unloaded by truck or cartons. A discussion of catalyst loading and hopper operation is provided in the procedure section. CATALYST LOADING AND UNLOADING LINES The catalyst transfer line at the bottom of the regenerator is normally used to remove catalyst from the regenerator when the unit is shut down. Most of the catalyst will be loaded and unloaded through a connection (4-6 inch or 100-150 mm) in the dense bed area. For rapid transfer, a large line is used. For gradual fresh catalyst addition, a smaller line is used which connects into the large line just before the regenerator at an angle to minimize erosion. The catalyst lines are usually carbon steel. The catalyst withdrawal lines from the regenerator should be 1¼ Cr, ½ Mo (up to the second block valve). It may be cased in places for personnel protection, but should be allowed to cool off to the atmosphere so that the temperature limits of the metal are not exceeded when hot catalyst is unloaded. Some units have finned pipe to help in cooling off the unloaded catalyst.


Figure 47 Catalyst Storage Hopper

Auto Gauge Instrument Vacuum Nozzle

Grate with Wire Mesh

Outlet

Manual Gauge Hatch

Relief Valve Inlet

Manway Catalyst Makeup (Fresh Catalyst Only)


CATALYST STANDPIPES AND EXPANSION JOINTS Catalyst flows through refractory-lined standpipes from the regenerator to the base of the riser, from reactor to the regenerator, and from the upper regenerator to the combustor. Expansion joints (Figure 48) are placed in these large catalyst-transfer lines to absorb thermal movement within the system and thereby minimize stresses imposed on vessel nozzles. The expansion joint is lined to protect the bellows from catalyst erosion. Pantographic linkages are used on the expansion joints to assure equal movement of both expansion bellows during temperature changes. Parameters affecting expansion joint design are movement to be absorbed, process environment, mechanical construction, and specified cycle life. During the last 10 years, advances have been made in the engineering and metallurgy of expansion joints. For a hot wall regenerator standpipe using 304H SS, the expansion joint had Inconel 625 bellows that were annealed after forming [for combined resistance to polythionic attack (PTA) and stress corrosion cracking (SCC), Inconel 625 is one of the recommended alloys]. The maximum allowable bellows temperature for this design was 1200°F. Today we use cold wall design for the standpipe and expansion joint. The expansion joint bellows use Inconel 625 LCF (low cycle fatigue) with a maximum allowable temperature of 1000°F and it is not annealed after forming. Annealing reduces the strength of the bellows as well as the fatigue properties. It is important that the bellows material not get too hot so age hardening does not occur. It is also important to stay above the dew point of the process so the bellows stay dry and corrosion is avoided. For example, the regenerator expansion joint bellows temperature would be close to the dew point so UOP specifies external insulation of 25 mm which is enough to raise the bellows metal temperature to 600800°F and to stabilize it from external changes caused by the weather.


Figure 48 Expansion Joint Location

Top of Support

Spring Hangers

Spent Catalyst Expansion Joint Recirculation Catalyst Expansion Joint Regenerated Catalyst Expansion Joint

Top of Support

Structural Bumper


The typical components of a UOP specified expansion joint are shown in Figure 49.

Figure 49 Expansion Joint Components

Convolution

Liner

Root Ring

Cover

Purge Connection

Bellows Flow

Bellows Design – UOP presently specifies single ply bellows. Dual Ply bellows are acceptable provided each ply is designed for full temperatures and pressure. Dual Ply bellows should include a method for pressure leak detection (preferably a positive pressure gage) as shown in Figure 50. When adding a dual ply replacement to an existing expansion joint, the piping system should be reviewed for the additional loads induced by the higher spring rates of the dual ply.


Figure 50 Two-Ply Bellows WEEP HOLE

TWO-PLY TESTABLE

FOR LEAK

BELLOWS

DETECTION

SEAL BAND TO ISOLATE BELLOWS ATTACHMENT WELD

PROCESS PIPE

ATTACHMENT WELD

Control Rod Design – The control rods are intended for control and stability of the expansion joint and are not intended to take the full bellows pressure thrust force. The control rods should be used for reference when determining the installed position of the expansion joint as well as a reference during operation and shutdown (for future trouble shooting information). Equalizing Rings – UOP presently specifies a self equalizing expansion joint. Root rings are an acceptable alternative to equalizing rings. (UOP specifies a minimum area requirement for root rings). Equalizing rings and root rings are intended for added protection in the event of over pressure. Bellows Packing Details – Several methods of packing the bellows are shown in Figure 51. External insulation is required on packed bellows to minimize the potential of condensation during operation. UOP currently uses packed bellows for all expansion joints except for the spent catalyst expansion joint which is continuously steam purged to remove hydrocarbon and minimize coking. Several refiners have successfully used packed bellows for the spent catalyst expansion joint. UOP is currently evaluating several FCC units before packed bellows for this service is incorporated in the project specifications.


Figure 49 illustrates a typical expansion joint with an aeration connection. The purge prevents an accumulation of catalyst between the bellows and inner wall which would restrict the movement of the expansion joint. The purge should be 5-10 psi (0.35-0.70 kg/cm2) above standpipe pressure; the purge is normally controlled with a restriction orifice.

Figure 51 Packing Details of Expansion Joints

The expansion joint of the spent catalyst standpipe is of hot-wall construction to match the spent-catalyst stripper. Based on the location of vessel supports, this


expansion joint absorbs both axial compression and lateral offset. These forces are due to the regenerator expanding upward, the spent catalyst stripper downward, and the standpipe compressing during operation. This standpipe is relatively short and therefore not subjected to out-of-plane movement that occurs on inlet and outlet nozzles connected to large-diameter vessels. Early units used carbon steel standpipes and slide valves for the spent catalyst to the regenerator. Most of the newer units use 1¼ Cr, ½ Mo; although a few have stayed with insulation lined carbon steel. On new units, the regenerated catalyst standpipe and the lower portion of the reactor riser are of cold-wall construction. On this type of unit, the expansion joint needs to absorb axial extension and lateral offset, mainly as a result of the large thermal growth from the hot-wall stripper and upper reactor riser. On older units having a hot-wall wye section and regenerated catalyst standpipe, the primary movement during operation is still axial extension and lateral offset. However, on shutdown of the unit, the expansion joint may need to absorb axial compression, which occurs when the raw oil is cut from the unit, steam is added at the bottom of the riser, and the regenerated catalyst slide valve is closed. The steam cools the reactor riser while the regenerator standpipe remains at operating temperature because of hot catalyst filling the standpipe above the closed slide valve. The regenerated catalyst standpipe can also be subjected to additional movements because of the long length of reactor riser and standpipe. Any bowing of the riser or the stripper or sagging of the standpipe affects the amount of axial movement and lateral offset the expansion joint has to absorb. The regenerated catalyst standpipes and slide valves range from 5% Cr on conventional units to stainless steel (Type 304) in older higher temperature units and cold wall design in modern high temperature units. Thermal movement is absorbed within the expansion joints by means of a flexible item referred to as bellows. The bellows are formed into a corrugated shape from thin-gauge material of a metallurgy selected to perform in the process environment encountered; the expansion joint bellows are normally Alloy 625. The geometry and number of corrugations used relate to the total movement capacity of a bellows. The bellows design is based on equations outlined in the Standards of the


Expansion Joint Manufacturers Association (EJMA). These equations were developed from engineering efforts with expansion joint manufacturers, extensive testing programs, and operating history. See Figure 52. The expansion joint must also conform to the ASME Code for Pressure Piping, B31.3. The expansion joint bellows absorb movement by means of axial compression or extension, angular rotation, and lateral deflection. See Figure 53. The expansion joints used in the FCC unit standpipe are subjected to both axial movement and lateral deflection. For this reason, two separate sets of bellows, called a dual element expansion joint, are incorporated in the design. The lateral deflection occurs in the vertical plane, parallel to the expansion joint pantographic linkage, and is absorbed by means of angular rotation of the bellows.


Figure 52 Bellows Movement 

D C



h E

L TAN  =  = 2

F

▀

+

h L ▀ 2

If Ends are Parallel:  = ▀ = 

A B



Ratios: C A F = = D B E   C-A Tan ( 2 ) = ( 2 )( F ) or

( 2 )( D E- B )


Figure 53 How Motion is Absorbed by Bellows


SLIDE VALVES Slide valves are used to control catalyst flow. These are gate-type valves driven by a hydraulic actuator. The primary problems encountered with slide valves were associated with erosion and corrosion. The elimination of guide steam purges have alleviated many of the corrosion problems. However, if guide steam purges are present, then the steam purge can be used once or twice a shift to keep the guides free of catalyst. In some cases, a continuous purge is used, but its flow should be restricted to prevent erosion of the guides. To prevent erosion, the valve disc is covered with an abrasion-resistant refractory anchored by stainless steel hexmesh. Hard metal surfacing is used on other parts of the valve exposed to catalyst flow (see Figures 54 and 55). The clearances between the support guides on the sides of the valve and the disc are set by the manufacturer. The entire system will expand when it gets hot, so these clearances should be checked to avoid binding or sticking. The new slide valves are designed such that no guide purge is required. Cold wall design which uses carbon steel (see Figure 54) has also eliminated some corrosion problems as well as many of the cracking problems generally associated with hot wall design (specifically polythionic attack on the stainless steel). By working closely with many slide valve suppliers, UOP has also incorporated design and geometry guidelines which minimize erosion during normal operation. These changes include locating the orifice plate upstream of the valve and sloping the bottom of the bonnet a minimum of 30° from the horizontal. The sloped bottom prevents catalyst accumulation in the bonnet. The support guides for the disc are recessed a minimum of 3" (75mm) from the sides of the inlet port. Other design improvements recently developed are as follows: 1. Five second stroke time for normal operation. 2. Two second stroke time for emergency shutdown. 3. Emergency shutdown features are testable on stream. 4. Offset port to center flow during normal operation. 5. Development of cold wall valves for spent catalyst standpipe service. 6. In shop hot stroke test required to guarantee trouble free operation in the field.


Figure 54

Cold Wall Slide Valve Flow

CL Port CL Port

CL Valve

CL Valve

Port Opening

Stem Stuffing Box Guide Bolting Bonnet

Drain Slots Guide

Bonnet Carbon Steel Shell

Disc

Orifice Plate


Figure 55 Hot Wall Slide Valve

HYDRAULIC OIL SYSTEM A hydraulic oil cylinder is used to drive the slide valve because of its size and the required fast response. A low capacity, high head pump supplies the oil through the controlling pilot valve. A manual control valve, the joystick, is also provided in case the pilot valve plugs. Most slide valves are equipped with a handwheel for use during total hydraulic oil failure. When the bypass valve between the two ends of the hydraulic oil cylinder is opened to equalize pressures, the handwheel is engaged. It is somewhat risky to attempt control on handwheel during normal operation, because the handwheel is usually very slow. A typical response time for a slide valve on a conventional hydraulic oil operation would be fully open to fully closed in 30 seconds maximum. Many older systems unfortunately are slower than this. Startup should never be attempted with handwheel control as valve closings will be slower than required. The handwheel should never be engaged until the hydraulic cylinder is bypassed. The equipment may be damaged if the hydraulic cylinder exerts force against the hand wheel.


A typical FCC slide valve hydraulic oil system, used in earlier designs, is shown in Figure 56. The hydraulic oil is pumped out of the low pressure tank through a filter to the slide valve. It circulates back from the valve or from two flow by-passes. One of these is a hand controlled valve at the end of the hydraulic header; the other is a minimum flow bypass on automatic control. These keep the oil circulating and prevent pump damage that might occur if the pump were to run against a closed system, such as when the pilot valve is holding the valve in one position. A spare pump with auto start is used to maintain flow if the first pump fails. The high pressure oil tank will supply sufficient hydraulic oil to close the slide valves if both pumps go down. This tank is not large enough to hold oil pressure for more than one to two minutes. If both pumps are lost, the unit should be brought down until at least one is running again. Modern plants are designed with a separate oil system for each valve, but it is more common to have one hydraulic oil system for all slide valves. Small multi-stage centrifugal pumps are gradually replacing the reciprocating pumps on older plants.


Figure 56 FCC Slide Valve Hydraulic Oil System N2

From Control Instrument

IA Must Drain F

FV RO

SV

HC FH

PI D HPR

HV

LPR F

FIC F

M

P

P

T

F&W

Legend D = Drain F = Filter FH = Flexible Hose FIC = Flow Indicating Controller F & W = Filling & Withdrawal HC = Hydraulic Cylinder HPR = High Pressure Receiver HV = Hand Valve IA = Instrument Air LPR = Low Pressure Receiver PV = Piolet Valve RO = Restriction Orifice SV = Safety Valve UOP 2119-44


UOP currently specifies electrohydraulic actuator assemblies with an individual hydraulic power source for each slide valve (see Figure 57). The individual hydraulic power source has several advantages in that the valves act independently and eliminate all interconnecting piping. A continuous running variable displacement pump maintains the system's hydraulic pressure. When no valve movement is required, the pump operates near its no-load current level and reduces heat generated in the reservoir. Hydraulic oil flows from the pump through a set of filters into the accumulators and actuator. The actuator is electronically controlled and directs hydraulic oil pressure to either end of the cylinder, positioning the slide valve in response to the process demand. An accumulator makes the fast response of the valve possible. When necessary, this accumulator will drive the valve two complete cylinder strokes at an approximate speed of 5 seconds per stroke. If the accumulator becomes depleted, the pump motor itself can supply power to move the slide valve 10% of its stroke every 10 seconds. There is an emergency accumulator, which normally is isolated from the hydraulic system and can be used only if directed from the control room. This accumulator will also generate two actuator cylinder strokes. If the pump and spare motors are lost and the emergency accumulator is not used, there are two means of manually positioning the actuator. One mode of manual operation uses a hand pump to move the cylinder. During normal operations, the hand pump is isolated from the system. The other mode uses a mechanical jack or manual handwheel to move the cylinder. When the handwheel is put into operation, the hydraulic cylinder is automatically bypassed to prevent damage should hydraulic power be restored. The hydraulic pilots are very sensitive mechanisms and it is imperative that the oil supplied to them is absolutely clean. It is important that the hydraulic oil used has the proper viscosity and meets all the specifications given by the valve manufacturer.


Figure 57 Slide Valve Actuator Component Arrangement and Hydraulic Circuitry

REACTOR RISER DISENGAGER DESIGNS The heart of any catalytic process is the reaction site. In the case of the modern FCC unit, this is the riser. With the introduction of riser cracking, the reaction site has changed from the reactor vessel to the riser. Today's reactor could be more properly called a disengaging vessel. A historical progression of riser termination devices (from the 70's to early 90's) is illustrated in Figure 58. These include: the traditional T-type disengager, downturned arms, vented riser, and direct-connected cyclones or suspended catalyst separation.


Downturned arms had essentially replaced T-type disengagers in revamp situations where the scope did not extend to a change in reactor cyclone design. The downturned-arm separation device began to replace the traditional T-type disengager in 1984. Currently, UOP has nine of these designs in operation. The downturned-arm design offers an increase in separation efficiency compared to the traditional T-type disengager. The T-type disengager has a separation efficiency in the range of 70%. Modeling work performed by UOP indicates that the separation efficiency of the downturned-arm separation device is about 80%. As a result of the improvement in separation efficiency, a reduction in catalyst loss from the reactor is observed if the existing cyclones are heavily loaded with catalyst. The reduction in abrupt changes of catalyst direction has reduced erosion compared to the T-type disengager.

Figure 58 Reactor Riser Disengaging Devices


Vented-riser systems, which use the momentum of the catalyst exiting the riser for ballistic separation, have been successfully operating for more than 10 years and are well proven in commercial applications. UOP has 11 vented-riser designs in operation. On later vented riser designs, no mechanical or erosion problems have been observed. However, limited erosion has been noticed on units revamped to a vented-riser when the catalyst flux is high. As a preventative measure, UOP now recommends additional ¾" (19 mm) abrasion-resistant refractory lining to be installed in the following areas on vented-riser reactor systems (Figure 59): 

The reactor head extending approximately 3 ft beyond the tangent line. The refractory lining is used to cover a target area on the head of the reactor where the catalyst impacts directly. The lining is a precaution since erosion has not been observed in this area.



DA points. Half-pipe shields coated with refractory lining are installed over the DA points to protect them against erosion.



Manways. Refractory lining is installed in the lower half of the manways in active areas. The most important site for installation of the refractory is on the large manway.



TI points. Half-pipe shields coated with refractory lining are installed over the TI points to protect them against erosion. The thermowells are to be the hard surfaced type. The control point is moved to the reactor plenum chamber.



The reactor shell above the cone section. A stainless steel taper bar should be used to terminate the abrasion resistant lining being added. The angle of this taper bar should be as small as possible. The abrasion resistant lining is usually extended a minimum distance of 3' (0.9 M) above the reactor cone.


Figure 59 Extent of Reactor Cladding and Hex Lining For Vented Riser Terminations

Direct-connected cyclones are now a well-proven technology. They represented a continuing evolution toward minimizing counterproductive post-riser residence time and maximizing separation efficiency.


In this design, the combined catalyst and product vapor mixture exits the riser and passes directly into the riser cyclones for separation. The product vapors exit the riser cyclone gas tube and pass through a second stage of cyclones before exiting the reactor. The catalyst flows down the cyclone diplegs into the catalyst stripper. The vapors carried down the diplegs and the vapors from the catalyst stripper are swept from the reactor system back into the cyclone system with steam into a vent tube entering the cross over duct between the cyclone stages. Because all of the catalyst flows into the cyclones, erosion was concern in the design of a direct-connected cyclone system. Reports from the operating UOP direct-connected cyclone systems that have undergone scheduled mechanical turnarounds indicated only negligible erosion in the riser and cyclone areas. Because of the transfer piping now incorporated between the primary and secondary cyclones, the hydrocarbon vapor product and hot catalyst are no longer in contact with the reactor shell. Consequently, a large thermal differential is introduced between the reactor internals and the adjacent reactor shell at startup, at shutdown, and during emergency situations. The mechanical design of the reactor must accommodate this thermal gradient. UOP's standard design practice is to use a thermal differential, which is based on the difference between the reactor design temperature and the steam condensation temperature. The thermal gradient is not limited to the reactor internals. The reactor riser, reactor vapor manifold, and reactor vapor line are also subjected to the same thermal differential. As a result, all guides, supports, and other attachments to the reactor shell must be designed to accommodate this condition. UOP's experience has been that all support guides, and sometimes the actual vapor line, require some type of modification when the direct-connected cyclone system is incorporated into the design. In summary, the addition of direct-connected cyclones affects the entire reactor system and is not just limited to the internals. During the 90's, UOP developed and put into commercial operation the suspended catalyst separation system which combines the features of the vented riser (allowing pressure upsets) and direct connect (high containment) systems (see Figure 60). This system is a two cyclone vented riser which operates with a similar flow pattern to the direct connect system, but the riser above the first stage is open.


The riser fills with a dense zone of catalyst above the inlet to the first stage cyclone during normal operation, but during a pressure upset the catalyst is free to discharge harmlessly into the reactor. It was also proven that the suspended catalyst separation system had the same advantage as the direct connect system; post riser cracking is minimized because all of the vapor goes to the first stage cyclone.

Figure 60 Suspended Solids Separation Riser Design


The most recent commercially proven riser termination technology developed by UOP is the Vortex Separation Technology. This includes the Vortex Separation system (VSS) for internal risers and the Vortex Disengager Stripper (VDS) for external risers. (see Figures 60 and 61). The operating principle of the VSS riser termination system is quite simple. Catalyst is discharged centrifugally in the horizontal plane, swirls downward along the wall of the chamber, and contacts the prestripping vapors before entering the stripper vessel. The vapor outlet from the VSS chamber is directly connected to a single stage of conventional cyclones. The base of the VSS is submerged in a fluidized dense phase of catalyst in the reactor cone. The catalyst discharging from the cyclone diplegs easily communicates with the catalyst level within the VSS chamber to maintain a controlled catalyst level. Tray and baffle arrangements at the bottom of the VSS chamber and top of the stripper direct stripped hydrocarbon vapors and stripping steam rising from the stripper up through the VSS chamber for use as prestripping media. Any vapors entrained down the cyclone diplegs, along with fluidization steam, and purge steam flow out of the reactor vessel on pressure balance through vent nozzles in the VSS outlet duct. The top of the vessel, the lower section above the stripper and the reactor riser above the stripper are lined with abrasion resistant lining for erosion protection. Material of construction is normally 1¼ Cr ½ Mo. The vessel normally operates at 510-530°C (950-990°F). External insulation is required not only for personnel protection but to prevent excess coke formation. Formation of coke in the reactor prevents the use of air for heat up during normal startup procedures.


Figure 60 Vortex Separation System (VSS)

Vortex Chamber A Prestripping Section Prestripping Steam

A

Dipleg

Gas Catalyst

Catalyst Level Internal Riser Spent Catalyst Stripper

Vortex Chamber Section A-A


Figure 61 Vortex Disengager Stripper (VDS) To Main Column Plenum

Cyclones

Vortex Disengager Riser Counter Weighted Flapper Valves Support Brackets

Gas Flow To Cyclones

Fluffing Steam Ring

Reactor Shell

Prestripping Section Catalyst Return Slots

Spent Catalyst Stripper Catalyst Flow to Stripper

Stripping Steam Ring Fluffing Steam Ring

Spent Catalyst To Regenerator


REACTOR RISER The reactor riser is a vertical pipe in which the desirable cracking reactions take place. Hot catalyst enters the “wye” section at the bottom of the riser and is lifted up to the feed injection point by the lift gas/steam mixture. Two wye designs, a hot wall and a cold wall, have been used. The two wye sections differ in wall temperature and resulting stresses. The modern cold wall wye with 5" (125 mm) of refractory lining has a lower wall temperature and reduced stress, permitting the metallurgy to be changed from 304H stainless steel to carbon steel. The cold wall wye and the hot wall wye are illustrated in Figures 62 and 63. A mixture of lift gas and steam is injected through the lift nozzle at the bottom of the wye section. Catalyst and lift media travel up the riser to the feed distributors where the riser diameter increases. This increase allows for the increased volume of hydrocarbon vapors as the oil is injected and vaporized when it meets the catalyst. Because the riser volume is small it limits the contact time between the catalyst and hydrocarbon. This prevents overcracking of the products. Below the point where the riser enters the reactor stripper, the riser is carbon steel with castable refractory lining. This lining is abrasion resistant and insulates the carbon steel from the high catalyst temperatures. Above the point at which the riser enters the reactor stripper, the riser becomes hot wall so the metallurgy is upgraded to typically 1¼ Cr ½ Mo.


Figure 62 Cold Wall Wye

5” High Density Refractory

Carbon Steel

Reinforcing Rings

Bumper

Lift Gas Distributor


Figure 63 Hot Wall Wye


REACTOR RISER FEED DISTRIBUTION The earliest feed distributors were simple open pipe bayonets located in the base of the wye section. Figure 64 shows an example of this type. Efforts were then made to generate some feed dispersion by using multiple nozzles at the exit point of the distributor. This led to the design of the "showerhead" distributor shown in Figure 65. This change generated a substantial improvement in overall process performance. Figure 66 shows the effect on cross-sectional temperature profiles in the riser when changing from a single nozzle bayonet to the multi-nozzle showerhead design. In the late 1970's and throughout the 1980's, much emphasis was placed on atomizing the oil into very fine droplets and evenly dispersing these droplets into the flowing catalyst. UOP's efforts led to the development of the WYE premix distributor shown in Figure 67. Steam is injected into the feed upstream of the distributor. The combined stream is then "pre-mixed" inside the distributor to obtain a pseudo emulsion phase before it exits the nozzles at the distributor tip. The expanding steam at exit conditions helps to break up the oil into fine droplets to achieve rapid vaporization and even mixing with the catalyst. Further refinements to the wye feed distributor were made by utilizing an annular lift around the base of the distributor. Either steam or gas can be used as the lift media. The intent is to provide some pre-acceleration of catalyst around the distributor before the catalyst is contacted with feed. This serves to reduce the degree of backmixing in the wye and lower riser. Figure 68 shows a wye premix distributor with annular lift.


Figure 64 Reactor Riser Feed Distributor Bayonet Type


Figure 65 Reactor Riser Feed Distributor Jet Nozzle Type


Figure 66 Effect of Feed Distributor

UOP 3110-9


Figure 67 Premix Feed Distributor (Wye)


Figure 68 Premix Feed Distributor with Annular Lift (Wye)


ELEVATED FEED SYSTEM During the 1990's, the industry widely accepted that elevated radial injection system on the riser is the preferred feed injection location. A typical elevated feed injection system is shown in Figure 69. The wye section, acceleration zone, and mix zone are the main elements. The term system must be emphasized because the best result will not be realized without all the parts functioning in the proper manner. The UOP system has evolved over many years from a combination of strong process understanding, commercial testing, cold flow modeling, and mathematical modeling. Acceleration Zone The base of the wye section is quite turbulent because the huge mass of catalyst changes direction here. Significant backmixing occurs as the catalyst begins to move up the riser. Clearly, injecting the feed at this location cannot be desirable. Restoring an even flow of catalyst is important before injecting the feed. Restoring an even flow is the function of the acceleration zone. A carrying gas, either steam, dry gas, or a combination of both, is injected at the base of the wye. Proper acceleration of the catalyst results in a more even catalyst flow distribution and a lower slip factor. The slip factor is the ratio of the gas-phase velocity to the catalyst-particle velocity. The catalyst is always moving at a lower velocity and “slips” relative to the gas phase. At higher gas-phase velocity, the slip factor decreases. A riser flow with a high slip factor is a less uniform and more unstable system. The goal is to minimize backmixing when the feed is injected. Uneven catalyst flow and high slip in the riser can lead to localized areas of high temperature and feed overcracking, which result in greater amounts of coke and dry gas and reduce selectivity to desired products. In addition, lift gas has proven to passivate metals which further reduces the dry gas yields.


Figure 69 Elevated Feed Injection System

Mix Zone

Feed & Steam Acceleration Zone

Catalyst

Steam or Gas UOP 2569B-1


Catalyst Velocity and Density Computational fluid-solid dynamic modeling work has indicated that the catalyst flow in the riser can be subject to undesirable slugging phenomena at low velocities. Increasing the catalyst velocity helps to reduce such slugging and improves the initial contact with feed. All evidence, both commercial testing and theoretical principles, confirms that an accelerated, moderate-density catalyst phase is the ideal environment for feed injection in a flowing riser. Mix Zone The last part of the feed distribution system is the mix zone, the point of contact between the flowing catalyst and injected feed. At this location, the feed is vaporized rapidly, and cracking reactions begin. This zone is a complex three-phase system: flowing solids, liquid feed spray, and vaporized gas phase. Rapid and intimate mixing at this point is critical to achieving the best yield performance. The number of feed nozzles, location and angle of injection, spray pattern and riser coverage, and droplet size and velocity are all important parameters that are optimized to generate the best performance. A final objective of the mix zone is to achieve a plug-flow environment as quickly as possible once the catalyst and feed have been mixed. An even distribution of catalyst and hydrocarbon vapor helps promote and complete the desired reactions as the mixture flows up the riser. Backmixing of catalyst at the point of contact must be minimized. Riser gamma scans at this location have helped identify and optimize the desired flow patterns. The first generation of elevated feed system used the elevated premix feed distributor shown in Figure 70.


Figure 70 Elevated Riser Premix Feed Distributor


Optimix Feed Nozzle After the elevated premix feed distributors were proven successful, UOP instituted a program to further improve the atomization and dispersion characteristics of its feed nozzles. The result of these efforts is the new Optimix feed nozzle (see Figure 71). The important features of the Optimix nozzle are: • • • • • • • •

Small average droplet size Narrow droplet size distribution Flat fan spray pattern Even liquid flux across spray Moderate pressure drop Three stages of atomization Short residence time after internal droplet generation High turndown efficiency

The Optmix nozzles have been commercially proven in more than twenty units and major performance benefits with the Optimix nozzle system have been achieved in dry gas reduction and gasoline yield improvement. The feature most recently added to the Optimix distributor is the DUR O LOK coupling. This coupling provides a means for replacing the tips for maximum flexibility in both maintenance and modification for changes in design charge rate.


Figure 71 UOP Optimix Feed Nozzle Steam DUR O LOCTM Coupling

Oil

REACTOR PLENUM The direct-connected cyclone reactor systems prompted the development of new reactor plenum designs. In a direct-connected system, the area surrounding the cyclones is relatively inactive. The traffic in this area is limited primarily to the vapors being carried out of the reactor stripper. During startups and shutdowns, the temperatures observed at the reactor shell and cyclones can therefore be substantially different. To accommodate the large differences in thermal expansion, new plenum designs were required. Different approaches to this problem are shown in Figure 72. UOP typically uses an internal plenum design to minimize overall vessel height and cost.


Figure 72 UOP Plenum Chamber Designs

External Plenum

Internal Plenum

External Manifold

SPENT CATALYST STRIPPER The spent catalyst stripper (Figure 73) surrounds the upper portion of the reactor riser. Catalyst descending from the reactor passes into the stripper where it flows over stripper baffles. The steam to the stripping section is distributed through a number of small holes in two opposing semi-circular steam rings. A rate of 1.5-2 lb/1000 lb catalyst is typical. The stripping steam displaces oil vapors from the voids between the catalyst particles and in the catalyst pores and returns this vapor to the reactor. The catalyst flows out the spent catalyst standpipe with some entrained steam. A small quantity of steam is injected into the base of the stripper cone through an additional semi-circular steam ring which maintains catalyst fluidization and assures an even temperature distribution at this section. Improved stripping of hydrocarbon from the spent catalyst provides significant advantages: catalytic coke from higher conversion replaces entrained hydrocarbon due to the increased catalyst-to-oil ratio resulting from lower regenerator dense bed


temperatures. Alternatively, a lower coke yield for the same conversion can be used to increase throughput on main air blower limited units. In 1998 and 1999 UOP conducted an extensive research program to evaluate the existing commercial stripper tray designs and develop a more efficient design. A commercial scale Plexiglas model was built to allow visual observations of the catalyst and gas flow patterns. Catalyst circulation rates greater than commercial flux rates were possible in this model. Helium tracer gas injected to the catalyst entering the top of the model was used to quantify the stripping efficiency for each tray type. The three stripper tray designs which were used commercially in the past by UOP and tested in the model are shown in Figure 74. The classic UOP tray, which had 2 rows or large holes on the top of the tray was used up to the mid 1980’s. The stripper tray style with three rows of holes in the skirt was used from the mid 1908’s to early 1990’s. In the early 1990’s the jets were returned to the top of the trays with longer skirts which allowed more pressure drop across the tray for better steam distribution. Each of these designs perform well at low to moderate catalyst flux rates, up to ~60,000 lb/hr/ft2 (290 kg/hr/m2). Above these rates the efficiency falls quickly and fluidization problems can occur which limit the capacity of the stripper. The modern tray design developed from the cold flow modeling is shown in Figure 75. This tray uses a larger number of smaller jets on the top of the tray spread out over a larger percentage of the tray than in previous designs. This provides for better steam/catalyst contacting resulting in improved stripping efficiency. Another advantage resulting from this is that all areas of the bed are uniformly fluidized resulting in smoother catalyst flows and uniform catalyst flux. This improved fluidization has allowed the upper limits on catalyst flux to be pushed above 100,000 lb/hr/ft2 ( 490,000 kg/hr/m2) while maintaining very high stripping efficiency. The catalyst flux is based on the annular area between the riser and stripper shell.


The modern stripper is also designed for higher catalyst residence times than the early models. This residence time is important to allow completion of bed cracking reactions of the absorbed hydrocarbons which continue to generate strippable vapors that can be removed before the catalyst enters the regenerator. The uniform catalyst flux in the modern stripper also increases the effective catalyst residence time by eliminating areas of stagnant catalyst and areas of high catalyst flux.

Figure 73 High Efficiency Spent Catalyst Stripper

Insulation

Riser Stripper Shell

Stripping Steam

Stripping Steam

Abrasion Resistant Lining Fluffing Steam

FCC-E005


Figure 74 Historical Stripper Tray Designs

Classic UOP Tray Up to Mid 1980's Short skirt Holes on top near the edge

Mid 1980's to Early 1990's Longer skirt Holes on vertical skirt

Modified Classic UOP Tray Early 1990's to 1999 Long Skirt Jets on top near the edge FCC-E006


Figure 75 Modern Stripper Tray Design C L Spent Catalyst Stripper Abrasion Resistant Lining

Detail

12-14 Gage Tubing

Riser

Stripper Shell FCC-E007


REFRACTORY LlNING Refractory lining is used extensively in the reactor and regenerator. The types of refractories generally fall into three categories: abrasion resistant, insulating, and castable. Castable refractory lining exhibits both abrasion resistant and insulating properties. In recent years, abrasion resistant plastic linings have entered the market. Plastics have become popular for small repair jobs because of their ease of application. The material is putty-like and comes ready for installation without addition of water. A listing of the various applications is given in Table 2. TABLE 2

REFRACTORY LlNING

Application Reactor Reactor Stripper Upper Regenerator Lower Regenerator Standpipes Upper Reactor Riser Lower Reactor Riser Cold Wall Wye Hot Wall Wye Feed Distributor Expansion Joint Slide Valve Cyclones External Mixer Flue Gas Line Catalyst Cooler

Abrasion Resistant X X X X X X X X X X X X X

Castable

Insulating

X X X X X

X X X X


UOP has made several modifications to the refractory specifications to ensure the proper use of materials, installation procedures, and inspection and testing methods. Some of the major revisions to the UOP Standard Specifications include the following: 

The anchor spacing has been reduced from 12” (300) for high density vibrocast refractory and gunned refractory: Anchor Location Top head of Vertical Vessels Plenum Areas Overhead Areas Horizontal Shells Catalyst Transfer Lines Overhead Vapor Lines Vertical Shells Bottom Head of Vertical Vessels Downhand Areas

Anchor Spacing 1.0 x Lining Thickness

1.5 x Lining Thickness

2.0 x Lining Thickness



The straight-leg anchor design has been replaced with a "steerhorn"-style anchor. The steerhorn-style anchor provides more positive anchorage and minimizes slippage.



The coating on the anchor has been reduced to the tip of the leg. Coating on just the top ½" (13mm) of the anchor leg allows a greater bond and allows thermal expansion of the tips without spalling the lining on the surface.



Metal reinforcing fibers are incorporated in the refractory. The use of stainless-steel metal fibers are required to improve the strength of the refractory lining for cold crushing, modulus of rupture, and thermal shock and to minimize cracking. The quantity of fibers is minimized (2 lb/ft3 or 32 kg/M3) to ensure uniform mixing and to reduce handling problems.




The heat drying procedure has been modified. The modified procedure is required to ensure refractory quality by incorporating a controlled, slow heatdrying rate.



The installation temperatures are controlled to ensure that the lining is not freezing or being subjected to incomplete hydration.



Mandatory testing of materials is now required to ensure the quality of the materials and skill of the installers. Materials and installers must pass all testing requirements prior to installation.

Figure 76 Steerhorn Anchors for Gunned and Vibrocast Refractory Lining

Cap

1-1/2” (38 mm)

1”(25 mm)

60º

5/16” (8mm) Diameter Anchor

3/8” (9 mm) Radius

Shell

Weld

1/2” (13mm) Radius


THERMOWELLS Reactor thermowells are protected with a hard metal surfacing. The number will depend upon the size of the unit. The most important temperatures are the riser outlets for a modern unit, or the dense bed for an older unit. These can be checked with the reactor over head and stripper bottom outlet thermowells. The catalyst in the stripper usually is 5-10°F (3-6°C) hotter than the cyclone outlets, because heat transfer from the catalyst to the oil is not perfect. Thermowells may inadvertently be placed in a dead area, where false, usually low, readings are obtained. False readings can be detected by comparing the different readings obtained around the reactor. The thermowells used in the regenerator are protected by a stellite hard surface coating. Wear will vary, depending on location. During the inspection, each thermowell should be carefully checked and replaced if necessary.

DRY STEAM, DRY AIR and DRY GAS SURGE POINTS The instrument connections on the reactor and regenerator must be protected from catalyst damage. The fluidized catalyst will behave like a liquid and "flow" into any instrument opening. Once the catalyst is out of the fluidized area, it tends to settle and pack. The instrument tap is blocked off and a false reading given. To avoid this problem, a stream of dry air, gas or steam is injected into the line between the vessel and the instrument to provide a buffer seal. This system is shown in Figure 76. It is important to point out that if air is used as the purge gas, its flow should be minimal to prevent localized burning or harmful oxidation reactions. The amount of steam injected is controlled by restriction orifices, 1/8" (3mm) for steam and 1/16" (1.6mm) for air or gas. The pressure to the orifice should be 5 psi (0.35kg/cm2) higher than the vessel pressure. For differential instruments, there is no error in measurement because the isolating medium injected is the same for each tap. For single instruments, any error introduced is negligible.


Figure 76 “DA”, “DG” and “DS” Piping 1/2" Pipe to Instrument

Provide 6' (1800) or More Clearance to Enable Use of Reamer

3/4" SCH 160 Piping Minimum Run Straight Piping Between Fittings

1/16" Restriction Orifice Strainer P

Nozzle on Vessle Or Standpipe 3/4 '' Instrument Air for "DA" Points 3/4 '' Purge Gas for "DG" Points 3/4 '' Purge Steam for "DS" Points

FCC-E008


During inspection, the pressure taps (and pressure taps used for level indication) should be cleaned and the surrounding refractory should be checked. The reamer, shown in Figure 77, can be used to clean out the taps when the unit is on stream, or when it is down.

Figure 77 Pressure Tap Reamer 6'' (150) DIAMETER HANDWHEEL

1/4 '' (6) STEEL ROD STEEL GLAND STEEL STUFFING BOX

FLEXIBLE GRAPHITE PACKING

3/4 '' FORGED STEEL SCREWED TEE

3/4 '' SCHEDULE 80 SEAMLESS STEEL PIPE NIPPLE 9'' (230) LONG

5'-0'' (1525)

ROUND HEAD STEEL PLUG

3/4 '' NPT (NATIONAL STANDARD PIPE TAPER THREAD) (NOTE 2)

1/4 '' (6) DRILL (WELD TO ROD)

FCC-E009


MAIN COLUMN The FCC main column is shown in Figure 78. The column may be divided into two sections: the regular fractionator and the lower disc and doughnut trays. The bottom six trays are designed for high vapor and liquid flow rates. The trays are shown schematically in Figure 79. A coke trap, shown in Figure 80, prevents large pieces of coke or other debris from entering the bottom line. The steam ring may be used on some units to steam strip the bottoms material of any lighter hydrocarbons. The column is fitted with three sample lines which are normally called try lines; the bottom try line is also used as a sample line. These are used to check the bottoms level because the level glass and indicator are subject to occasional plugging by fines or coke particles. Both level instruments are protected with an LCO flush which flows into the level instrument from the top. The FCC main column is mounted on a table top instead of the skirt mounting seen on most columns. Table top mounting simplifies maintenance work on the bottoms lines, which are subject to plugging from coke or catalyst fines. A separate suction line to each pump allows operation to continue through one line while the other is cleaned. As an additional note, every valve in the bottoms circuit should be installed with the stem up to keep catalyst fines out of the bonnet. If plant layout does not allow a stem to be straight up, it should be as close to the vertical position as possible. The upper part of the tower is similar to any other fractionator. Sidecut streams must be stripped to remove absorbed light material. Stacked heavy cat naphtha and light cycle oil strippers are shown in Figure 81. Steam is the most common stripping method, although reboiler strippers are sometimes used. Reboiler strippers decrease the amount of sour water that must be treated. The overhead vent from each stripper returns to the main column in a vapor space.


The overhead from the column is condensed by air or water coolers, normally a combination of the two. Water from the gas concentration section washes the ammonia and some other salts from the last bank of condensers. If it were injected before this, much of the water could vaporize, decreasing its effect. The sour water is separated in the overhead receiver (see Figure 82) and sent to a sour water stripper. Gasoline and gas go to the gas concentration section, with some gasoline returned to the tower for reflux. The main column is made of regular or killed carbon steel, with a 1/8 inch (3 mm) thick Type 405 or 410 (12 Cr) cladding from below the naphtha or LCO draw tray to the bottom outlet. Any nozzles or manways in this clad section should also be alloy lined. All the trays and caps in the column are Type 410 stainless. The inspector should look for any evidence of pitting, cracks or bulging in the alloy lining. Metal thickness should be checked and recorded for future reference. A general inspection should be made for loose caps, clips and bolts, and for general metal loss or tray distortion. The upper part of the column is usually not subject to severe corrosion. The bottom of the column should be checked for integrity of the lining and for any signs of coke buildup. Excess amounts of coke may indicate that the bottoms temperature was too high during operation. The coke trap and steam ring should be cleaned and checked for corrosion. The steam ring should have the nozzles pointing upward, to avoid erosion to the bottom head lining.


Figure 78 Main Column Manway and Vent Reflux

Vapor Out

1

3

HCN Liquid Return

4 5

HCN Stripper Vapor Return

6

HCN Reflux Draw HCN Reflux Return

HCN Pumparound and Product Draw 7 18

LCO Pumparound Return

19 20

LCO Stripper Vapor Return LCO Reflux Draw LCO Reflux Return

21

LCO Pumparound and Product Draw 410S or 405 cladding

17

26

HCO Pumparound Return

27 28 29

Feed Surge Drum Equalizing Line HCO Pumparound and Reflux Draw HCO Reflux Return

30 31 32 33 34

MCB Pumparound Return

35 36 37

38

Reactor Vapor Inlet

Try Lines Quench

Steam Out Coke Trap

Bottoms Pumaparound and Product Draw

FCC-E400


Figure 79 Typical Disc and Donut Pans C VESSEL 1-3/8" DIAMETER HOLES ON 5" EQUILATERAL 1-3/8" DIAMETER HOLES ON 2-1/2" EQUILATERAL

TRIANGULAR PITCH

TRIANGULAR PITCH

OMIT HOLES FOR TOP DONUT ONLY

SUPPORT RING

A

C VESSEL

A OPEN AREA 4" HIGH WEIR FOR TOP DONUT ONLY

PLAN OF DONUT C VESSEL DOWNCOMER OF TRAY

TRAY

SUPPORT RING

DONUT TRAY DISC TRAY

SECTION A-A C VESSEL 1-3/8" DIAMETER HOLES ON 2-1/2" EQUILATERAL OPEN AREA

TRIANGULAR PITCH

C VESSEL SYMMETRICAL ABOUT CENTERLINE

*

*

1-3/8" DIAMETER HOLES ON 5" EQUILATERAL TRIANGULAR PITCH

PLAN OF DISC NO HOLES AT FEED POINTS

FCC-E401


Figure 80 Typical Coke Trap 120

40

150

SUPPORTS

OPENINGS

25 R 150

50 (TYPICAL)

OD OUTLET NOZZLE

100

PROVIDE 2 DRAIN OPENINGS, LOCATED 180 DEG APART.

(BOTTOMS OUT)

FCC-E402


Figure 81 Heavy Cat Naphtha Stripper (Top) Light Cycle Oil Stripper (Bottom) Manway and Vent

Vapor to Main Column BAFFLE HCN from Main Column 1

6

Stripping Steam

LC-LG

Vortex Breaker Vent Vapor to Main Column Stripped HCN Product

1 LCO from Main Column

6

Stripping Steam

LC-LG

Vortex Breaker

Stripped LCO Product

FCC-E403


Figure 82 Main Column Overhead Receiver Vent and Ventilation Vapor Outlet Manway PC

Steam Out Inlet Distributor (Detail)

LC & LG

LC & LG Water Water Boot Water Outlet

Slots Must Face Nearest Head 1/2" Drain Hole

Hydrocarbon Outlet Vortex Breakers

FCC-E404


MAIN COLUMN BOTTOMS The main column bottoms pumps are detailed in Figure 83. The minimum distances shown are important to prevent catalyst fines from settling in a line that is not in service and plugging it. The draw stream for heavy oil out of the unit (i.e. to the slurry settler or storage) always comes off the bottom of the main column bottoms header. Any catalyst or coke which has a tendency to settle will be drawn off and removed from the circulating loop. Steam and LCO lines are provided for flushing on shutdown and for cleaning. If the unit is shut down without flushing, catalyst fines may settle, or the heavy oil may set up causing problems with cleaning. The suction and discharge lines of the main column bottoms pumps range from 5 Cr, ½ Mo to 9 Cr, 1 Moly. The exchanger metallurgy ranges from carbon steel to 5 Cr, ½ Mo. Carbon steel does not last long if it is exposed to hot main column bottoms so the slurry service side of the tubes, tube sheets, and heads should be clad with a chrome alloy, such as Type 410. After the first exchanger, most of the piping is carbon steel. All piping and exchangers should be inspected for corrosion and fouling at each turnaround, and the appropriate records kept. The recommended main column bottoms velocities through the tubes are 3.75 to 7.0 ft/sec for straight tubes and 3.75 to 5.75 ft/sec for U-tubes. Deviations outside of these ranges will show up as excessive amounts of settled catalyst fines or erosion to the tubes. The main column bottoms pumps normally have steam turbine drivers. Speed is limited to 2000 rpm to minimize erosion. The shaft is tungsten carbide or stellite coated, with a light cycle oil flush to the throat bushing and wear rings and HCO to the mechanical seal. UOP recommends aluminum foil packing with light cycle oil flush for cooling and lubrication. Graphite or asbestos have occasionally been used in this service. Many refiners have installed mechanical seals with good success.

STEAM

LP

RETURN HEADER

MAIN COLUMN BOTTOMS

TO CIRCULATING

STEAM CONDENSATE

3/4 ''

MP STEAM

T

TI

PUMP WARMUP/COOLDOWN

HEAVY CYCLE OIL FOR


FLUSHING OIL FOR

PDI

P

PG

MIN

TO CIRCULATING MCB HEADER

MIN

COKE STRAINER

OIL

FLUSHING

STEAM

OIL

FLUSHING

MIN

BLANKOFF

STEAM

MP

COKE STRAINER

MIN

MIN

P

PG

PDI

TI

M


HEAVY CYCLE OIL FOR


FLUSHING OIL FOR

3/4 '' Try Lines

MP STEAM

Figure 83 Main Column Bottoms Circulating Pumps

M

MOV

MIN

FCC-E405

OIL

FLUSHING

MOV

M


MIN


Besides the turnaround, the bottoms pumps usually require some repair work during the course of the run. The seal flush system should be checked for cleanliness, and the packing replaced if necessary. The packing is especially vulnerable to the hot, erosive conditions. There may also be wear to the pump case and rotor. If this is severe, a hard surfacing material may be applied to the rotor. The inspector should also check the pump suction strainer for corrosion and holes.

SEAL AND GLAND FLUSH The main column bottoms pump is protected by gland and seal flushes to keep catalyst fines out of the throat and to lubricate and cool the packing. HCO is used as the flushing medium. LCO is not used as a flush to the main column bottoms pumps because suction will be lost when vapors are formed in the pump casing. For earlier units, all main column bottoms and slurry flow instruments and control valves are flushed with LCO to keep out fines that could plug up or erode the equipment. The total amount of LCO required will depend on the number of flushing points. The flush to each instrument will not be exactly the same because of the pressure drops through the system, although this can be partially balanced by the globe valve provided at such point. The primary control is a 1/16 inch (1.6 mm) restriction orifice after the globe valve. These orifices must be kept clean. A rough number for total LCO required is 1000-2000 BPD (6.5-13 M3/hr). Smaller units usually have fewer exchangers and, therefore, require fewer control valves and less oil. The LCO which enters the main column bottoms stream will flash off when it is returned to the column. Excessive flush will cause cooling problems in the bottoms, and should be avoided. Currently, UOP designs slurry flow instruments which do not use flushing as shown in Figure 84.


Figure 84 Slurry Service Orifice Meter Piping Assembly Closed Coupled Above Orifice for ANSI Class 300/600 Screwed Cap (do not seal weld)

¼" Globe

¼" minimum

2" Field Fabricated Instrument Pipe Support

Pipe

Transmitter

Mounting Bracket

P Capsule

L

Single Valve Equalizing Manifold

H

minimum

Flange Adapters

½" Gate (Install on Vertical Tap)

½" Pipe

Flow

Pipe Strap ½" Solid Plug

FCC-E406


SLURRY SETTLER In earlier units when the cyclone efficiency was not as high, the main column bottoms stream contained catalyst fines which had to be removed to meet certain product specifications. Part of the bottoms stream was sent to the carbon steel settler (sometimes called a clarifier) where the fines would settle out and were returned to the reactor with a slurry recycle. Clarified oil came off the top of the vessel. The settler operated liquid full, with a cross-sectional area designed for an upward oil velocity of 30 BPD/ft2 (2.1 M3/hr/M2) or less at normal charge rates. Heavy cycle oil or raw oil diluent could be used to flush the fines back to the reactor.

Figure 85 FCC Slurry Settler Vent

Clarified Oil Outlet Relief Valve

Manway Main Column Bottoms Inlet Nozzle and Wear Plate

Dilute Oil Inlet and Distributor

Flushing Nozzle

Steamout Slurry Outlet UOP 2119-67


SLURRY FILTRATION Filtration of the slurry product to achieve product specifications of 100 wppm total solids or lower is becoming increasingly common. Sales of the slurry product as carbon black feedstock and desire to minimize the increasing cost of tank cleaning are driving this trend. The most common type of filters use pourous, sintered metal filter elements. A typical filter system will have 2 or 3 vessels with a number of filter elements in each. One vessel is typically in filtration mode while another is in backflush mode to remove the filter cake from the elements. When enough catalyst fines have deposited on the filter elements to increase the pressure drop across the filter to a pre set limit the vessel is taken off line for back flushing. Once the filter vessel is off line and drained the vessel is filled with backflush liquid, either HCO or LCO, and allowed to soak to help dissolve any heavy aromatic compounds on the elements. The top of the vessel is then pressured up with either fuel gas or nitrogen to provide the driving force for a high velocity back flush. The back flush material is collected in receiver vessel and pumped back to the reactor riser or to the spent catalyst stripper. A typical arrangement for the filtration equipment is shown in Figure 86.


Figure 86 FCC Slurry Filter BACKFLUSH GAS BACKFLUSH GAS ACCUMULATOR

CLEAN PRODUCT OUT

FILTER VESSEL #1

FILTER VESSEL #2

FILTER VESSEL #3

VENT

SLURRY IN

BACKFLUSH RECEIVER

HCO or LCO SOAK IN BACKFLUSH TO REACTOR

FCC-E407


GAS CONCENTRATION SECTION The gas concentration unit is relatively easy to operate and usually trouble free. It may be forgotten in the planning for the turnaround. This can prove to be a serious mistake because the vessels and lines are sometimes subject to severe corrosion. The amount and type of corrosion will depend on the feedstock, vessel metallurgy, and the methods used to combat this corrosion Corrosion in this unit can be broken down into three classifications: 1.

Hydrogen blistering and/or embrittlement.

2.

Corrosion of steel by hydrogen sulfide, cyanides, and other acids.

3.

Ammonia attack on Admiralty tubes.

The first of these, hydrogen blistering, is the most common problem with general metal attack by the various acids a close second. Hydrogen blistering occurs in steel where surface corrosion is active. Blisters are formed by the diffusion of atomic hydrogen into steel and then accumulate at slag or inclusions within dirty steel. The atomic hydrogen then converts into molecular hydrogen, and the resultant pressure forms the blister. Blisters may rupture to the thin side of the defect. The presence of cyanides (Prussian blue) greatly increases susceptibility of H2 blistering in steel. The inspector should look for a bluish color when the plant is first opened, because this is a good indication of areas where corrosion and/or H2 blistering may be found. Hydrogen sulfide (H2S) also contributes to blistering and corrosion; it is present in most gas concentration units. The inspector should look for H2S pitting, blisters, and general metal loss. Inspection of the equipment should be thorough, and when corrosion is found, the type and extent of the corroded areas should be determined. It is not unusual to find one head or one particular plate in a vessel full of blisters, with the remaining portion of the vessel in good condition. Repairs can be made by replacing the damaged areas and still maintain the equipment in service.


Ammonia attack on Admiralty tubes in the condensers or coolers is not severe if there is sufficient wash water during operation. Stress corrosion of brass tubes may be seen in exchangers where the pH is high. Galvanic corrosion and dezincification of brass tubes may be found where dissimilar metals are used in the condensers or coolers. Water side erosion or corrosion of the tubes may occur if dirty or contaminated water is used. This could be caused by suspended solids, or if there were salts or other chemical species, for which the exchanger was not designed. The type and amount of problems will vary between refineries. It may even change over a period of time for one unit, as feed or cooling water quality changes. Every turnaround a good inspection should be made with precise records on each piece of equipment. Previous records should be available, and corrosion rates determined each time. Changes of materials in bundles or shells, should be considered if exchanger life is too short. Gas plants are usually easy to gas free and open for inspection. Some units that have been coated with an epoxy type paint cannot be subjected to steam for gas purging. These items must use water or inert gas to clean for entry, which makes the job more difficult. Knowledge of protective coatings and locations is a must for the refinery safety and inspection departments. Operations supervisors should have written orders for each piece of equipment which detail the correct procedures to follow in flushing and opening these items. Admiralty tube bundles should be handled in special slings to eliminate overstressing and potential stress cracking of the tubes at the lifting areas. Good supervision of opening and removing of equipment is critical in this unit. The inspector should follow the flow from the low and high pressure receivers to the final condensers. Corrosion found in any item should be followed to the next item in the flow pattern to determine the extent and area the corrosion has covered. Because there is a certain amount of crossover flow, such as wash water from the HPS to the MC overhead, there may be some backtracking. The corrosion products, such as iron sulfide scale, will not necessarily be in the same area as the most severe corrosion. Should hydrogen blistering be found, the blisters should be measured, drilled if necessary with air drills, and the ruptured metal should be measured to determine if the corrosion allowance has been exceeded. If blisters are small, ½" (13mm) or less, they can be covered with a protective lining to eliminate


future growth. Corrosion allowance in the blistered areas should be maintained if possible. General corrosion, such as H2S pitting, may require replacement of all or part of a vessel or line. Scabbing is usually not effective, but can be used if time is short and if the scabbing is done well. If repairs are not made, the item should be programmed for the next turnaround as necessary work. Various metal corrosion test samples can be installed in the corroded area to determine the best material to be used to resist the corrosion. Hydrogen probes, electrical resistance corrosion probes, and corrosion coupons can be used to indicate the rate and severity of corrosion activity during the run. Onstream inspection of piping can also be used to monitor present problems and to find areas that should be checked carefully or shutdown. If corrosion or blistering is found, there is usually a question concerning replacement with a higher quality material. All equipment eventually wears out; it is a matter of comparing cost with possible benefits. Replacement of trays and other tower internals is easier than replacing the vessel shell. If the refiner feels that the equipment is wearing out too quickly, then a higher alloy may be justified. Before the decision is made, a careful investigation should be conducted to determine other ways of resisting corrosion. For the FCC gas concentration section, the best method to resist corrosion is to maintain the wash water injection at a rate of 6.5-7.0 vol.% of the feed. This has greatly limited or eliminated corrosion in most cases. Inhibitors have been used in some cases, but they are usually not necessary. Most of the material used in the gas concentration section is carbon steel. Killed steel is common for towers and the upper trays in the absorbers. The debutanizer reboiler is very often 5% chrome, ½% moly steel. A lower alloy steel, such as 1¼% chrome, ½% moly, may be used in areas that have had corrosion problems. In severe cases, the affected area may be lined with a 12% chrome, such as Type 405 or 410. The compressor rotor is usually a specially tempered 1¼% chrome, ½% moly steel.


WET GAS COMPRESSOR The compressor of an FCC unit separates the relatively low pressure of the main column from the higher pressure gas concentration and recovery section. Positive displacement reciprocating or centrifugal machines can be used; the latter have become more common because of better performance at lower cost. During the turnaround, thickness measurements of piping headers and manifolds should be made each down period and compared to previous readings. Severe corrosion may be found in the discharge manifolds and piping to the receivers. The manufacturer's instructions should be followed for the gas compressor inspection. This machine normally requires maintenance work on the seals or packing.

CENTRIFUGAL MACHINES Centrifugal compressors (Figure 87) are multi-stage machines, contained within the two external stages. Each of the two stages is protected by an anti-surge device which will over-ride normal controls to protect the machine. Gas leakage out of the casing is prevented by suction and discharge end labyrinth seals. These seals are backed by either buffer gas or seal oil, normally oil. The seal oil system runs 5 psi above suction gas pressure, using a seal oil head tank or differential pressure controller to hold this differential pressure even if suction pressure varies somewhat. A balancing drum is used to offset axial thrust, so suction pressure is maintained at both ends of the compressor. The seal oil system must be kept in operation while the casing is under pressure or gas will leak out.


Figure 87 Centrifugal Compressor

FCC/DS-R00-45


RECIPROCATING MACHINES Reciprocating compressors can be driven by gas engines, turbines, or electric motors. They are usually constant speed machines. Suction valve unloaders and clearance pockets are provided on some machines for greater operating flexibility. A reciprocating machine should never be run with discharge valves blocked in. Liquid slugging is also very dangerous, as the liquid is incompressible and will lead to broken piston rods or a cracked head. Suction knockout drums should be checked frequently for any liquid. The reciprocating machine uses packing to prevent gas leakage from the compressor. This simple system should work quite well as long as the packing is tightened correctly. Over tightening will lead to excess wear and then leakage.


FIRST AND INTERSTAGE SUCTION DRUMS These vessels are designed to remove any liquid from the gas streams to the compressor. They are normally carbon or killed steel, with a stainless steel mesh blanket to remove any entrained droplets. The first stage liquid is drained to a blowcase, and pressured back to the overhead receiver. The liquid from the interstage drum is pumped to the HPS. A typical drum is shown in Figure 88. Pitting corrosion has been found in the lower portion of the shell and bottom head. Hydrogen blistering and embrittlement has also been found.

Figure 88 Compressor Suction Drum Ventilation and Vent Nozzles

Gas Out

Mesh Blanket (Demister Pad)

Gas Inlet with Distributor

LC-LG

Manway

Steamout Vortex breaker

Liquid Out

FCC-E700


WASH WATER The overhead stream from the main column contains various contaminants which may cause corrosion, plugging, or fouling. These would include ammonia, sulfides, cyanides, chlorides, and phenols. A wash water stream is used to remove them, because most of them are ionic or polar species, readily soluble in water. Flow rates are usually maintained at 6-7 vol% of the fresh feed rate. The water should be clean, preferably steam condensate, to prevent adding more problems such as salts or dissolved oxygen to the system. The water is injected after the compressor first stage and is pumped out of the interstage suction drum to the HPS. Here it collects in a water boot and is pressured back to the main column overhead condensers. It collects in the overhead receiver water boot and is pumped out to the sour water stripper for disposal. There is always water present in the main column and gas concentration section from stripping steam and other sources. If the wash water is not used to flush out the sulfides, ammonia, cyanides, and other species, the water present can become highly corrosive from absorption of these contaminants. Sulfide levels of greater than 20,000 ppm have been reported in the overhead receiver water. Hydrogen blistering and general corrosive attack may become quite severe, especially if feed sulfur is greater than 1%, or nitrogen greater than 1000 ppm. Also, while the overhead receiver water may be basic, pH >7, most of the ammonia that is responsible for this will drop out in the main column receiver. The water in the gas concentration section may become acidic from H2S and CN-. If there is any oxygen present, elemental sulfur may be formed from oxidation of the sulfides. This will cause problems in meeting gasoline product specifications. Wash water will solve many of these problems by diluting the corrosives, and keeping the water pH at 8-9, where sulfide oxidation is greatly reduced.


HIGH PRESSURE SEPARATOR The HPS Figure 89 receives the liquid streams from the interstage suction drum and primary absorber bottoms, and the gas streams from the second stage compressor discharge and the stripper overhead. The vessel is carbon or killed steel; it looks very similar to the main column overhead receiver shown in Figure 82, with a slotted pipe distributor instead of a baffled inlet. Sour water is pressured from the boot to the main column overhead, while gasoline is pumped to the stripper. The gas goes to the primary absorber. Pitting corrosion has been found in the lower section. Hydrogen blistering and embrittlement has also been found.

Figure 89 High Pressure Separator Inlet Pressure Relief

PI & PC

LI & LC

Ventilation and Vent

Vapor Out

Slotted Pipe Distributor Manway

Steam Out

Vortex Breaker

LI & LC Liquid Out Water Boot Water Out

FCC-E701


PRIMARY ABSORBER The primary absorber (Figure 90)is a killed steel column with about 30-40 trays. The trays are normally carbon steel with Type 410 stainless steel fittings. Valuable light products such as LPG are absorbed from the gas as it travels from the bottom to the top outlet. The most common tower design uses two intercoolers to remove the heat of gas absorption from the gasoline as it fails down the tower. This temperature rise is roughly 20-25°F (11-14°C). These intercoolers are located roughly one-third and two-thirds of the way down the tower. The gasoline circulation through the water coolers may be pumped or gravity driven. Water drain pots are required on the gravity circulation systems to prevent accumulation of water as the gasoline cools; a water head would stop gasoline circulation. Pumped systems are better from a velocity, and thus heat transfer, viewpoint. Another design that has become popular recently uses debutanized FCC gasoline as a final wash before the gas leaves the tower. A cooled slipstream from the bottom of the debutanizer is introduced at the top of the column, with the gasoline from the main column overhead receiver entering the tower 5 to 7 trays lower. This system allows the use of only one intercooler with the same or better C3 recovery efficiency, i.e., the amount of C3 recovered in the debutanizer overhead compared to the total amount produced. The gasoline from the bottom of the primary absorber is pumped to the HPS; the gas flows to the sponge absorber. Serious corrosion can occur in this column from H2 attack. Hydrogen blistering can be severe in the upper section and top head, and is also seen occasionally in the middle of the tower and in the intercoolers. The shell and heads should be inspected at each turnaround. The inspection of the internals should pay particular attention to stagnant areas around weirs, downcomers, and nozzle extensions. Bolts and trays should be inspected for tightness and embrittlement of materials.


Figure 90 Primary Absorber Vent and Ventilation Vapor Outlet Distributor

Recycle Gasoline Inlet 1

Distributor Unstabilized Gasoline Inlet

8

Manway

9

12

Upper Inter Cooler Return

13 14

26

Upper Inter Cooler Draw Lower Inter Cooler Return

27 28

Lower Inter Cooler Draw

40

LC & LG

Vapor Inlet

Liquid Outlet

FCC-E702


SPONGE ABSORBER The sponge absorber (Figure 91) is a "guard" tower which uses LCO to absorb any remaining liquid from the gas. If any significant quantities of C5 or C6 remain in the gas stream, these will cause problems with treating. If the gas is sent directly to the fuel system, condensation of heavy material is dangerous. The sponge absorber is fabricated with killed steel. It used to be designed with 25 carbon steel trays; the tray fittings were Type 410 stainless. Currently, UOP designs all sponge absorbers with packing towers which minimizes the risk of foaming. A mesh blanket, Type 304 stainless, is installed at the top of the absorber to trap any entrained liquid. The gas leaves the tower on pressure control. Absorption LCO is flow controlled, with a lean oil/rich oil exchanger to heat the rich sponge oil as it leaves the absorber. This decreases the temperature effects when the LCO is returned to the main column. The lean oil is water cooled after it leaves the rich oil exchanger, to 100°F (38°C) maximum. The inspector of the tower should follow the same pattern of inspection as the primary absorber. The sponge absorber is subject to corrosion and hydrogen blistering, but usually somewhat less than the primary.


Figure 91 Sponge Absorber Vent and Ventilation

Vapor Outlet Mesh Blanket

Pre-Distributor

Re-Distributor Hold Down Grid

Manway Bed Support Vapor Inlet

LC-LG

Steam Out

Liquid Out FCC-E703


STRIPPER The stripper (Figure 92) removes light hydrocarbons and H2S from the gasoline. A LCO reboiler is used to heat the tower and to preheat the gasoline feed. The off-gas goes to the HPS; an FRC in the gas line controls the LCO to the reboiler. The stripper is made of killed carbon steel with about 30-40 carbon steel trays. Tray fittings are Type 410 stainless. Some units have experienced high corrosion rates in the stripper. The reboiler is usually carbon steel, but may be 5% Cr ½% Mo in cases where severe corrosion has occurred. This corrosion is not common and can usually be corrected without the additional expense of a chrome reboiler. Inhibitor has been used in some cases, but this protection can be lost with unstable operating conditions. The inspector should check metal thickness and the reboiler should be checked for unusual wear during the turnaround.


Figure 92 Stripper Vapor Outlet Vent and Ventilation

Distributor Liquid Inlet 1

18

Manway 19

36

Reboiler Return Reboiler Draw Off Well LC & LG

Liquid Out to Reboiler

Steam Out Liquid Out Vortex Breaker

FCC-E704


DEBUTANIZER The stripped gasoline is pressured to the debutanizer for vapor pressure adjustment. The LPG product after treating can be used for fuel or further processed for petrochemicals. The gasoline product usually requires treatment for sulfur, and the addition of inhibitors for better stability. The most common method of pressure control is a hot vapor bypass around the overhead condenser. The reflux back to the tower is temperature controlled by a TRC device located on the fourth tray from the top. This scheme gives a better split than overhead temperature control. Heat is supplied by a HCO reboiler, occasionally supplemented by a main column bottoms reboiler. Hot fluid to the reboiler tubes is FRC controlled. Insufficient heat will not give good fractionation, too much heat will flood the tower. The debutanizer (Figure 93) is killed steel with 35-40 carbon steel trays. Tray fittings are Type 410 stainless. The overhead system is usually carbon steel, with inhibited Admiralty tubes in the water condenser. The shell of the reboiler is carbon steel; the tubes, tube sheet and floating head cover are 5% Cr ½% Mo. Carbon steel clad with Type 410 stainless may be used instead of 5% Cr. Hot fluid flow is through the tubes. This vessel and associated piping are occasionally subject to severe corrosion and hydrogen blistering. The best solution to this and to the corrosion problems of the other vessels is water injection at the interstage receiver. As mentioned above, some refiners have also used alloy steel or monel instead of carbon steel for severe cases of corrosion in the debutanizer.


Figure 93 Debutanizer Vent and Ventilation

Vapor Outlet Reflux 1

7

TIC 8

Stripper Bottoms Inlet

19

Manway

20

35 36

40

LC & LG

Vortex Breaker

Reboiler Return Manway Steam Out Liquid Out to Reboiler and Product

FCC-E705


CORROSION Steel Corrosion It has been established that steel corrosion, hydrogen blistering, and weld cracking problems come from a water, hydrogen sulfide, ammonia, and/or cyanide environment. The overall corrosion reaction for steel is: Fe° + 2HS- = FeS + S-2 + 2H°

(1)

The amount of bisulfide ion (HS-) formed depends on the pH, temperature, and hydrogen sulfide (H2S) partial pressure and comes from the H2S dissociation: H2S = H+ + HS-

(2)

The iron sulfide (FeS) scale provides some protection against corrosion if the system pH is about 8. However, dissolved hydrogen cyanide (HCN) accelerates corrosion by destroying the protective FeS film and converting it into soluble ferrocyanide complexes: FeS + 6CN- = Fe(CN)6-2 + S-2

(3)

The ferrocyanide (Fe(CN)6-4) is easily recognized by its blue color (Prussian blue) when water samples are allowed to dry. Hydrogen Blistering Some portion of the hydrogen atoms formed in the corrosion process (Equation 1) gradually penetrates the steel through surface imperfections and diffuse into the steel as atomic hydrogen. This diffusing hydrogen reacts to form molecular hydrogen: 2H° = H2


The hydrogen molecules are larger in size and are trapped in the steel cavities causing local pressure to build up. The result is blistering and fissures in the metal. High quality killed carbon steel usually has less inclusions today than in earlier times, and blistering is usually less of a problem than before. Water Wash Most refiners feel that wash water is adequate for cyanide control. Hydrogen blistering has not been a problem in most of the FCC Gas Concentration units. Cyanide is effectively kept out of the Gas Concentration Unit through the use of an adequate wash to the Main Column overhead receiver and to the wet gas compressor Interstage suction drum. An effective water wash dilutes and scrubs the corrosive species such as hydrogen sulfide, ammonia, chlorides, and cyanides from the FCC light hydrocarbon streams. As stated previously, the water wash rate should be at least 6-7 Vol% on fresh feed to the FCC Unit. There should be no entrainment of water from the overhead receiver to the primary absorber column. Recommended water wash sources include oxygen free water, steam condensate, and boiler feed water before alkalinity adjustment. If the wash water contains oxygen, gum formation will increase and wet gas compressor fouling may occur. If the wash water contains magnesium or calcium salts, these components will precipitate increasing the fouling problems. Also, if oxygen is present, elemental sulfur may be formed causing copper strip corrosion in the gasoline. Condensate pH Overhead streams in the fractionator and gas concentration section consist mainly of hydrocarbon vapors, steam, and relatively small amounts of contaminants. During condensation, the water phase absorbs some of the contaminants and can become highly corrosive. The FCC Process condensate develops a pH of 7 to 9 because of the higher concentration of ammonia relative to the acid contaminants and because ammonia is more soluble in water than hydrogen sulfide or hydrogen cyanide.


It is recommended to maintain the pH of the condensate in the range of 8.0 - 8.5 to prevent elemental sulfur formation and reduce the corrosion rate. Polysulfide can decompose at a pH below 7.8 forming H2S and free S. This will cause the gasoline to be corrosive. The pH measurement of the high pressure separator and main column overhead receiver water must be made at site with portable pH meters. The pH of a water sample can change as much as 2 pH units if the sample is taken to the laboratory. Polysulfide Injection Some Refiners who are processing feed with high nitrogen content and have high cyanide concentrations in the Unit are using ammonium polysulfide to reduce the overall corrosion rate. The ammonium polysulfide is an additive that converts cyanides to thiocyanates by the reaction: (NH4) 2S3 + HCN = NH4SCN + S2-2 The refiner should monitor the Gas Concentration Unit carefully when injecting polysulfide, since these compounds could sometimes cause more problems than benefits. The following are some disadvantages of the ammonium polysulfide: 1)

The pH of the system should be higher than 8.0. At lower pH values the polysulfide decomposes into ammonia (NH3), hydrogen sulfide (H2S), and elemental sulfur (S°). The H2S will increase the corrosion rate and the elemental sulfur (S°) will give corrosive gasoline and equipment plugging problems.

2)

The polysulfide will commence salting out at temperatures below 38°F and plug pump suction and discharge lines.

3)

The polysulfide will decompose at temperatures close to 250°F producing H2S, NH3, and S°. As stated previously, the H2S will increase the corrosion rate and the S° will give corrosive gasoline and equipment plugging problems.


4)

An excess of 10 to 15 ppm of polysulfide over the stoichiometric amount of cyanide is desired. If over 30 ppm is present, there will be enough found in the sour water stripper bottoms to give problems in the effluent treating. Upon neutralization the (NH4)2S3 decomposes into NH3, H2S, and S°. This can result in bacterial kill in biotreating and equipment plugging.

5)

Wash water requirements are generally higher to assure that the polysulfide stays in solution. If the polysulfide does not stay in solution equipment plugging will increase. Also, processing the additional sour water can be costly.

157048 Fluidized Solids Page 1

FLUIDIZED SOLIDS Introduction A fluidized bed is formed by passing a gas upwards through a bed of solid particles. If the gas velocity (U) is above a minimum gas velocity, Umf, the solids become fluidized and the mixture behaves like a liquid.

Figure 1 Fluidized Beds Behave Like a Liquid

No Flow

U Umf

Flow

U Umf


A fluidized catalyst bed provides a number of key processing advantages. Fluidization facilitates the transfer of catalyst, which permits the continuous reaction/regeneration cycle of the FCC unit. Catalyst addition and removal is easy, simplifying catalyst management. High mass and heat transfer in a fluidized bed allow for efficient operation. For a given material and gas velocity, pressure drop is lower in fluidized beds than in fixed catalyst beds. These characteristics make the fluidized bed an attractive choice as a chemical or physical processing tool. Fluidized Bed Theory An unfluidized bed of catalyst is referred to as a fixed bed. The catalyst particles are in close contact with each other and contain void spaces. The voids are due to the spherical shape of the catalyst and the efficiency with which spheres can be packed. By introducing a fluidization medium – air, hydrocarbon vapor, or steam – the catalyst can be made to behave like a liquid. Once fluidized, the catalyst will assume a level surface and exert a pressure profile similar to that of a liquid. In this state, the catalyst can be made to flow through a pipe as in the case of the standpipes and reactor riser. In essence, it is a liquid stream. It is this principle which permits circulation of catalyst between the reactor and regenerator. When a fluidization gas is initially introduced to a fixed bed of catalyst, there is no change in the bed; the gas simply flows through the void spaces. As the velocity of the gas increases, it begins to exert a force on the bed to lift the catalyst. The void volume begins to increase as the bed expands. With a further increase in gas velocity, there is a point at which the drag (friction) force exerted on the particle is just equal to the force exerted by gravity on the catalyst. This gas velocity is referred to as the incipient fluidization velocity or minimum fluidization velocity (Umf). Any further increase in gas velocity allows the catalyst to freely move in the flowing gas, constantly colliding with other catalyst particles. At this point the catalyst bed is fluidized and behaves like a liquid.


This behavior will change as more gas is pressured through the bed. Bubbles of gas begin to appear in the bed. The point at which this occurs is defined as the minimum bubbling velocity (Umb). The bubbles may be small ones which disappear after a short rise, or larger ones that grow as they pass through the bed. As the velocity is increased further, the bubbles begin to occupy more of the bed. The bubbles bursting through the surface throw catalyst into the area above the bed. The surface of the bed is violently disturbed by the slugs, yet remains fairly distinct. As gas velocity is further increased, the bed enters a turbulent state, where the upper surface of the bed is not well defined. Slugging by large gas bubbles decreases. Bubbling bed regenerators typically operate in this region. Advantages of this regime are high solids mixing and therefore high heat and mass transfer rates, contributing to the combustion of coke off the catalyst. The mixing is primarily in the vertical direction and is driven by catalyst particles being pushed upwards on top of the bubbles then falling off and dropping down behind the bubbles as shown in Figure 2. Eventually, the gas velocity exceeds the terminal velocity of some of the particles in the bed. This regime is known as fast fluidization. Combustors typically operate in this regime. At this point, a fluidized bed is not maintained unless catalyst is recycled to the bed. Further gas rate increase moves into the pneumatic conveying regime. Without recirculation of catalyst, all catalyst will be ‘blown’ out of the vessel.


Figure 2 Bubble Driven Catalyst Mixing

Catalyst Flow Bubble Wake Drift

Gas FCC-F001


Figure 3 illustrates the pressure drop across a bed as a function of fluidization gas superficial velocity. Initially, the pressure drop increases with velocity to the point of initial fluidization, Umf. For typical FCC catalyst this value is around 0.01 ft/sec. Once fluidized, the catalyst exerts a pressure profile similar to that of a liquid. The pressure head exerted by the bed will be about equal the weight of catalyst and the gas contained in it. As a result of frictional losses, the pressure drop across the bed is slightly greater than its weight. Between the point of initial fluidization and the point at which entrainment begins, the behavior of the catalyst may be described by the following equation: P = (cat + gas) * h where  gas  0 P = Pressure Drop, psi  = Density, lbs/ft3 h= Bed Height, ft/144 At higher velocities, we approach the pneumatic conveying regime in which the entire bed is eventually entrained; hence a sharp pressure drop.


Figure 3 Pressure Drop Across a Fluidized Bed Bubbling Bed Regenerator

U-Umf < Umb

Increasing Pressure Drop Across the Bed

U>Umb Combustor U
Particulate Regime

Bubbling Regime

Slug Flow Regime

Turbulent Regime Fast Fluidization

Fixed Bed

Gas Initiation of Fluidization

Entrainment Begins

Increasing Gas Velocity P = (cat +gas) * h P = Pressure Drop, psi

gas 0

U = Gas Velocity Umf = Minimum fluidization Velocity Umb = Minimum bubbling velocity

Density, lbs/ft3 hBed Height, ft/144

Pneumatic Conveying


FCC Catalyst Fluidization FCC catalyst is a fine powder, generally smaller than most refinery catalysts. The physical properties of the catalyst effecting the fluidization characteristics are density, particle shape, and size distribution. In the Geldart powder classification system, FCC catalyst is considered a Group A material. The “A” representing ‘aeratable’. The classes of particles under the Geldhart classification system are shown in Figure 4. FCC catalyst bulk density (non-fluidized) is about 50-55 lb/ft3 (800-880 kg/m3) and is a result of catalyst composition and the manufacturing process. FCC catalyst is roughly spherical in shape. This shape is advantageous to fluidization as it tends to pack less tightly when defluidized. Spheres also lack the sharp edges which can contribute to plant erosion problems. FCC catalyst is a mixture of particle sizes ranging from 10-130 microns. The presence of fines (particles < 40 microns) is helpful for fluidization. Fines are introduced with the fresh catalyst and are produced during operation of the unit by attrition. These smaller particles move more easily in the gas and act as a lubricant between the larger particles to lower the minimum velocity required for fluidization. Generally, the more fines in the catalyst inventory, the easier the catalyst is to fluidize and the longer it takes to defluidize once the aeration medium is stopped. The physical properties of the gas also have a strong influence on fluidization. The primary factors are the gas viscosity and density. FCC unit design takes into consideration all the fluidization parameters. Changing temperatures, pressures, bed heights, and gas velocities are considered when sizing and orienting process equipment.


Figure 4 Geldart’s Particle Classification

Pp, Pg, Kg/m3

10,000 5,000 3,000 2,000

D B

1,000

A

500 300 200

C

100 10

20

50

100 200 500 1,000 2,000 dp, Microns

A: Aeratable (Umb> Umf) Have Significant Deaeration Time B: Bubbles Above Umf (Umb= Umf)

C: Cohesive D: Spoutable


Fluidization in Regenerator Superficial velocities in the FCC regenerator are considerably higher than minimum fluidization. This is because the air rate is set by the coke make in the unit, not the fluidization requirements. Less air would leave the unit behind in burning, with most of the coke still on the catalyst. The regenerator diameter could be increased to drop the superficial velocity but this would add considerable cost in construction. Additionally, uniform distribution of air is more difficult in very large vessels. The FCC design takes into account the need to regenerate catalyst, maintain fluidization, retain catalyst inventory, and minimize capital expenditures. The conventional or “Bubbling Bed” regenerator, shown in Figure 5, generally operates with superficial air velocities of 2-3.5 ft/sec. This velocity is in the turbulent fluidization regime. This regime exhibits violent bubbling and gas slugging which causes catalyst to be thrown upward into the freeboard area above the bed. Most of the catalyst settles back to the bed by gravity. However, some of the catalyst, especially fines, is carried up above the freeboard area. The regenerator vessel can thus be divided into two sections, the dense bed and the dilute phase. The dense bed is all the catalyst contained below the established bed level. The dilute phase is where larger catalyst particles separate from the gas and fall back to the bed. Any catalyst particles that do not separate in the dilute phase, enter into the regenerator cyclones. Catalyst entering the cyclones is separated by centrifugal force with the larger particles being returned to the bed via the cyclone diplegs. Catalyst fines too small to be separated by the cyclones are carried out of the regenerator with the flue gas. In a bubbling bed regenerator it is important that the coke and air are evenly distributed. The air bubbles rising through the bed result in thorough mixing in the vertical direction but little mixing horizontally. Therefore if the spent catalyst, and therefore the coke, is not uniformly distributed the areas with more coke may be short of air allowing CO to breakthrough the bed. The other areas with less coke operate with excess oxygen. This combination results in afterburning and high


temperatures in the dilute phase. For this reason a spent catalyst distributor or “ski jump” is used on all spent catalyst standpipe outlets. In many cases the air grid jet plugging pattern will be shifted to areas expected to have less coke present to reduce the air in that area. The design distance between the cyclone inlet and the surface of the dense bed is determined by the Transport Disengaging Height, or TDH. The TDH is a function of the superficial gas velocity, vessel diameter, and the particle size distribution. The amount of catalyst entrained in the gas above a fluidized bed decreases with the height above it. A given particle reaches a distance above the bed where gravitational forces overcome the upward drag forces of the gas, and the particle falls back to the bed. The smaller the particle, the greater the distance. A height is reached where the amount of entrained solids becomes constant, no more particles are overcome by gravitational forces. The particles here are to small to settle. This height, the TDH, determines what minimum distance above the bed the cyclones inlets must be placed. Other considerations for setting the cyclone design inlet level include dense bed level variations and minimum required dipleg length. To account for these considerations, the cyclone inlet height will be greater than the actual TDH. As illustrated in Figure 6, if a regenerator is operated in such a manner that the distance between the catalyst bed and the cyclone inlet is less than the TDH, the catalyst density at the cyclone inlet will be higher. This will increase the catalyst loading to the cyclone and potentially increase catalyst losses from the cyclone. In a bubbling bed regenerator the discharge of the primary cyclone diplegs which are returning hot catalyst to the bed can be directed to heat colder areas of the bed, typically near the spent catalyst inlet.


Figure 5 Conventional FCC Regenerator Flue Gas

Transport Disengaging Height

Freeboard

Catalyst Movement

Dense Bed

Gas Bubbles

Air


TDH

Height

Figure 6 Schematic Depiction of TDH

Density of Solids

U


The combustor section of aHigh Efficiency regenerator operates at much higher velocities than the conventional regenerator. The lower part of the regenerator, the combustor, operates at the low end of the fast fluidization regime at a velocity on the order of 5-6 ft/sec. At this velocity catalyst is carried upward. The catalyst travels up the combustor riser in pseudo plug flow. Much of the catalyst slips behind and backmixes as it moves upward. At the top of the combustor riser the regeneration gas and catalyst undergo a primary separation. The upper regenerator vessel is designed to operate with superficial gas velocities on the order of 2-3 ft/sec. Since the regeneration gas does not pass through the catalyst bed in the upper regenerator, a small stream of fluffing air is directed into the dense bed to keep it aerated. Since the combustor velocity is so high and the net catalyst flow is up, external catalyst recycle is necessary to maintain density in the combustor. The density in a high efficiency combustor is controlled with a slide valve on a recycle standpipe. The amount of catalyst circulated from the dense bed in the upper regenerator controls the density and temperature in the combustor. Figure 7 represents fluidization data for a commercial combustor operation at various catalyst loadings and superficial air velocities. The figure shows the pressure gradient, or fluidized density in lb/ft3, measured across the combustor section. The catalyst loading, W (lb/ft2/sec), is referred to as the flux and is the summation of both the reactor-regenerator catalyst circulation and the combustor’s external recirculation. Lines A through D represent lines of constant gas velocity. The gas superficial velocity A lies well below the catalyst transport velocity; however, at low solids flux (region X-Y) dilute phase flow exists. At condition Y, the solids flux is sufficient to choke the system and any further increase in solids loading (region Y-Z) results in a substantial density (inventory) increase. At the higher gas velocities B & C, choking takes place at much higher solids flux and results in a less abrupt change in combustor density with further increases in flux. Such a region can be referred to as fast-fluidized. Eventually, at higher velocities like D, true transport flow will be generated under which conditions no solid flux can choke the system.


This operating data demonstrates how flexible the system is in adjusting combustor inventory/coke burning capacity for various operating conditions. Consider the operating condition E at gas velocity B. If the superficial velocity is now increased to C either by a change in regenerator pressure or air rate, the combustor density will decrease to condition F at constant solids loading. However, if necessary, the combustor density may be reestablished at condition G by means of increasing the solids flux (opening the external recirculation slide valve). Note that an adjustment in combustor density also changes the combustor inventory and a change in the upper regenerator level (surge inventory). The catalyst bed density in a regenerator is primarily a function of superficial velocity. For typical bubbling bed regenerator velocities, bed densities would range from 30 to 40 lb/ft3. By contrast, typical combustor densities range from 10 to 15 lb/ft3. Figure 8 gives a correlation between bed density and superficial velocity. Figures 9, 10, & 11 show normal catalyst densities in various unit configurations.


Figure 7 Combustor Operation A

Gas Velocity, ft/sec

(Z)

B

3 3 Combustor Density, lb/ft

Combustor Catalyst Inventory, I = Density x Volume

E

C

G

F

D (Y) (Y) (X)

Catalyst Loading (W) lb / sec/ft2


Figure 8 Catalyst Bed Density

Bed Density, lb/ft 3

60 50 40 30 20 10 0 0

0.1

1.0

3.0

10.0

Superficial Velocity, ft/sec *Adapted from a paper "Fluidization and the FCC Process" by W.S. Letzch of Katalistiks


Figure 9 Catalyst Densities Side by Side FCC with Bubbling Bed Regenerator/ Tee Disengager Riser Termination

2 (30)

40-45 (640-720)

3 (50)

35 (560)

Densities in lb/ft3 (kg/m3)

35 (560)

5 (80)


Figure 10 Catalyst Densities High Efficiency Regenerator/ Highly Contained Riser Termination/ Elevated Feed

<1 (<15)

1 (20)

3 (50)

40 (640)

40-45 (640-720)

35 (560) 5 (80)

8-12 (130-200)

15-20 (240-320)


Figure 11 Catalyst Densities RFCC 2 Stage Regenerator

<1 (<15)

40-45 (640-720)

3 (50) 5 (80)

35 (560)

3 (50)

35 (560)

35 (560) 15-20 (240-320)


Fluidization in Standpipes Fluidized catalyst can be made to flow through pipe in the same manner as a liquid. Standpipes generate static head which allows transport of catalyst between the reactor and regenerator. The static head generated must be enough to overcome any pressure differences between the reactor and regenerator. As catalyst travels down the standpipe it carries gas with it which keeps the catalyst fluidized and flowing smoothly. The pressure will increase as the catalyst progresses down the standpipe because of the additional head. This increase in pressure causes the entrained gas to compress and the void volume to decrease. If the gas is compressed too much the catalyst in the standpipe becomes dearated, it will no longer flow smoothly and can cause circulation problems. A condition know as ‘slipstick’ flow can ensue where the catalyst ‘breaks’ away from the dearated mass and passes down the standpipe. Reactor temperature and reactor level swings can be a symptom of this phenomenon. Standpipe design criteria are set to ensure that pressure head build up, flux rate and velocity, and aeration levels are adequate to ensure stable operation. Figure 12 illustrates the pressure profile in a standpipe. Troubleshooting of low or erratic standpipe flow should start with a single gauge pressure survey to ensure that the right amount of head is generated in the standpipe. The head generation should be the density in the standpipe, typically 35 lb/ft3 (560 kg/m3) times the elevation change. Another useful troubleshooting exercise is to check the flow through the slide valve versus the expected flow predicted by the slide valve equation:

A x Cd x (DP x )1/2 W= 1.5 Where:

W = Catalyst Mass Flow (lb/sec) A = Total Slide Valve Open Area, in2 Cd = Valve Coefficient (0.9 Typical) DP = Slide Valve Pressure Drop (psi)  = Standpipe Density (35 lb/ft3 Typical)


Flow problems in standpipes can have several causes including:     

Too much gas entrained into the standpipe Insufficient gas to keep the standpipe fluidized Shallow standpipe angle Elbows in standpipes Poor catalyst fluidization properties (typically lack of fines <40 )

Figure 12 Standpipe Pressure Profile

Standpipe Entrance Effects

Head Generation

Height Slide Valve Entrance Effects Slide Valve DP

Pressure


Fluidization in Reactor Riser Elevated feed systems use steam and/or lift gas to accelerate catalyst up the riser. Typical velocities in the ‘lift zone’ are 12-18 ft/sec (3.6-5.5 m/sec). This velocity is in the conveying regime and transports the catalyst up the riser with minimum slippage. This velocity region results in a moderate and uniform catalyst density profile at the point of feed injection. The moderate density allows the feed droplets to penetrate the catalyst more easily resulting in more uniform catalyst/oil contacting. Feed is radially injected roughly 1/3 the distance up the riser where rapid vaporization and reaction take place. The vapor expansion results in a dramatic increase in velocity up the riser. Riser exit velocities of 60 ft/sec are typical. Feed riser residence times are on the order of two to four seconds. This would be considered dilute phase pneumatic conveying. During startup, shutdown, or emergency situations, it is important to maintain lift media in the riser to ensure that catalyst does not ‘slump’ and plug the riser. Typically, lift steam is added to the wye to ensure adequate velocity in the riser in such situations. Because fluidized solids flow like a liquid they will flow into any stagnant area. Therefore it is important to keep the feed nozzles purged with steam any time there is catalyst in the riser without feed. Fluidization in Reactor Depending on the reactor riser termination, catalyst and oil exiting the riser are separated in a ballistic manner or by centrifugal force in a vortex chamber or cyclone system. The catalyst, once separated from the reaction products, falls down into the reactor stripper. With modern, highly contained riser terminations, the reactor is essentially a large disengaging device as the majority of the reaction takes place in the riser. As in standpipes, the catalyst flowing down the diplegs is kept fluidized by the reactor vapors which are entrained with the catalyst. In direct connected cyclone systems with almost all of the catalyst circulation flowing down the primary cyclone


diplegs this gas volume can be very large, as much as 6 wt% of the total product. Once the catalyst exits the dipleg the entrained gas is exposed to high temperatures for an extended period of time in the large reactor shell volume resulting in higher light ends and coke make. This is one of the reasons for development of the vortex separation technology which provides a highly contained riser termination without the entrainment of hydrocarbons back into the reactor through the diplegs. Fluidization in Reactor Stripper The reactor stripper is similar in principle to a dense bed in the regenerator. The key difference is that the fluidization, or stripping, medium is steam and not air. Steam is introduced through pipe distributors and flows up through the bed displacing hydrocarbons from the interstitial and void spaces of the catalyst. Roughly 1.5 to 2.5 pounds of steam per thousand pounds of catalyst is required for a well designed stripper. The stripping action helps to recover valuable reaction products and maintains proper fluidization of the catalyst as it moves back to the regenerator. Typical superficial velocities are in the range of one to three feet per second. Poor stripper design can result in uneven distribution of the vapor flow. As a result of this, some of the catalyst may become defluidized while the catalyst flux in other areas of the stripper increases. This can reduce stripping efficiency due to poor catalyst/steam contacting, excessive gas entrainment down the stripper and lower residence times in the active portion of the stripper. As the catalyst flux in the active area of the stripper increases more gas can be entrained down the stripper and into the standpipe resulting in lower densities, less standpipe head generation and lower slide valve pressure drop.


Fluidization in Catalyst Coolers As stated earlier, bubbles rising through a catalyst bed result in thorough mixing in the vertical direction. In a catalyst cooler this phenomena is used to continually replace cooled catalyst with hot catalyst from the regenerator vessel by bubbling air injected near the bottom of the tubes. As the fluffing air rate is increased the rate of hot catalyst entering the catalyst cooler increases. The heat transfer coefficient between the catalyst particles and the tubes also increase with increasing fluffing air as the bed becomes more turbulent. Flow through type catalyst coolers also use a standpipe and slide valve to continually circulate hot catalyst from the regenerator through the cooler. This increases the temperature of the catalyst in the cooler further so that ~50% more duty can be obtained from a flow through cooler than a back mixed cooler.

157048 Catalyst Page 1

CATALYST INTRODUCTION The FCC catalysts in use today are a complex blend of technologies. The workhorse catalyst in this complex system is, itself, a complex blend of technologies. The main component in the workhorse catalyst is a crystalline silicaalumina material known as a zeolite, or alternatively known as a molecular sieve. The quantity and the properties of the zeolite in the catalyst can be altered to fit the individual activity and product yield requirements of the refinery. A second active component in the workhorse catalyst is typically an active alumina, included in the workhorse catalyst to provide conversion of the very large and heavy molecules in the feed, which are difficult for the molecular sieve to process. This second component also can protect the molecular sieve from contaminants in the FCCU feed, which can damage the performance of the molecular sieve. A third component is added to this catalyst cocktail to make the catalyst hard enough to meet the stringent particulate emissions requirements of the FCCU. In today’s world, other catalytic components are frequently combined with the workhorse catalyst, as catalyst additives. These additives can be included to provide increased gasoline octane, increased yields of light olefins, principally propylene, enhanced coke burning characteristics in the regenerator, and decreased SOx and NOx emissions from the regenerator.

HISTORY The first FCC catalysts were finely ground, naturally occurring clays. By today's standards, these catalysts were inexpensive, but they suffered from poor stability, activity, and integrity. Activity was less than half what it is today, and erosion caused


by the jagged edges of natural catalyst created maintenance problems which required high makeup rates to keep up unit inventory. The first major improvement in FCC catalysts was the introduction of the allsynthetic catalysts in 1946. These new catalysts were amorphous silica-alumina materials containing approximately 12% alumina and 88% silica. They were spray dried as the final step in their manufacture, which resulted in a roughly spheroidal particle. This improved particle shape provided improved fluidization characteristics and decreased equipment erosion problems. These low alumina catalysts were more active than the ground clays and produced higher octane gasoline. The early 1950's saw the next major improvement in these amorphous FCC catalysts. The catalyst alumina content was increased to 25%. Catalyst activity increased with increased surface area. Shortly after the development of the higher alumina catalyst, a silica-magnesia catalyst was introduced. Though advertised to be selective for middle distillate (light cycle oil) production, these catalysts did not regenerate sufficiently to be commercially acceptable. A breakthrough in FCC catalyst technology for maximizing gasoline selectivity occurred during the early 1960's, when it was discovered that molecular sieves could be excellent cracking catalysts. The silica - alumina chemical composition of the catalyst remained basically the same, but the structure was radically different. The crystalline zeolite has a well ordered, repeatable framework structure in contrast to the amorphous, sponge-like structure of the previous low and high alumina catalysts. The zeolite, when first synthesized, contains ion exchangeable Na+ cations. In that form the zeolite is a poor cracking catalyst. The break through discovery occurred when it was found that the replacement of Na+ with either H+ or rare earth (principally lanthanum and/or cerium) cations by ion exchange converted the zeolite into an outstanding cracking catalyst.


There are over 30 different naturally occurring zeolites, but only 8 exist in sufficient quantities to permit commercial exploitation. Over 100 have been synthesized in the laboratory. The zeolite used in today’s FCC catalyst is called a Type Y sieve and is close to a rare natural mineral called faujasite. Large quantities of this material are synthesized by the catalyst industry; roughly 50 million pounds per year are used for FCC catalyst. When zeolites were first tested as FCC catalysts, a pure zeolite was used. In the pilot plants in operation at that time, this zeolite proved far too active, more than 100 times as active as amorphous catalysts. But when the zeolites were incorporated into an amorphous base, the catalyst showed great promise. The first commercial catalysts were made with 8-10% zeolites, and showed an activity 1.5 to 2 times that of the amorphous types. To utilize this higher activity, UOP introduced the short contact time, all riser cracking concept. The short contact time minimized undesirable over cracking, while the high activity maintained good conversion. Since virtually all the cracking could be done in one pass, it was no longer necessary to recycle large quantities of heavy oil from the main column. Obviously, the fresh feed rate was increased. However, for distillate operations, the low reactor severity desired still required a high recycle rate. The Y-zeolite is made up of regularly reoccurring cage-like structures. The cage has openings into the cage of approximately 7Ǻ in diameter. The zeolite is, thus, sometimes referred to as a molecular sieve, since the constrained entrance into the zeolite cage acts as a sieve. Molecules up to a certain size can enter the cage while larger molecules are kept out. Cracking occurs inside the cage at the locations of the active sites, which are associated with the aluminum atoms inside the cage. The cracked products must then exit the catalyst. At this time, the accepted theory is that the reaction is not diffusion-controlled or limited, although the effect is present. When exposed to the severe hydrothermal conditions that exist in the regenerator, the crystalline structure of the zeolite is susceptible to significant destruction of its structure, resulting in loss of catalytic activity. A major advancement in catalyst technology occurred in the late 1960s, when it was discovered that a controlled


hydrothermal pretreatment could improve the stability of the zeolite. The resulting product was designated as “ultra-stable Y-zeolite”, commonly referred to as US-Y. The 1970s ushered in the age of Environmental Sensitivity. Air pollution concerns impacted the operations of the FCCU. This in turn ushered in the age of Catalyst Additives. The first environmental concerns that effected FCC operations were in the areas of CO pollution from the FCC regenerator and lead pollution from car exhausts. Prior to 1970, burning coke in the regenerator resulted in equal volumes of CO and CO2 in the flue gas. It was impossible to convert the CO to CO2 in the FCCU without creating unacceptable temperatures in the upper part of the regenerator.. A modestly sized FCCU (20,000 BPD) without a CO boiler typically emitted over 20,000 lbs/hr of CO into the atmosphere. In the early 1970s it was discovered that a small amount of platinum (~ 1ppm) in the FCC catalyst inventory allowed the CO to be completely converted to CO2 without causing excessive temperatures in the regenerator. In addition to eliminating CO, “complete combustion” had economic advantages as well. Today, the 1 ppm of platinum is typically added as a catalyst additive, which contains a high concentration of Pt (500 – 1000 ppm). Prior to the 1970s, tetra ethyl lead (TEL) was used to increase gasoline octane but concerns were raised that it also resulted in toxic lead emissions in automobile exhaust. In the 1970s, legislation in the U.S. was passed that reduced, and eventually eliminated, the amount of TEL that could be used in gasoline. Refiners had to take steps to regain the lost octane. The use of US-Y zeolites in the FCCU was found to help increase FCC gasoline octane. At the same time, a new zeolite with a smaller pore than Y-zeolite, designated as ZSM-5, was developed which, when added to the FCC catalyst inventory, was found to increase gasoline octane. An additive containing ZSM-5 was frequently used for this purpose in the 1980s. Sulfur oxide emissions (SOX) from the regenerator flue gas became an environmental concern in the 1980s. This led to another catalyst additive which refiners could blend with their FCC catalyst. The additive captured the SOX


produced in the regenerator and carried it back to the reactor, where the sulfur was released as H2S and left the FCCU with the product gases. The sulfur could then be recovered as elemental sulfur in a Klaus plant. In the late 1990s, concerns about nitric oxide (NOX) emissions from the regenerator began to be raised. Catalyst suppliers began to provide additives that could be used to reduce NOX. With the aid of the NOX additive NOX was converted to N2 and left with the regenerator flue gas. Also in the late 1990s, demand for propylene increased to the point where propylene became an FCC product of considerably greater value than FCC gasoline. Refiners who could produce a high quality propylene product began to run their cat crackers to maximize propylene. The use of a ZSM-5 containing additive became the variable that had the biggest impact upon increasing FCC propylene yields. Whereas, in the 1980s, a ZSM-5 additive was used to increase gasoline octane, at the start of the 21st century, ZSM-5 additives were primarily being used to increase FCC propylene yields. The trend towards increasing value of propylene continued during the period 2000 – 2005, during which time the price of propylene increased from ~$500 / metric ton to ~$900/ metric ton. Refiners have increased their use of ZSM-5 additives during this period and catalyst suppliers have responded by increasing the content of ZSM-5 zeolite in their additives. In the 1980s, ZSM-5 additives typically contained 10% ZSM-5 zeolite. By 2005, additives containing 40% ZSM-5 were common. Catalyst suppliers were also beginning to supply a multi-functional catalyst which contained, within one catalyst particle, both a Y-zeolite component and a ZSM-5 zeolite component.


MODERN FCC CATALYSTS PURPOSE OF THE CATALYST Understanding some basic principles regarding fluid catalytic cracking (FCC) catalyst performance is important to understanding catalyst technology. The modern FCC catalyst carries out a variety of functions. One important function is to provide all the process heat requirements This is achieved by the burning of coke on the catalyst in the regenerator, which: •

Heats the hydrocarbon feed up to the reaction temperature

•

Provides for the endothermic heat of cracking

•

Compensates for all the unit's heat losses

•

Heats the air from the air blower up to the temperature of the regenerator

On the reactor side, the catalyst must have sufficient activity to carry out catalytic conversion of the hydrocarbon feed before any significant amount of thermal cracking occurs and must have the selectivity characteristics that provides the type of products required by the refinery. Thus, the catalyst must have the thermal stability to maintain particle and catalytic integrity under severe regenerator conditions. It must have the physical strength to maintain particle morphology under the severe impact and erosion forces so that it remains in the unit, and it must have the proper flow characteristics to allow it to readily flow between the regenerator and the reactor. Thermal (free radical) cracking is the predominant cause for the removal of hydrocarbon groups from the ends of a hydrocarbon chain, producing most of the methane, and a large portion of the ethane and ethylene produced. Significant quantities of the larger fragments (C3 and greater) are generally not produced by


thermal cracking. Small amounts of C4 to C16 -olefins are produced. A high percentage of the olefins that are formed condense directly to coke. Catalytic cracking, by comparison, produces fewer C2 fragments and relatively few C1 fragments. With catalytic cracking, cracking occurs at the strong acid sites in the zeolite cage where the Brønsted acid sites occur. Cracking is by beta scission. The fragments that are cracked off the large gas-oil molecule are mainly in the C3 to C6 range. A large number of olefins are produced. Because of the ability of the catalyst to achieve rapid double-bond shifts, the linear olefins are generally in thermal equilibrium with each other. However, since hydrogen transfer (H-transfer) is a principal reaction and is selective for tertiary olefins, the isomeric olefins are present in less than thermal equilibrium, since some of the isomeric olefins are converted to the saturated isomer before desorbing from the catalyst surface. Basically, the catalyst carries out two classes of reactions: • •

Primary reactions, which involve only a single molecule. Secondary reactions, in which bimolecular reactions take place. These reactions involve molecules formed from the primary reactions.

Cracking of the original large gas-oil molecule is a primary reaction. Such reactions can be: • • •

•

Paraffin  smaller paraffin + olefin Alkyl naphthene  naphthene + olefin Alkyl aromatic  aromatic + olefin Multiring naphthene  alkylated lesser-ring naphthene

The secondary reactions are mainly associated with H-transfer reactions of one kind or another and generally result in the saturation of an olefin: • • •

Olefin + paraffin  paraffin + olefin Olefin + naphthene  paraffin + aromatic Olefin + olefin  paraffin + diolefin (or coke)


•

Olefin + olefin  paraffin + aromatic

The isomerization of a straight-chain olefin, after the initial formation of the olefin, is another important secondary reaction. CATALYST COMPONENTS To carry out its functions, the modern FCC catalyst is typically made up of four separate but important components: • • • •

Zeolite (molecular sieve) Active matrix component Inactive matrix component Binder

The inactive matrix component and the binder control the overall activity of the catalyst by diluting the highly active components down to the proper activity level and provide the proper particle strength and morphology. In the area of particle strength, all catalyst suppliers have made major advances in catalyst attrition resistance. In the 1990s, achieving good catalyst attrition resistance is generally not a problem. From a catalytic point of view, the zeolite and the active matrix components are the items of principal interest. The Zeolite The zeolite provides controlled Brønsted (proton donor) acidity from its crystalline structure and both Brønsted and Lewis (electron acceptor) acidity from the nonframework alumina. Non-framework alumina exists as a result of the hydrothermal removal of alumina from the silica - alumina zeolite framework. Some basic zeolite concepts help to understand how the zeolite functions.


Zeolite Reaction Pathways Zeolitic cracking is believed to begin with the transfer of a proton from the catalyst surface to an olefinic double bond in the hydrocarbon molecule, thus forming a carbenium ion. The double bond was initially created through a free radical thermal cracking step or by a Lewis acid reaction initiated by an active alumina surface. These steps are illustrated below. •

Initiation: nC10H22

CH3–(CH2)6–CH=CH–CH3+H2

•

Carbenium ion formation (Brønsted Acidity) (Hydrocarbon)

H

H

C7H 15 – C – C – CH3 + H

C7 H15 – CH = CH – CH3

+

O O | – Si – O – AL – O | | O O

H+ O O | | – Si – O – AL – O | | O O (Protonated Catalyst Surface)

Hydrocarbon Adsorbed on Catalyst Surface in Carbenium Ion Form

The carbenium ion migrates freely through the hydrocarbon molecule to increasingly stable locations, and arrives at a position where the flexing of the C–C in the β position sufficiently weakens the bond, cracking the bond to a smaller, stable carbenium ion and an olefin. This mechanism is known as Beta scission. H

H

C6 H13 – C – C – CH2 – CH3 H +

 O O | – Si – O – AL – O | | O O

C6H 13 + CH 2 = CH – CH 2 – CH 3

+

 -

O

O – AL – O | O


In this illustration, the -CH3 fragment at the other β C-C position can not generate enough vibration to crack the bond. The carbenium ion also promotes rapid skeletal isomerization to the tertiary carbon location, which is the most stable configuration, as: CH 3

|

C 2 H 5 – CH 2 – CH 2 – CH 2

C 2 H 5 – C – CH 3

+

+

O

-

-

O – AL – O | O

O

O – AL – O | O

It also promotes the bimolecular H-transfer reaction:

CH3

CH3



C2H5  C  CH3 + C10H21 +

-

+

O

-



C2H5  CH  CH3 + C10H20 +

O

O AL O

O AL O





O

O

O O AL O 

O

+

O O AL O -



O

In this illustration, the hydrocarbon molecule receiving the H atom is no longer attached to the zeolite particle and desorbs into the product stream. The molecule which donated the H atom is now doubly attached to the zeolite particle, making it more difficult to desorb. As the molecule continues to donate H atoms, it becomes more and more strongly attached to the catalyst particle, to the point where it may be impossible to desorb. It is then classified as coke.


Zeolite Structure The zeolite used in today's FCC catalyst is the Y-type faujasite. The tetrahedral structure is the basis for the entire geometry of the Y-type faujasite. The tetrahedron is present at the beginning when silicon and aluminum atoms are combined with four oxygen atoms (Figure 1) in a tetrahedral arrangement. The silica and alumina tetrahedra are then arranged at the vertices of a truncated octahedron known as the sodalite cage. This cage has 8 hexagonal faces, 6 square faces (Figure 1), and 24 silica and alumina vertices.

Figure 1 Zeolite Molecular Structure

Tetrahedron

Sodalite

Silicon or Aluminum Oxygen

The sodalite cages also combine tetrahedrally, through the hexagonal faces, with other sodalite cages (Figure 2) to form the supercage, also known as the unit cell. Eight sodalite cages make a unit cell. Some important dimensions of these structures in a Y-type faujasite crystal are: •

Sodalite cage – Entrance diameter 2.2 Å


•

•

Supercage – Entrance diameter 7.4 Å – Internal diameter 13 Å Unit cell (as synthesized) – External diameter 24.67 Å

Figure 2 Zeolite Molecular Structure Sodalite Cages

Supercage

Hexagonal Prisms Zeolites are frequently referred to as either a large pore zeolite, with a 12 member ring pore entrance; as a medium pore zeolite, with a 10 member ring opening; or as a small pore zeolite, with an 8 member ring pore opening. The Y-zeolite is a large pore zeolite with a 12 member ring.


The key to the performance characteristics of this crystal is that aluminum is trivalent and thus does not fit comfortably into a tetrahedral arrangement with four oxygen atoms, whereas silicon, being tetravalent, does. In such an arrangement, the silicon atom is electrically neutral, but each aluminum atom takes on a negativeone charge. This charge is counterbalanced by a cation, such as Na+, NH4+, Ce+2 or La+3. The presence of these charged particles directly or indirectly produces the protons on the catalyst surface that create the catalytic properties of the zeolite. Zeolite Ion Exchange When the catalyst is first manufactured, the charged particle on the catalyst surface is not a proton. Rather it is a sodium cation that comes from the sodium aluminate and sodium silicate used to produce the zeolite. Each supercage thus initially has the formula: Na54 (AlO2)54 (SiO2)138 • (H2O)250 Each aluminum atom has a corresponding sodium atom. Also, when manufactured, approximately seven aluminum atoms are in each of the eight sodalite cages, for a total of 54 aluminum atoms per super cage. In the sodium form, the crystal has poor hydrothermal stability because sodium promotes dealumination of the crystal lattice. The sodium cation is therefore removed by ion exchange during catalyst manufacturing and is replaced with either ammonium cations, which form the Brønsted acidity protons directly upon heating in the FCC unit, or with rare earth cations (principally lanthanum or cerium). The rare earth cations hydrolyze water molecules on the catalyst surface and thereby create the necessary protons. The rare earth cations are the most successful in preventing crystal dealumination. Consequently, the early zeolitic catalysts used in the 1960s and 1970s were fully rare earth exchanged.


Zeolite Dealumination Under the conditions of steam and high temperature that exist in an FCC regenerator, all Y-type faujasite zeolites dealuminate to some degree, even if fully rare earth exchanged. The crystal structure is attacked by water molecules, which extract the aluminum and deposit it within the supercage as Al(OH)3: O | O — Al- — O | O

OH + 4 H+OH-

OH

OH OH

+ Al(OH)3 + Cation • OH

The non-framework alumina (NFA) that is deposited has catalytic activity (Lewis acidity). It tends to catalyze the formation of C2 and lighter gases, olefins, and coke. The hydroxyl nest, which is formed when aluminum is removed from the framework, represents a point of framework weakness that can lead to crystal collapse. In some cases, silicon atoms migrate from the crystal surface into the crystal lattice in an annealing type of operation and fill the vacancies left by the departing aluminum atoms. Also, in other cases, silicon atoms from a collapsed framework react with the NFA to form silica-alumina compounds. A further effect of dealumination is that the crystalline structure shrinks in size, the O-Si-O bond being smaller than the O-Al-O bond. The greater the degree of dealumination, the smaller the crystalline unit cell size. Also, as aluminum is removed from the crystal structure, the ratio of SiO2 to Al2O3 in the remaining framework increases. The measurement of unit cell size by X-ray diffraction techniques then becomes a convenient way to determine the SiO2/Al2O3 ratio of the zeolite in a catalyst. Such a relationship is shown in Figure 3.


Figure 3 Effect of Dealumination on Unit Cell Size

24.70

Unit Cell Size (ao) of NA-Y Zeolite, Å

24.66 24.60 24.50

24.52 (US-Y)

24.40 24.30 24.20

5 8 10

20

30

40

50

60

SiO2/Al2O3 mol Ratio in Unit Cell

0 30 50 60

70

80

85

90

% Dealumination (Starting with 5.0 SiO2/Al2O3)

Because rare earth exchange prevents dealumination, the degree of dealumination can be controlled by controlling the amount of rare earth exchange. Thus, decreasing the degree of rare earth exchange decreases the number of aluminum atoms in the unit cell and results in a smaller unit cell size (Figure 4).


Figure 4 Effect of Rare Earth Exchange on

Equilibrated Unit Cell Size

Zeolite Unit Cell Size, Å (After 5 hr 815oC Steaming)

24.45 24.40 24.35 24.30 24.25 24.20 0

2

4

6

8

10

12

14

16

% RE2O3 on 100% Y-Zeolite One of the important catalyst developments in recent years was the realization that important selectivity benefits can be achieved by reducing the amount of alumina in the crystal structure. Because only the presence of aluminum atoms causes the protons to exist on the catalyst surface, decreasing the aluminum atoms causes the catalyst activity to decrease. Reducing the number of aluminum atoms also spreads the active sites further and further apart, making bimolecular secondary reactions (H-transfer) more difficult. This site reduction results in less olefin saturation (more olefin production) and less coke formation and also increases the strength of the individual acid sites by decreasing the interaction between sites. The net result is an increase in C3 and C4 production and an increase in aromatic hydrocarbons in the gasoline fraction. An important benefit is an increase in gasoline octane. These trends are seen in Figures 5 to 9.


Figure 5 Delta Coke at 65 wt% Conversion


Figure 6 Gasoline Olefin Content at 65 wt% Conversion


Figure 7 C6-C8 Aromatics at 65 wt% Conversion


Figure 8 Gasoline RONC at 65 wt% Conversion


Figure 9 Gasoline MONC at 65 wt% Conversion


Changing the degree of zeolite dealumination has some other subtle effects on the cracking reaction and on FCC operating conditions. Increased dealumination (lower unit cell size) increases the endothermic heat of cracking and also increases cracking activation energy. The multiple effects of changing unit cell size are illustrated in Figure 10. Although the higher heat of reaction and lower delta coke resulting from lower unit cell size increases the catalyst circulation rate and lowers regenerator temperature, it also increases coke yield because of the effect of increased heat of reaction on the FCC heat balance.

Figure 10 Effect of Equilibrium Zeolite Unit Cell Size

Specific Catalyst Activity  Coke

Heat of Reaction Octane Activation Energy

24.25

24.30

24.35

24.40

Equilibrium Zeolite Unit Cell Size , Å

24.45


The ZSM-5 Zeolite for Octane and Olefin Production The ZSM-5 zeolite cage, with a 10 member ring pore entrance, has a smaller pore mouth (~ 5.5 Ǻ) than the Y zeolite, with a 12 member ring and a 7.4Ǻ pore mouth. ZSM-5 is designated as a shape selective zeolite, meaning that the hydrocarbons that can enter the Zeolite cage are limited to the smaller molecules, such as the linear and singly branched molecules in the gasoline boiling range. Large molecules, such as found in VGOs, can not enter the ZSM-5 cage. ZSM-5 is considered to be a medium pore zeolite, compared to the Y-zeolite, which is known as a large pore zeolite. ZSM-5 also has significantly fewer aluminum atoms in the cage structure than the Y-zeolite, hence it has higher SiO2/ Al2O3 ratios. Depending on the zeolite synthesis procedure, ZSM-5 type zeolites, as produced, can have SiO2/ Al2O3 ratios in the 30 -50 range (the most common range) up to a 300 – 500 ratio. The latter are commonly designated as Silicalites, since they are nearly 100% silica. Despite the fact that these zeolites have very few aluminum atoms in the cage structure, and hence, very few active sites, they still have surprisingly good activity. ZSM-5 mainly cracks the larger olefinic molecules (C7= to C12=) in the gasoline range, reducing them to smaller olefins, mainly C3= and C4=, with C3=/ C4= ratios typically > 1. Some ethylene is also produced from the cracking of the large gasoline molecules. At typical FCC riser temperatures, ZSM-5 does not crack the saturated paraffins nor the aromatics in the gasoline boiling range. Since the high molecular weight , paraffinic molecules in gasoline are low octane components, removing them via ZSM-5 cracking, with the corresponding concentration of aromatics, results in an increase in gasoline octane, both RON and MON. Consequently, in the 1980s, ZSM-5 containing additives were used to enhance gasoline octane. They did, however, have the disadvantage of reducing the yield of gasoline.


As the demand for propylene increased in the 1990s, ZSM-5 additives began to be used to increase the yield of FCC derived C3=. That trend has continued to accelerate in the 21st century. Placing a high concentration of ZSM-5 crystalline zeolite into the FCC catalyst blend can produce high C3= yields. Yields of up to 20 wt% C3= can be achieved with the proper feed and operating conditions, compared to a C3= yield of 3-4% at conventional FCC conditions. Catalyst suppliers are actively working to produce more effective ZSM-5 additives. The additives produced in the 1980s typically contained 10% ZSM-5 crystalline zeolite. As the demand for C3= increased, the need arose for FCC catalyst blends containing higher amounts of ZSM-5. However, using large quantities of the 10% additive resulted in a dilution of the base Y-zeolite catalyst. Since only the Y-zeolite could crack VGO, the dilution resulted in an unacceptable loss of VGO cracking activity. Additives containing higher contents of ZSM-5 were developed. A second generation of additives containing 25% ZSM-5 was provided to the industry. Going higher in sieve content and maintaining adequate attrition resistance proved to be difficult, but eventually that problem was resolved. Additives containing 40% ZSM-5 with good attrition resistance are now available. Catalyst suppliers are now working on developing a single particle catalyst which contains both a Y-zeolite and a ZSM-5 zeolite. Today’s ZSM-5 additives are modified by the addition of phosphorous to the zeolite. Phosphorous combines with aluminum atoms in the zeolite crystal structure, through an oxygen bond, and enhances the stability and acidity of that alumina site. Catalyst suppliers now routinely include the equilibrium catalyst phosphorous content on their equilibrium catalyst report. Knowing the phosphorous content of the fresh additive allows the refiner to tract the amount of ZSM-5 additive in their equilibrium catalyst blend. A typical set of results for increasing amounts of ZSM-5 additive is shown in Figures 11 and 12.


Figure 11

20

50

18

48

16

46

14

44

Gasoline →

12

42

10

40 Butylenes

8 6

38 36

Propylene

4

34 Ethylene

2

32

0

30 0

0.5

1

1.5

2

2.5

ZSM-5 Crystal Content in FCC Catalyst - wt%

3

3.5

Gasoline Yield - wt%

C2=, C3=, & C4= Yields - wt%

Effect of ZSM-5 on Product Yields


Figure 12 Effect of ZSM-5 On Gasoline Octane


The Matrix The matrix is defined as the entire catalyst particle except the zeolite. Its function is to provide catalyst hardness, as well as to adjust the catalyst activity to the proper level by diluting the zeolite, and to provide some special properties, such as bottoms cracking activity and metals traps. Zeolites have one major drawback: their porosity. The pore opening is too small (7 to 8 Å) to allow the large molecules with high molecular weight to enter the zeolite cage. For this reason, some precracking must occur before the zeolite can play a major role. To some extent, the zeolite itself can provide this precracking function. The zeolite has some external, active surface area that can be reached through the large inert pores provided by the kaolin-silica binder matrix. Some mesopores in the 20 to 50 Å region are also created within the zeolite as a result of sieve collapse. These mesopores can then provide some access to the zeolite interior. Consequently, the zeolite itself can precrack a large percentage of the paraffinic molecules in the feed. The large, multi-ring aromatic molecules create a more-difficult problem. Active Alumina Matrix The active alumina matrix component provides the catalyst with the ability to crack these large molecules. As seen in Figure 13, the active alumina matrix provides significant surface area and hence active sites in the 50 to 200 Å region. These pores are sufficiently large to provide easy access to the large molecules and thus allow precracking to occur.


Figure 13 Effect of Matrix on Pore Size Distribution

When precracking is insufficient, as it usually is when no active alumina matrix is present, too few intermediate products are fed to the zeolite, and the zeolite tends to overcrack these intermediates. The net result is a loss in LCO yield (poor LCO selectivity). Thus, the addition of an active alumina matrix benefits LCO selectivity, and the benefit is greater for an aromatic feed than for a paraffinic feed (Figure 14).

Figure 14


Effect of Matrix on LCO Yield

27 Aromatic Feed @ 60% Conversion

LCO, wt-%

25 23 21

Paraffinic Feed @ 75% Conversion

19 17 15 20

40

60

20

100

120

140

160

Matrix Area, m2/g

The typical active alumina matrix achieves its activity to a large degree through Lewis acid sites. These sites are well known to cause coke and H2 formation. Thus, the selectivity achieved via active alumina cracking is expected to be substantially poorer than with zeolitic cracking. This reaction occurs for both paraffinic feeds and aromatic feeds (Figures 15 and 16).

Figure 15


Effect of Matrix on Coke Yields

5.2

Coke, wt-% FF

4.4 Aromatic Feed

3.6 2.8

Paraffinic Feed

2.0 1.2 20

40

60

20

100

Matrix Area, m2/g

120

140


Figure 16 Effect of Matrix on Dry Gas Yields


Catalyst Contaminants and Poisons Contaminant Metals (Coke and H2 Production) A number of metals typically found in FCC feed are deposited on the catalyst. These metals, including nickel, vanadium, copper and iron can catalyze unwanted dehydrogenation reactions to produce large quantities of coke and H2. Nickel is typically the strongest dehydrogenation catalyst of these metals. Vanadium generally considered to be ¼ as strong. Iron, which originates as organically bound iron in the hydrocarbon molecules of the feed, deposits on the catalyst and is an active dehydrogenation agent. However, most of the iron deposited on the catalyst originates from equipment scale and is inactive. The combined iron content, which is thus relatively inactive, is typically considered to be approximately 1/10 as strong a dehydrogenation agent as nickel. Copper has as much dehydrogenation activity as nickel but is usually present in much smaller amounts. A frequent method for expressing the combined contaminant potential of these metals is through the use of an equivalent nickel value, where: EqNi = Ni + Cu + V/4 + Fe/10. Since Copper is usually present in very low quantities and iron is such a weak dehydrogenation catalyst the equivalent nickel is often expressed simply as Ni + V/4. The ratio of hydrogen to methane is an indication of metals catalyzed dehydrogenation reactions. A sponge absorber off gas H2/C1 ratio for uncontaminated catalyst should be 0.1 to 0.3. A figure of 1.0 or greater would indicate metals contamination. One method of evaluating the effect of contaminant metals on catalyst selectivity is to monitor the equilibrium catalyst performance in a standardized laboratory test (see page 61) as the metals content changes. Some typical catalyst coke and H2 responses to increasing equilibrium catalyst (E-cat) metal contamination are shown in Figures 17 and 18. As seen in these figures, the response varies significantly from catalyst to catalyst.


Figure 17 Effect of Metals on Coke Factor Coke Factor of Equil. Catalyst

3.0 C atalyst F A B

C atalyst F

2.5

V /N i 3.3 2.4 0.5

2.0 C atalyst A & B

1.5 1.0 0.5 1,000

2,000

3,000

4,000

5,000

E q u ivalen t N ick el (N i + V /4), w t-p p m

Figure 18 Effect of Metals on H2 Factor H2 Factor of Equil. Catalyst

8 .0 7 .0

C a ta ly st F A B

C a ta ly st F

6 .0

V /N i 3 .3 2 .4 0 .5

C a ta ly st A

5 .0 4 .0 C a ta ly st B 3 .0 2 .0 1 ,0 0 0

2 ,0 0 0

3 ,0 0 0

4 ,0 0 0

E q u iv a le n t N ic k el (N i + V /4 ), w t-p p m

5 ,0 0 0


Extrapolation of Figures 17 & 18 would indicate that nickel levels above 10,000 ppmw would result in totally unacceptable H2 and Coke Factors. Recent experience has shown that these factors do not continue to increase linearly with increasing nickel content. Rather, as seen in Figure 19, they reach an upper limit, which depends upon catalyst design. It is believed that there is a limiting catalyst surface area for supporting the deposited nickel. Once the metal has covered the available area, additional metal deposits on top of already deposited metal and does not generate any additional catalytic metal surface area.

Figure 19 Effect of Nickel Upon E-Cat H2 & Coke 0.9

4.5

0.8

4 Coke

0.6

3.5 3

Hydrogen

0.5

2.5

0.4

2

0.3

1.5

0.2

1

0.1

0.5

0 0

5000

10000

15000

20000

E-Cat Coke Yield - wt%

E-Cat H2 Yield - wt%

0.7

0 25000

Nickel on E-Cat - ppmw

Metals Passivation The FCC feed additives that form chemical complexes with nickel have been successful in reducing the dehydrogenation activity of nickel. Antimony compounds, which have been in use for more than 30 years, are effective chemical additives.


Data showing the reduction in H2 yield as a result of using antimony are shown in Table 1. Bismuth-containing additives have also been found to reduce the dehydrogenation activity of nickel. Although somewhat less effective than antimony, the bismuth additives have the advantage of being less of a concern from an environmental point of view.

TABLE 1 HYDROGEN YIELD DECREASE EXPERIENCED BY PHILLIPS LICENSES USING METALS PASSIVATION Hydrogen Yield, SCFB FF

License

Catalyst 4 Ni + V, Ppm

Unpassivated

Passivated

% Charge

1

3,600

92

58

37

2

16,200

202

104

49

3

6,540

64

21

67

4

10,820

126

62

51

5

8,800

87

55

37

6

6,140

161

103

36

7

8,300

105

85

19

8

10,800

71

19

73

9

9,300

159

109

31

Average

44

Reference: W.C. McCarthy, et al. Paper No. 13, Katalistiks 3rd FCC Symposium, 1982.


Perhaps the most-active area in catalyst development in recent years has been the development of active matrices that minimize the adverse effect of metals. One advancement has been achieved through pore-structure adjustments and alteration of the surface chemistry of the aluminas, using a large-pore boehmite alumina. Such a matrix alumina is believed to encapsulate the nickel, so that the nickel surface is no longer exposed and cannot make contact with the hydrocarbons to catalyze dehydrogenation. With this technology, antimony additives provide only a small additional benefit. Equilibrium catalyst material data illustrating the performance of such a nickel trap are seen in Figure 20.

Figure 20 Catalyst Performance with Nickel Trap and Antimony 0.25 V=1,000-2,000 wt-ppm

Without Nickel Trap & with Sb

H2 Yield, wt-%

0.20

0.15

Without Sb With Nickel Trap

0.10

{

With Sb

0.05 0

1

2

3

4

Nickel, ppm (thousands)

5

6

7


Contaminant Metals (Catalyst Deactivation) Vanadium destroys the zeolite. In the presence of steam and high temperature (regenerator conditions), vanadium forms vanadic acid, a highly mobile compound that moves freely across the catalyst surface. It reacts with the aluminum in the zeolite structure to form a low melting point eutectic compound that causes the crystal structure to collapse. This collapse destroys activity. A typical example of the effect of vanadium on equilibrium catalyst activity is shown in Figure 21.

Equilibrium Catalyst MAT Activity

Figure 21 Effect of Vanadium on Catalyst Activity 70 68 66 64 62 60 58 56 54

Catalyst Addition Rate ~ Constant at 1.7% of Inventory/Day

52 3,000 3,500

4,000

4,500

5,000 5,500

6,000

6,500 7,000

Vanadium Content of Equilibrium Catalyst, wt-ppm Vanadium traps have also been an area of considerable R&D activity. One family of vanadium traps that appeared to have great promise was the titanate family, particularly barium titanate. Laboratory testing in 1984-1985 indicated that catalysts containing a barium titanate metals trap could sustain high levels of vanadium with


little loss in activity and with little increase in H2 and coke production. Unfortunately, when placed in actual commercial use, the titanates appeared to be sensitive to sulfur poisoning and have not lived up to their laboratory performance. As yet, they have shown little trapping capability in refinery use. Rare-earth-based vanadium traps, which are much less susceptible to permanent sulfur poisoning, have been considerably more successful in commercial refinery use. This technology, which was originally commercialized by Katalistiks in the late 1980s, is now being used by Grace, designated as their RV technology. Grace catalysts that incorporate the RV trap are called “Residcats.” The RV trap can be used either as a separate additive or incorporated into the cracking catalyst particle. Because of its high mobility, vanadium readily jumps from particle to particle, seeking those particle locations where it has a high affinity. Separate particle vanadium traps thus work nearly as well as in-catalyst traps. An illustration of the effectiveness of the rare earth vanadium trap is given in Figure 22.

Equilibrium Catalyst MAT Activity

Figure 22 Effect of Vanadium Trap 72 70 68 Vanadium Trap

66 64 Base Case

62 60 58 1,000

2,000

3,000

Vanadium on Equilibrium Catalyst, wt-ppm

4,000


Sodium and other alkalis or alkaline earths such as calcium, potassium and lithium are strong catalyst poisons which can have an immediate and significant impact on catalyst activity. Sodium is usually present in the feed as salt resulting from operational problems in the crude unit desalter or from purchased feeds from tankers. Poor steam quality can also be a source of these contaminants. An increase in sodium of just 0.1 wt% can cause a drop of up to 3 numbers in catalyst activity. Iron is another potential catalyst poison in high concentrations. If enough iron accumulates on the surface of the catalyst the access to the active catalyst sites may be blocked resulting in lower effective catalyst activity. Several units have reported significant loss in catalyst activity when the iron on the equilibrium catalyst exceeds ~3000 – 5000 wt ppm above the fresh catalyst iron level. Basic nitrogen compounds in the FCC feed are temporary poisons that bond with the active acid sites making them inaccessible for cracking reactions. The nitrogen is oxidized off the catalyst during regeneration and leaves as NOx compounds in the regenerator flue gas. The deactivation effect, thus, lasts only as long as the basic nitrogen compounds are present in the feed. Typically, basic nitrogen compounds make up about 1/3 of the total nitrogen compounds in the feed. Coke can also be considered as a temporary catalyst poison, which sits on the catalyst active sites and is removed during regeneration. If coke is not completely removed during regeneration, a loss in actual catalyst activity will result. Typically a loss of 1.0 – 1.5 activity numbers occurs for every 0.1wt% coke left on the catalyst. Note that the activity numbers reported on a catalyst vendor’s equilibrium catalyst report are determined after any remaining carbon is burned off. If the catalyst leaving the regenerator does have a significant amount of unremoved coke, the reported activity on the equilibrium catalyst sheet will thus not reflect the actual lower, working activity in the reactor.


CATALYST ADDITIVES CO Combustion Promoters In the early 1970s, researchers discovered that certain Group VIII metals, particularly platinum, can be incorporated into an FCC catalyst system at low concentrations (13 wt ppm) to catalyze the combustion of CO to CO 2 effectively. Of even greater importance was their discovery that at this low incorporation, the Group VIII metals did not catalyze undesirable dehydrogenation reactions during the cracking reaction. The cracking-selectivity characteristics achieved by the cracking catalyst in use in the FCCU are thus not altered when such a CO combustion promoter is added to the catalyst inventory. The use of 12 ppm platinum in FCCU circulating catalyst inventories is now widely practiced throughout the world to catalyze the combustion of CO to CO2 in the FCCU regenerator. The promoter is added to the FCC catalyst inventory either as an integral part of the fresh FCC catalyst, where it is included in the fresh catalyst at 12 wt ppm, or as a separate additive. The additive, which typically contains 5001000 wt ppm of platinum, is added to the FCCU by a small metering system that is independent of the fresh cracking-catalyst addition system. In the United States, the separate additive approach is generally used; in Europe, the additive is most commonly incorporated with the fresh cracking catalyst. Through the use of a combustion promoter, CO combustion occurs readily in the dense phase at temperatures well below 700C (1292˚F). It has been reported that promoted CO combustion occurred as low as 650C (1202˚F) in a commercial FCCU operation. Promoted combustion has a number of important benefits. First, the afterburn problem, i.e., burning in the dilute phase, is greatly reduced. By consuming CO in the dense phase, the potential heat release from burning in the dilute phase is significantly decreased. A commercial example of a 90F reduction in regenerator afterburn by using a CO combustion promoter is as follows:


Without promoter

With promoter

Typical flue gas temperature, ºF

1290

1220

Typical dense phase temperature, ºF Afterburn ΔT, ºF

1225 + 65

1245 - 25

The increased flexibility of FCCU operations resulting from using a CO combustion promoter is an even greater advantage. When using a promoter, the air rate to the regenerator can be varied to achieve any degree of CO combustion that is desired. By this mechanism, heat can readily be added to or removed from the regenerator to produce changes in regenerator temperature, catalyst circulation rate, coke yield, and reactor feed conversion. In actual practice, the regenerator temperature is normally controlled at the maximum temperature allowed by the regenerator metallurgy. Variations in feed quality in the FCCU can result in significant regenerator temperature excursions if the degree of CO combustion is unchanged (constant heat of combustion). With a CO combustion promoter present, the degree of CO combustion can be varied to hold the regenerator temperature constant at the desired maximum value even though significant changes in feed quality have occurred. This flexibility is achieved mainly through regenerator air rate control. To a lesser degree, additional flexibility is achieved by controlling the addition rate of fresh CO combustion catalyst to the catalytic cracker. When a CO combustion promoter is used, changes in the air rate affect the amount of CO converted to CO 2 both in the dilute and in the dense phase. Changing the amount of CO burned in turn affects both the dense- and dilute-phase temperatures. Increasing the amount of excess O2 in the regenerator causes an increase in CO burning in the regenerator dense phase as indicated by an increase in dense-phase temperature. Additional burning in the dilute phase also occurs, causing an increase in T (afterburning) between dense and dilute phases. As a typical example, a decrease in CO content in the regenerator flue gas from 5 to 3 vol % would result in a dilute-phase temperature increase of 55˚F, a densephase temperature increase of 29˚F, and an increase in dilute-dense T of 26˚F.


The amount of promoter used can also be a variable in influencing the degree of CO combustion. A promoted catalyst system can be classified as fully or partially promoted. A partially promoted system is one in which an increase in promoter content results in a decrease in the dilute-dense T. Conversely, a fully promoted system sees no effect on afterburn T when the promoter level is changed. An increase in the promoter content increases the proportion of CO that is burned in the regenerator dense phase, and decreases the dilute-dense T, as seen in Figure 23 . The change in dilute-dense T is mainly because of changes in the dilute-phase temperature. Changes in promoter concentration have only a small effect on dense-phase temperature but greatly affect dilute-phase temperatures.

Figure 23

Additives to Reduce Sox in Regenerator Flue Gas Recently, governmental regulations, particularly in the United States, have significantly reduced the allowable FCCU emissions for sulfur oxides. A new 50 MBPD unit or an existing 50 MBPD unit being significantly revamped in 2002 would be required to meet <0.7 t/d of sulfur oxides in the regenerator flue gas.


Currently, many refiners find that the use of a SOx adsorbing additive is the most effective way to comply with the EPA regulations.. With such an additive, the following steps occur: SO 2 is oxidized to SO3 in the regenerator: 2 SO2 + O2 → 2 SO3 SO3 is adsorbed in the regenerator by the SOx additive SO3 + MO → MSO4 MO = metal oxide and M is most commonly magnesium: The metal sulfate is reduced in the reactor MSO4 + CH4 → MS + CO2 + 2 H2O Release of H2S is released in the stripper The metal oxide adsorbent is regenerated MS + H2O → MO + H2S The H2S is carried out with the reactor products, goes through the product-recovery system of the FCCU, and eventually to further processing for sulfur recovery. The metal oxide adsorbent recirculates with the spent cracking catalyst back to the regenerator for the next SOx adsorption cycle. The first commercially effective metal oxide adsorbent consisted of a solid solution of a pure magnesium aluminate spinel (MgAl2O4) with MgO. Such a solid solution ( Mg2AlO5) does not destroy the spinel framework. The adsorption activity of the dispersed MgO in the spinel is much greater than that of pure MgO itself. Cerium is effective in oxidizing SO2 to SO 3. Consequently, a cerium-impregnated Mg2Al2O 5 actively converts the SO 2 to SO3, which is then strongly adsorbed by the dispersed MgO as MgSO4. A completely cyclic SOx removal catalyst contains, in addition to the above component, a fourth metal component such as vanadium to catalyze the conversion of MgSO4 to MgO. Multifunctional SOx removal catalyst systems have been in commercial use since 1985 in the United States. Such systems have successfully reduced SOx


emissions in the FCCU regenerator by typically 20-60%, with reductions up to 90% reported. Additive levels in the circulating catalyst inventory range from 1-10%. The additive level and the amount of SOx reduction depend on conditions, such as feed quality, the presence or absence of a CO combustion promoter, regenerator temperature, regenerator mixing efficency, and excess O2 content. Hydrotalcite, a MgO/Al2O3 containing compound, has also been found to have SOX adsorbent capability and is now being used in some SOX adsorbent additives. Recently, both spinel and hydrotalcite types of SOx reduction technology have been upgraded to higher performance standards, frequently by the addition of more magnesium oxide component. Because O2 is necessary to convert SO2 to SO3, decreasing O2 in the regenerator has been found to reduce the effectiveness of the SOx removal additive. The SOx additives used in regenerators operating in a partial CO combustion mode, where excess O2 is frequently limited to 0.2vol  in the flue gas, are less successful in reducing SOx. In such cases, SOx removal is typically 2030% less than for a full CO combustion ( 1+  excess O2) case. Additives to Reduce NOx in Regenerator Flue Gas NOx (nitrous oxides ) emissions are now recognized as a significant contributor to photochemical smog and acid rain, and therefore have come under greater regulatory scrutiny than in the past. In terms of tons emitted per year, the FCCU regenerator is one of the largest point source of NOx emissions. Only a small portion of the NOx present in the flue gas is produced through the oxidation of N2 in the regenerator air stream. The main source of the NOx comes from the combustion of organic nitrogen, originating in the FCC feed of which approximately 50% ultimately reaches the regenerator section in the coke. However, only 5% to 20% of the organic nitrogen compounds entering the regenerator end up as NOx, predominantly NO. The remainder is converted to N2. Since most of the organic nitrogen ends up as N2, there is likely a secondary reaction with CO or coke to reduce NO to N2. Although feed nitrogen is the source of NOx, researchers have found that final NOx concentrations are due more to the type of N in the feed and to


the regenerator conditions, than they are to the absolute content of N in the feed. The amount of excess oxygen may have a more direct correlation with the final NOx concentration leaving the regenerator. Increasing the O2 in the regenerator flue gas from 0.1 to 1.6% has been reported to double NOx emissions when no CO combustion promoter is present. Adding a platinum-based combustion promoter, which increases the available atomic oxygen, can increase the NOx content much more. An emerging picture of the complex NOx chemistry in the FCCU regenerator includes the following reaction networks: Pyrolysis

Oxidation

N (Coke)  HCN



N2, NO, N2O

N (Coke)  NH3



N2, NO, N2O

NO Reduction 2NO + 2 CO coke  N2 + 2 CO2 NO + CFAS*  1/2N2 + CO

* FAS = Free Active Surface

The use of platinum based CO combustion promoters, which decrease the CO concentration in the regenerator, have been found to result in increased NOx emissions. It is also believed that platinum promotes the combustion of HCN, further increasing NOx. Catalyst additives have had mixed results in reducing flue gas NOx emissions. Recent research in NOx additives has taken two approaches: the development of CO combustion promoters that do not catalyze NOx formation and additives that directly reduce NOx emissions. Non-platinum combustion promoters such as palladium and cerium on alumina have had some success. Copper on alumina has had some success in converting NOx to N2. Additives to Reduce Sulfur in Gasoline


The latest additive development has been directed at reducing the sulfur content in gasoline, which is believed to be largely due to the presence of thiophene and alkylated thiophenes in the FCC gasoline. The additives are intended to dehydrogenate the thiophenes which then can crack, releasing H2S. Zinc based catalysts on an alumina support have had mixed results. CATALYST SELECTION An enormous amount of flexibility is available in the design of FCC catalysts. The catalyst designer can choose from a large variety of zeolites. Depending on the zeolite chosen, the designer can vary the amount of zeolite, the type of ion exchange, and the type of active matrix and can include metals traps or additives. This complexity is illustrated in Figure 24. Each unique combination of these factors provides a uniquely different result, and each has some specific positive and some negative benefits. Catalyst suppliers are continuously working to improve the catalyst to provide more flexibility to meet the specific needs of each individual refiner. The catalyst advancements are mainly related to improved coke and gas selectivity in the presence of contaminant metals and in the presence of increased matrix activity for better bottoms cracking and LCO selectivity. Because of the wide variability in catalyst options and the individuality of each refinery's situation, generalizing as to which catalyst is best should be avoided. The choice is best arrived at when the refiner and catalyst supplier work closely together to take advantage of the catalyst flexibility available. UOP can also assist in this selection process by providing an assessment of the various catalyst options based on commercial experience in other UOP units and on standardized testing in UOP pilot plants.


Figure 24 Catalyst Formulation



ReHY



USY



Other Zeolite Treatments



ZSM-5





Low-Activity Matrix



Active Alumina Matrix



Nickel Trap



Vanadium Traps

Plus     

Rare Earth Variations Variations in Total Sieve Content Combustion Promoter Bottoms Cracking Additives Environmental Additives


TIME AND TEMPERATURE EFFECTS High temperatures in the presence of steam will result in loss in catalyst activity. This condition exists in the regenerator where steam is formed as a result of burning the hydrogen in the coke and in the lift zone of the reactor. The rate of hydrothermal deactivation is relatively slow at temperatures below 1300 ºF (700 ºC) and increases rapidly at temperatures greater than ~1350 – 1400 ºF (730 – 760 ºC) although thermal stability varies from catalyst type to catalyst type. Generally, hydrothermal deactivation in the FCC unit is minimal because the temperatures are typically lower than 1350 ºF (730 ºC). Steam used in the reactor for feed atomization and spent catalyst stripping does not contribute significantly to hydrothermal deactivation because of the relatively low temperatures in the riser and stripper. For a given catalyst over a set period of time (say one hour), a graphical representation of the resultant catalyst activities at various temperature levels is shown in Figure 25.

Figure 25 Catalyst Hydrothermal Deactivation

M.A.T. Conversion, wt%

80 75

Catalyst A Catalyst B Catalyst C

70 65 60 1350

1400

1450

1500

Deactivation Temperature, ºF 5 hours - 100% Steam

1550


Use of torch oil can also result in thermal deactivation if the oil is not properly atomized. CATALYST MANAGEMENT The type of catalyst used in a particular FCC unit depends on feedstock properties, the desired product make, and unit bottlenecks. The choice of catalyst should be made after careful study of all factors, including cost. Any catalyst changeover will take significant time to be effective. Varying feedstocks and operating conditions may also change yields and cloud the results of catalyst changes. Fresh catalyst is typically added to the unit for two reasons: 1. 2.

To maintain catalyst quality. To replace physical losses.

If the unit holds catalyst well, it may be necessary to withdraw equilibrium catalyst and add fresh to maintain a desired activity. If the catalyst has been damaged by excessive use of steam or torch oil, or contaminated by metals, it may be necessary to increase the catalyst makeup rate. The desired catalyst activity is based on several factors, the first of which is economics. The refiner should examine. 1.

Relationship between yield and activity.

2.

Value of any extra conversion.

3.

Rate of catalyst degradation at higher yields.

4.

Ability of downstream units to handle extra conversion.

5.

Cost of catalyst.

The addition of fresh catalyst should be done as evenly as possible. There are a variety of continuous loaders on the market which make this quite easy.


If a large batch of fresh catalyst is added too quickly, the relative activity increase may cause overcracking with lower gasoline and higher dry gas yields. The unit should be monitored with plots to follow: 1. 2. 3. 4. 5.

Catalyst activity. This is normally done weekly by the catalyst supplier. Makeup and withdrawal rates. Reactor and regenerator temperature. Conversion levels and product yields. H2/C1 ratio – This will indicate possible metals poisoning.

These procedures are required for routine monitoring of the unit. Moreover, without them, it is impossible to make intelligent decisions on catalyst management. It may be necessary to add catalyst to make up physical losses. These can be due to excessive or improper steam usage, equipment damage – such as a bypassed cyclone If catalyst losses are severe, or fresh catalyst addition is not desirable, equilibrium catalyst can be added, either from a newly purchased stock or from onhand stock. In some units, low metals equilibrium catalyst is added to replace higher metals catalyst from the inventory. This is an effective way to reduce the levels metals and therefore minimize the negative effects of the metals without the high cost of adding large amounts of fresh catalyst. This is especially true in units with resid feed stocks which are typically higher in metals than VGO. Catalyst loading and use should be reviewed periodically to ensure proper unit operation. This review should normally be done about every two months. If there are special problems such as metals, the policy will have to be reviewed more often. Care should be taken, however, that other factors such as different feedstocks do not lead to hasty, inappropriate decisions.


CALCULATION OF FRESH CATALYST ADDITION RATE The fresh catalyst addition rate is a critical factor in the operation of an FCC unit and in maintaining a desired catalyst activity. The following sample calculation illustrates a method of approximating the addition rate to achieve a desired activity. Basis of calculation: 1.

Fresh catalyst activity is 79.

2.

Equilibrium catalyst activity is 69.

3.

Desired catalyst activity is 73.

4.

Present addition rate is 2.5 short tons/day.

5.

Unit inventory is 200 short tons.

6.

Catalyst retention factor is 0.80.

The retention factor is an index attrition number which accounts for the weight fraction of fresh catalyst that is not lost from the regenerator during loading. Generally, the retention factor will vary from 0.70 to 0.80 depending on the specific catalyst. 1.

Calculate average catalyst age ACA = INV/CAR where:

ACA INV CAR

= = =

Average catalyst age in days Unit inventory in short tons Present catalyst addition rate in short tons per day


In our sample calculation: ACA = 200 tons/2.5 tons per day = 80 days 2.

Calculate the deactivation constant DC = [LN ( FCA ) - LN (ECA)] /ACA where:

DC

=

Deactivation Constant

FCA ECA

= =

Fresh Catalyst Activity Equilibrium Catalyst Activity

ACA

=

Average Catalyst Age in Days

In our sample calculation: DC = [LN (79) - LN (69)]/80 DC = 0.00169 3.

Calculate the new catalyst addition rate NAR = [INV x DC x RF]/[LN (FCA) - LN (DCA)] where:

NAR INV DC RF FCA DCA

= = = = = =

New Catalyst Addition Rate in short tons per day Unit Inventory in Short Tons Deactivation Constant Retention Factor Fresh Catalyst Activity Desired Catalyst Activity

In our sample calculation: NAR = 1200 x 0.00169 x 0.80]/lLN (79) - LN (73) NAR = 3.43 short tons per day


4.

Calculate the time required for 75% turnover at the new addition rate.  PCT  Inventory Turnover = - LN 1  INV / NAR  RF   100 

where:

PCT NAR RF

= = =

Percent Turnover desired New Catalyst Addition Rate in short tons per day Retention Factor

In our sample calculation: Time Required for 75% Inventory Turnover 75   = - LN 1 200 / 3.43  0.80   100 

= 101 days


CALCULATION OF METALS ON EQUILIBRIUM CATALYST Excessive metals concentration on equilibrium catalyst results in undesirable dehydrogenation reactions and loss of catalyst activity. The following sample calculation outlines a method of predicting the current metals concentration on catalyst. BASIS: Feed Rate Feed Specific Gravity Metals in Feed Catalyst Addition Rate Metals on Catalyst Addition Initial Metals on Equilibrium Catalyst Equilibrium Catalyst Inventory

20,000 BPD 0.9042 3 ppm 2.5 Short Tons/day ppm 1,000 ppm 200 Short Tons

Find metals concentration after 100 days of operation at the above conditions.  0.175  SG  FFR  FFM  CAR  TS -  EMI FCM  EXP         EMF =      CAR INV  0.175  SG  FFR  FFM  +  + FCM    CAR where: EMF EMI FCM SG FFR

= = = = =

Final Equilibrium Catalyst Metals Concentration in ppm Initial Equilibrium Catalyst Metals Concentration in ppm Catalyst Addition Metals Concentration in ppm. Feed Specific Gravity Feed Rate in BPD


FFM CAR INV TS

= = = =

Metals Concentration on Feed in ppm Catalyst Addition Rate in Short Tons per Day Reactor/Regenerator Catalyst Inventory in short tons Time Period in Days

 0.175  0.9042  20000  3   2.5  100  1000 0  EXP        EMF =      2.5  200   0.175  0.9042  20000  3 + 0 +     2.5

EMF

=

2996 ppm

CATALYST PROPERTIES AND TESTING Catalyst manufacturers routinely test for their customers catalyst samples taken from the FCCU circulating catalyst inventory. These are referred to as equilibrium catalyst samples, or E-cat samples. Samples are taken generally on a weekly or biweekly basis, but, in special situations, can be taken more frequently. Samples are also sometimes taken of the fines leaving the FCCU, when the FCCU appears to have a catalyst loss problem. The evaluation report which a catalyst manufacturer prepares contains valuable information that provides the refinery with a better understanding of the unit operation and enables him to improve the operation of the FCC unit. If the FCCU is having problems, the E-cat data can tell the refiner if the problems are related to deteriorating catalyst performance or whether he should look to mechanical problems in the unit. At present, there are no standard catalyst testing procedures used by the various catalyst manufacturers. Results for the same sample will most likely differ from one laboratory to another. Catalyst reports are more useful as trend indicators than as reliable guides based on their absolute values.


A catalyst test report will include information on the physical, chemical, and catalytic properties of the equilibrium catalyst. Under the category of physical properties, information will frequently be found on: Catalyst surface area – total area, zeolite area, and matrix area Zeolite unit cell size Catalyst bulk density Catalyst pore volume Catalyst particle size distribution Catalyst fluidization properties Following is a description of the various tests performed by laboratories: PHYSICAL PROPERTIES Surface Area The body of catalyst particles is made up of pores which contain the active sites where the cracking reactions occur. Nearly all (90 – 95%) of the catalyst's surface area is internal. The total surface area, measured in square meters per gram of catalyst, is made up of two components, the zeolite surface area and the area of the material around the zeolite, known as the matrix surface area. The zeolite surface area is the larger of the two and is a good indicator of the catalyst activity. The zeolite surface area measures the area within the pores of about 50Ǻ or less. The matrix surface area measures the surface area in the pores > 50Ǻ, which provide the channels through which the hydrocarbons can reach the zeolite. The active alumina incorporated for bottoms cracking is found in this higher pore range. Surface area is measured by nitrogen adsorption at very low pressure. Total surface area is typically determined by the BET method which measures adsorption at a single pressure. Matrix surface area is determined by the t-plot method, which


measures adsorption over a wide range of pressures. Zeolite surface area is typically obtained by difference. Typical surface area values for Y-zeolite based catalysts range from 70-200 m2/gm for equilibrium catalyst and up to 400 m2/gm for fresh catalyst. ZSM-5 has a much lower surface area than does Y-zeolite. Unlike Y-zeolites, ZSM-5 loses little, if any, surface area upon exposure to regenerator conditions. A ZSM-5 additive containing 25% crystalline ZSM-5 zeolite would be expected to have a surface area in the 70 90 m2/gm range, both in the fresh condition and in the “deactivated” condition. Apparent Bulk Density (ABD) The apparent bulk density of the catalyst is its density measured in grams / cc. Fresh catalysts have ABD's ranging from 0.7-1.0 grams/cc, while equilibrium catalysts vary from 0.75-1.0. The ABD depends upon the chemical composition, pore volume, and particle size distribution. Hydrothermal deactivation causes the fresh catalyst to both lose moisture content and to shrink somewhat in size. With regard to bulk density, these are counteracting influences. Bulk density, thus, does not change greatly due to hydrothermal deactivation. Thermal deactivation, which can occur with an extreme regenerator temperature excursion, does cause significant collapse of both the sieve and matrix and hence causes a significant increase in bulk density. Monitoring the equilibrium catalyst bulk density can provide an indication of such an event. FLUIDIZATION CHARACTERISTICS Some catalyst suppliers routinely test the catalyst in a fluidization test stand and report the ratio of minimum bubbling velocity / minimum fluidization velocity UMB /UMF. This ratio is a measure of the range of catalyst densities within which a smooth standpipe operation is obtained. The larger the value, the better. Particle Density


Particle density is the weight per unit volume of catalyst. Particle density is defined as: Skeletal Density + 1 (Skeletal Density  Pore Volume)

Particle density is always less than skeletal density. A comparison of density would be: Particle Density ≈ 2  ABD ≈ 1.7  CPD Compacted Particle Density (CPD) is similar to ABD, but the catalyst is tapped until it settles to a minimum volume. Skeletal Density Skeletal density is the density of the solid portion of the catalyst, exclusive of the volume of pores or voids. The values are obtained with a pycnometer after determining pore volume. Typical skeletal density for low alumina catalyst is 2.35 g/ml; for high alumina the density is 2.5 g/ml. The zeolites fall somewhere in between these two.

Pore Volume Pore volume is another indicator of the zeolite content of the catalyst. Higher pore volume would usually indicate higher zeolite content for two catalysts of the same type, but would not necessarily indicate higher activity. Pore volume can be determined by N2 adsorption, or by water titration. The values reported on the equilibrium catalyst sheet are obtained from a water titration.


The pore volume of equilibrium catalyst generally ranges from 0.2-0.5 cc/gm. The variation here is because of the many different types of catalyst, and because of different manufacturing techniques. Hydrothermal deactivation causes a slight decrease in pore volume, while thermal deactivation causes a greater decrease. Pore Diameter This is a convention used to describe the average pore size. The assumption is made that the pores are in the form of minute cylinders of diameter, PD, and length, L. By definition then, the following equations apply: SA = (PD)L Pore Volume (PV) = Therefore:

 (PD)2L 4

 PV  (PD)2   1 PD = L =     SA 4 4   (PD) L 

Therefore: PD =

4 PV SA

Using: SA in m2/g PV in cc/g PD in Angstroms The equation then becomes: PORE DIAMETER (PD) =

4 PV  104 SA


Attrition Resistance The attrition resistance of catalyst is an indication of its strength and hardness. A more attrition-resistant catalyst will erode at a lower rate as it circulates through the unit. Each catalyst supplier has their own technique for measuring attrition resistance. Comparing one supplier’s attrition value with another supplier’s value is not recommended. Typical FCCU catalyst loss due to attrition is approximately 1 wt% of the catalyst inventory per day. Particle Size Distribution The fluid properties of the catalyst are largely a function of the range of particle size. Typical particle size distributions are: % under 20 microns % under 40 microns % under 60 microns % under 80 microns % under 100 microns Average Particle Size (APS)

1 10 33 55 73 75 microns

The average size, in microns, of the particles contained in a catalyst sample is determined by the 50% point on a weight distribution plot. Equilibrium catalyst usually has an APS of 65-85 microns. The desirable particle size distribution is the coarsest one that still gives good fluidization. This coarse distribution will also result in minimum catalyst entrainment with the exiting gases. The catalyst will erode with time and provide its own supply of fines, leading to a gradual catalyst loss.


Satisfactory fluidization demands a range of particle sizes. The percentages of particles under 40 microns and over 100 microns are good indicators of the performance to be expected from the catalyst. UOP FCC units are designed for relatively coarse catalyst and normally operate quite satisfactorily with 5% or less of particles smaller than 40 microns. LOSS ON IGNITION This is the amount of material which the catalyst loses after heating for a specified period at temperatures in the range of 1000° to 1100˚F. Catalyst is usually sold on a dry basis but shipped with some moisture present. Typical values are 10-15% by weight. If the moisture content is too high, there may be problems with catalyst packing in the hopper. If the value is too low, less than 5%, the catalyst has been over dried. This could adversely affect activity and make smooth addition of fresh catalyst difficult, due to static electricity. CATALYTIC PROPERTIES The catalyst's conversion ability is determined by testing in a small micro reactor system at standard conditions. The test measures the equilibrium catalyst's activity and selectivity. A decoked catalyst sample is used to crack a typical FCC feedstock in a laboratory reactor. The resulting liquid and vapor products and the catalyst coke content are analyzed and the results compared to a laboratory standard. A catalyst activity number is reported, which is related to the conversion achieved in the lab test, while the selectivity values relate to the catalyst's undesirable characteristics of producing coke, light hydrocarbon gases, and hydrogen. These data are found on each catalyst supplier’s equilibrium catalyst sheet. Each laboratory has its own unique feed and operating conditions for this test. Although each catalyst supplier’s reported values on the same sample will be different, their values will not be greatly different, and changes reported from one sample to another should be relatively the same.


The test method does not duplicate commercial performance so the results are only relative. By comparing the results to a standard, however, the performance of the sample in a commercial unit can be predicted. It is important to remember that the sample is tested after any remaining coke is burned off so in units with high carbon on regenerated catalyst the reported activity will not correlate with the activity in the commercial reactor. SELECTIVITY The selectivity values reported for an equilibrium catalyst relate its tendencies to form coke, C1-C4 gases, and hydrogen to those of a standard reference catalyst when corrected to the same conversion. Selectivity is affected by the catalyst type as well as varying degrees of metals contamination. The catalyst's coking characteristic is measured by the coke factor, CF, which is dependent on the accumulated Ni + V level. This coke factor can be used to separate the effect the catalyst type and its accumulated metals level has on coking from the effects due to processing conditions such as the actual FCCU feedstock's coking tendency and the FCCU reactor temperature. Hydrogen (H2) production is also very sensitive to the catalyst metals contamination. The H2 producing tendency is measured by the H2/CH4 factor which usually varies from less than one up to two for metals levels up to 1500 ppm. Some catalyst suppliers provide a H2 yield value from their standard test. The other selectivity value which relates to the catalyst's gas yield is the gas factor, GF. This, like the coke factor, varies with the fresh catalyst type, but it is not as metals sensitive as is the coke factor or the H2 value . Generally, the gas factor will not give an increasing trend until the metals accumulation exceeds 2000 ppm. This factor is used to separate the effect the catalyst type and its contamination has on light gas production from FCCU processing conditions such as reactor temperature, regenerator temperature, or the feedstock quality.


CO COMBUSTION TEST Some catalyst suppliers test the equilibrium catalyst for its ability to combust CO to CO2. The reported value is given as the performance of the catalyst relative to a catalyst that is considered to be fully loaded with a CO combustion promoter.

157048 Process Variables Page 1

PROCESS VARIABLES INTRODUCTION Proper control of an FCC unit requires careful balancing of many variables. Within certain limits, the system can be controlled to provide optimum performance. The limits, or restrictions, include feedstock quality, equipment limitations, and environmental constraints. Of primary interest is the unit enthalpy balance which results from the many unit operating conditions. The amount of coke produced in the reactor is a direct result of the operating conditions imposed on the unit by the operator and is relatively independent of the feed quality. The unit yields are directly related to the quality of the feed and the operating severity set by the enthalpy balance. As these topics are developed in this chapter, it will become clear that many of the assumptions made by FCC operators are incorrect and a better understanding of the relationships of the process variables is necessary to properly predict unit operation.

REACTOR-REGENERATOR HEAT BALANCE The most fundamental principle in the operation of the FCC unit is that in steady state operation the reactor will produce just the amount of coke necessary for the regenerator to burn to satisfy the reactor energy demand. This is called the heat balance. The calculation of the heat balance is shown in the calculation section of this book. In this section, the relationship of coke burning to coke production will be discussed.


The amount of energy flowing in the unit operation can be summarized as: Energy In + Energy Produced = Energy Out + Energy Consumed Energy In

= Energy in the air to the Regenerator, raw oil, steam, and lift gas

Energy Produced

= Heat of Combustion from Coke

Energy Out

= Energy in the Flue Gas, Reactor Vapor, Steam from the Cat Cooler, Radiation Loss

Energy Consumed

= Heat of Reaction, Sensible Heat of the Feed, and the Feed Latent Heat of Vaporization

At steady state the net heat of combustion must equal the heat consumed by the reactor. Stated mathematically: NET HEAT OF COMBUSTION

BTU [∆Hcomb - ∆Hair - ∆Hloss] LB COKE MUST EQUAL TOTAL REACTOR HEAT LOAD

BTU [∆Hfeed + ∆Hdiluent + ∆Hrecycle + ∆HRx] LB FEED


The coke yield from the unit can then be written as: BTU 100  Hfeed +  Hdiluent + H recycle +  HRx  LB FEED WT% COKE YIELD ON FEED = BTU  Hcomb - H air -  Hloss  LB COKE

Note that the coke yield depends on the energy balance of the unit and the only term that represents feed quality is the heat of reaction. Thus as feed quality changes and the heat of reaction changes, there will be a change in coke yield. Otherwise, the coke yield is set by the operating conditions imposed by the operator. THE ENTHALPY OPERATING WINDOW Let us take as an example a unit operating in full CO combustion with a heavy gas oil feed in once-through (no recycle) operation. As shown in the calculation section, the enthalpy changes for diluents and normal vessel heat losses are negligible, so the coke yield equation can be simplified to:

WT% COKE =

100 H feed + H Rx  H Comb Net 

The unit control variables then become the feed temperature and the reactor temperature. The usual range of feed temperature is 350-520°F and 970-990°F for the reactor so a table showing the enthalpy changes for these ranges is:

CONTROL VARIABLE

DEPENDENT VARIABLE

REACTOR TEMP.

970 - 990°F

∆H REACTION

FEED TEMP.

350 - 520°F

REGEN CAT. TEMP.

409 - 530

BTU/LB FEED

∆H FEED


Thus, the maximum range in coke yield variability for these operating ranges is 4.8-5.8 wt%. The potential change in coke yield due to any variation in dependent variable such as heat of reaction, flue gas temperature, feed enthalpy and heat of reaction is extremely limited. The following table shows the range of coke yields for various variable changes.

DEPENDENT VARIABLE HEAT REACTION REGEN TEMP. H2 IN COKE

RANGE

ENTHALPY

COKE YIELD

LOW/HIGH CONV.

100 - 200 BTU/Lb Fd

4.6 - 5.5

1250 - 1400°F

3407 - 4010 BTU/Lb Coke

5.14 - 5.39

6.0 - 7.0

16547 - 17000 BTU/Lb Coke

5.5 - 5.3

The net coke yield is thus essentially independent of feed quality. The conversion and the cat/oil ratio are the variables that change with varying feed quality. In essence, conversion coke from the quality feed is replaced by contaminant coke from the poor feed at a correspondingly lower conversion. Another way to look at this balance is to look only at the components of the reactor side heat balance.

Feed Enthalpy Requirement Stripping Steam Enthalpy Feed Steam Enthalpy Requirement Heat of Reaction Heat Loss TOTAL

BTU/Lb Feed 530 5 13 180 2 730

% 72.6 0.68 1.78 24.67 0.27 100


Over 70% of the coke produced in the unit is used just to heat up and vaporize the feed. Only about one fourth of the coke is used to provide the heat of reaction. As we saw above, the heat of reaction does not vary very much from feed to feed, so the range of coke production can be seen clearly to be limited to less than half of 25% or about 10% of its typical value. In most units this corresponds to 5.5 wt% plus or minus 10% (5.0 to 6.05) even for wide swings in feed quality and conversion level. THE DELTA COKE OPERATING WINDOW Now that the coke yield on feed is known to be set by the operating conditions in the unit, the next important term in the heat balance is the 'delta coke'. Delta coke is the difference in coke content between the regenerated catalyst and spent catalyst. Another way to express this is the coke yield on feed divided by the catalyst/oil ratio.

DELTA COKE =

ENTHALPY COKE (or COKE YIELD) CAT/OIL

Since the coke yield is set by the operating conditions, the cat/oil and the delta coke must vary proportionally opposite to each other. The delta coke term is strongly related to the regenerator temperature and thus the product selectivities.

C TRegen = TRx + Hcomb - Hair - Hloss Cp where:

Cp = catalyst heat capacity ∆C = delta coke

Rearranging the equation, the cat/oil ratio can be calculated knowing the regenerator and reactor temperatures, and the coke yield from the heat balance calculation. Cat /Oil =

Coke Yield wt% 100 Cp

H Comb Net

TRegen

- TReactor


At constant operating conditions if the delta coke is changed the regenerator temperature and cat/oil ratio will vary as follows:

DELTA COKE

0.5

0.775

0.928

1.137

REGEN TEMP °F

1250

1330

1390

1470

CAT/OIL

11.4

7.6

6.5

5.4

5.7

5.91

6.03

6.18

COKE YIELD WT%

BASIS: 980°F Rx, 350°F Feed, 190 BTU/Lb Feed ∆HRx, FULL CO COMBUSTION, 1.0 CFR

The delta coke function therefore impacts conversion and selectivity enormously due to its influence on regenerator temperature and corresponding cat/oil. At the same time, since the regenerator temperature has little influence on the enthalpy coke, the coke yield changes very little for large variations in delta coke at constant heat of reaction. HEAT REMOVAL FROM THE REGENERATOR BY CATALYST COOLING Since we have just seen that the coke yield is set by the operating conditions and not the feed quality, what can be done when a poor quality feed must be processed and the conversion is not sufficient at the current operating conditions? At the same time, this poor quality feed usually will result in a higher regenerator temperature as well, causing a decline in liquid product yield. If a constant coke yield is required due to an air blower limitation, any heat input to the reactor side must be offset by a similar heat removal on the regenerator side. With the installation of a catalyst cooler, the feed preheat can be increased to reduce the coke make, and a corresponding heat removal via catalyst cooling can be done in the regenerator to increase the cat/oil ratio and increase conversion to maintain constant coke


production. The overall yields can be greatly improved through the increase in catalyst circulation. The higher E-cat activity resulting from the lower temperature also improves the yields. COKE YIELD FLEXIBILITY FOR OPTIMUM PERFORMANCE The prime objective of most FCC operations is the conversion of feedstock to gasoline and other valuable products, while minimizing the production of less valuable products such as clarified oil and coke. The unique feature of an FCC unit is that it supplies its own fuel, not only for the conversion to the products it produces, but also for the fractionation of the products as well. This fuel is the byproduct of the cracking reactions left on the catalyst, commonly referred to as coke. It has been the quest of designers not only to minimize coke on catalyst from the cracking reactions, but to minimize coke on catalyst from other sources as well. It has been documented that coke on catalyst can come from feed contamination such as metals, coke added by the high boiling fraction of the feed, and by entrainment from the catalyst circulation through the reactor stripper. The objective is to produce the most desirable yield pattern for a given feedstock with the least amount of coke possible. Once a unit is designed, it has a certain coke burning capability which the operator can utilize to increase cracking severity as much as desired for each feedstock. With heavier feeds that contain less hydrogen and inherently produce less liquid yield volume, high coke burning capabilities are needed and a way to increase the coke yield to improve yields is needed. The key is to produce enough coke to increase the conversion of the feed until the optimum yield pattern is achieved. QUALITY AND CONDITION OF CHARGESTOCK The typical feed to an FCC unit is a heavy gas oil, such as heavy atmospheric, light vacuum, and heavy vacuum gas oil. The unit can accommodate a range of different rates and types of feed within its design limitations. Typically a unit may run at design conditions, at a higher charge rate with moderate conversion, or at low charge rates with very high conversion levels.


Some FCC units recycle heavy oil from the main column back to the riser. The amount is defined by the combined feed ratio:

CFR =

BPD raw oil + BPD total recycle BPD raw oil

Typical combined feed ratios for riser cracking units operating with a higher activity catalyst would be 1.0-1.10. Older units running on low activity catalyst could range from 1.2 to 1.8. If the oil can be cracked on one pass and not recycled, more fresh feed can be processed. The recycle is generally heavy material which tends to be more difficult to crack, and when it does crack, it makes more coke and light gas than does the fresh feed. Many refiners run only enough recycle to return catalyst fines to the reactor from the main column. This generally means a CFR of 1.05 or less. Proper control of upstream units, such as the vacuum column, is essential to good feed quality. This control is mentioned only briefly in the following discussion because it is beyond the scope of this work, but will affect all the variables listed below. There are several important characteristics which are used to describe the raw oil charge. These relate to its ease of cracking and to most potential problems. The major elements are: RAW OIL CHARGE CHARACTERISTICS 1. 2. 3. 4. 5. 6.

°API GRAVITY AND UOP K (HYDROGEN CONTENT) BOILING RANGE AVERAGE BOILING POINT CARBON RESIDUE METALS SULFUR, NITROGEN, AND OXYGEN


°API GRAVITY API gravity is a specific gravity relating the density of oil to the density of water. Feed to an FCC can range from 15° to 45° API, with 20-25° API the most common. Any change in this number is due to a change in boiling range, crude type, or both. If the API gravity increases because the feedstock is more saturated (paraffinic and less aromatic), the following changes can be expected: 1.

The charge will crack more readily for the same reactor temperature and there will be greater conversion.

2.

At a constant conversion level, there will be a greater gasoline yield, with a slightly lower octane.

3.

Products will be less olefinic.

A rough indication of the quantities of paraffins present is a characterization factor which relates boiling point to specific gravity. This is the UOP K factor, which is given by:

(CABP) UOP K = SG60

1/3

where: CABP = Cubic average boiling point, °R SG60 = Specific Gravity at 60°F A detailed example of this calculation is given earlier. A UOP K factor of 11.2 would show a more aromatic stock, while a K factor of 12.5 would indicate a more highly paraffinic stock. One key feature of the FCC unit is that the hydrogen in the product streams must come from the hydrogen in the feed. There is no added hydrogen in this process. One good way to check the quality of FCC data and yield estimates is


to do a hydrogen balance across the unit as shown in the calculation section. Gasoline yield is a very strong function of the H2 in feed as shown:

Gasoline Yield, Vol-%

GASOLINE YIELD vs. HYDROGEN CONTENT

70

60

50 11

12

13

Hydrogen in FCC Feed, Wt-% BOILING RANGE The boiling point range of FCC feed usually varies from an initial point of 500°F (260°C) to an endpoint of about 1050°F (565°C). The distillation must be conducted under vacuum and corrected to atmospheric pressure, because thermal cracking will occur above 700°F (371°C). There are two boiling point ranges which are used to describe the lighter material in the feed. These are "percent over 430°F" (221°C), and "percent at 650°F" (343°C). The first quantifies the amount of gasoline in the feed. This material may be cracked, but only at a very slow rate. Most of it merely passes through the unit, with perhaps some small octane improvement. The octane improvement is somewhat


higher for cracked gasoline, as compared to straight run material. Gasoline in the FCC feed is not considered desirable because it occupies space which could be used to process gas oil, and because it usually has a low octane number. It would be better to remove this material in the crude unit and send it to a reformer. The "percent at 650°F" (343°C) is a measure of the light fuel oil in the charge. The boiling point of 650°F is chosen because it corresponds to the normal LCO endpoint. This material will crack, but not to the same extent as the heavier molecules. It will produce more coke than will the gasoline, but again, less than the heavier material. The endpoint of FCC charge stock may vary, depending to some extent on the suitability of the material for cracking operation. The presence of coke precursors, such as polynuclear aromatics, organometallic compounds, and high sulfur material, are, in many cases, good reasons for avoiding the inclusion of high boiling point compounds in FCC feed. This depends very much on the individual stock. AVERAGE BOILING POINT The average boiling point of the FCC feed depends on the average molecular weight. An increase in API and the molecular weight will typically cause the following: 1.

3.

The charge will crack more readily, so at constant reactor temperature the conversion will increase. At constant conversion, the gasoline yield will increase about 1% for an increase in the molecular weight of 20. This would correspond to an increase of 2° API. At constant conversion, the yield of C4 and lighter will decrease.

4. 5.

The olefinic content of the products will decrease. The regenerator temperature will tend to rise.

2.


There are upper limits to these increases. An exceptionally heavy feed might prove to be undesirable because of high Carbon Residue, Sulfur, and similar factors. The exact changes will depend on the individual feedstock. CARBON RESIDUE The carbon residue of a feedstock is an indirect measure of its coke producing nature. The value may be determined by either the Conradson or Ramsbottom methods. An increase in residue of a feed from one crude source will generally result in an increase in regenerator temperature. The exact nature of coke laydown is somewhat complicated so this characteristic is not always reliable for comparing feeds from different crudes. The carbon residue can be a useful number for determining possible contamination in storage, or of problems in the upstream feed preparation units. Because entrainment in the vacuum tower is a common cause of increased carbon residue, a higher metals level may be observed at the same time. Gas oil having a carbon residue over 0.5% should be considered suspect.



COLOR Color may be used as a possible indication of problems. Darker stocks tend to have higher carbon residues, sulfur, and metal problems. This method is especially valuable because it is a relatively quick and easy test. METALS Organometallic compounds in the FCC feed can cause serious overcracking if the metals deposit on the catalyst. The cleanliness of a chargestock is given by a metals factor:

Fm = V + 10 (Ni + Cu) where: Fm V Ni Cu

= = = =

Metals factor Vanadium concentration, wppm Nickel concentration, wppm Copper concentration, wppm

All metal concentrations are ppm by weight in the feed. A factor of 1.0 is considered safe, over 3.0 indicates a danger of poisoning. Normal dry gas make at 60-70% conversion is 50-100 SCF/bbl. A heavily contaminated stock would produce 200 SCF/bbl or more. The ratio H2/C1 will increase as dehydrogenation reactions are catalyzed by the metals, especially Ni. An H2/C1 ratio of 1.0, as compared to the normal 0.3-0.5, would again indicate metals poisoning. A third indication of metals would be an increase of 10-15% in the olefin content of the C3 stream, to as high as 85%. Sodium and vanadium pose a different threat to the catalyst than nickel. These metals are mobile at high temperatures and can destroy the zeolite in the catalyst causing low activity and conversion. Even with metal traps in the catalyst, high


catalyst usage will be required to flush these metals from the unit to maintain reasonable activity, conversion and selectivity. Proper control of the feed preparation units is necessary to prevent these metals from contaminating the feed. Most of the metals are concentrated in the heavier fractions, although a few may be volatile at lower temperatures. Keeping entrainment in the feed towers to a minimum will keep metals level down to a safe value. SULFUR Sulfur is as undesirable in cat cracker charge as it is in the feed to most refining units, causing corrosion of the equipment and increased difficulty in treating products. At 50% conversion, about 35% of the sulfur charged is converted to H2S, and at 70%, this figure will rise to about 50%. The sulfur content of a 400°F endpoint cat gasoline will be about 10% of that of the raw oil charge, but this fraction increases rapidly as the endpoint is raised above 400°F. As would be expected, the higher the sulfur content of the gasoline, the lower will be its lead response, although lead response is no longer important in many parts of the world. Hydrotreating will significantly improve FCC feedstock. The effect is twofold: the removal of impurities and the hydrogen addition to saturate molecules. The first of these is important when the charge is contaminated with sulfur, nitrogen or metals. These poisons may cause both process and environmental problems. Hydrogen addition to the feed, especially to the large polynuclear aromatics, will give higher conversion and a decreased coke yield by making these heavy compounds easier to crack. The FCC product sulfur distribution is not even; it tends to concentrate in the heavy products and as H2S. Table 1 is a typical product sulfur distribution for a nonhydrotreated feed.


TABLE 1 COMMERCIAL SULFUR BALANCE (80.7 LV-% CONVERSION) STREAM COMPOSITION WT-% SULFUR

PERCENTAGE OF FEED SULFUR

FEED:

VIRGIN GAS OILS

1.04

100.0

PRODUCTS:

SPONGE ABS. OH DEBUT. OH H2S

— —

— — 44.4

RSH GASOLINE LIGHT CYCLE OIL CLARIFIED OIL FLUE GAS SOUR WATER

0.15 1.35 2.08

3.8 6.5 16.7 17.1 9.1 1.9 99.5

Much of the feed sulfur comes off as H2S because H2S, once formed, is fairly stable under the conditions encountered in the reactor. This desulfurization is beneficial because concentrating the feed sulfur in one product stream decreases the difficulties in treating other products. Higher sulfur feedstocks that are mildly hydrotreated will tend to produce a lesser percentage of H2S, and leave a larger portion of the feed sulfur in the heavy products. The effects are shown in Table 2 for cycle oil sulfur content.


The results of this sulfur concentration in heavy products will vary with the refinery's needs. Product specifications and uses must be balanced against changes in corrosion rates and required metallurgy, and should be examined on a case by case basis. TABLE 2 SULFUR DISTRIBUTION EQUIVALENT PROCESSING CONDITIONS

SULFUR, WT-%

TOTAL CYCLE OIL SULFUR CONTENT, WT-%

PERCENTAGE OF FEED SULFUR IN CYCLE OILS

RAW

2.68

4.31

43.7

HYDROTREATED

0.56

1.40

55.8

HYDROTREATED

0.26

0.76

67.6

FEEDSTOCK TYPE

Metals The reduction of metals by hydrotreating follows the same pattern as nitrogen and oxygen removal. It is possible to reduce the feed metals factor, Fm, by 50-90% in a moderate severity hydrotreater. Hydrogen Addition Hydrotreating an FCC feedstock to improve quality is generally more attractive if gasoline or LPG is the desired product. Moderate severity LCO production may not justify hydrotreating unless there is a high concentration of feed sulfur or other contaminants. Table 3 shows the yield changes for a charge stock under equivalent cracking conditions at three levels of hydrotreating.


The increase in conversion in Table 3 is accompanied by an increase in gasoline efficiency, except for the severe hydrotreating case where potential alkylate (C3 and C4) is significantly higher. The decrease in sulfur content would be accompanied by a decrease in nitrogen and metals, while the decreasing carbon deposition will indicate a slightly higher product yield, with less feed being converted to coke.

TABLE 3 EFFECT OF FEEDSTOCK HYDROTREATING: EQUIVALENT CRACKING CONDITIONS FEED

RAW

HYDROTREATED

GRAVITY, API SULFUR, WT-% CONVERSION (LV-%) C5+ GASOLINE (LV-%)

21.7 2.68 78.1 55.5

25.5 0.53 80.7 60.1

26.4 0.28 81.7 61.0

33.3 0.01 92.0 68.2

POTENTIAL ALKYLATE (LV-%) RELATIVE CARBON DEPOSITION

33.6 1.0

37.2 0.78

37.1 0.76

41.8 0.43

FCC GASOLINE RESEARCH CLEAR OCTANE NUMBER

93.8

93.9

94.0

92.0

RATIO ISOBUTANE/BUTYLENES IN C4 FRACTION (MOL-%)

0.68

0.72

0.79

0.81

GASOLINE EFFICIENCY VOL-%, GASOLINE/CONVERSION

0.71

0.74

0.75

0.74

A detailed analysis for two different feedstocks is given in Table 4. Two obvious changes are the increase in API gravity and UOP K, indicating a greater degree of hydrogen saturation. There is some small degree of hydrocracking, as shown by the decrease in average molecular weight and boiling point.


TABLE 4 FCC FEEDSTOCK PROPERTIES GACH SARAN (HEAVY IRANIAN)

L. A. BASIN SEVERELY HYDROHYDROTREATED TREATED

RAW

HYDROTREATED

RAW

22.9

26.3

22.2

26.3

35.0

11.78 385 1.83

11.98 380 0.14

11.45 316 1.30

11.7 298 0.1

11.99 266 0.001

CON. CARBON, WT-%

0.34

0.06

0.1

0.1

0.0

N2, PPM

1880

1570

3380

2190

2

AROMATICS, WT-%

48.5

42.1

53

51

30.5

ASTM D 1160 IBP 10 30 50 70 so 95 % RECOVERED

504 661 760 822 877 958 983 98

458 632 739 799 856 933 990 98

382 548 653 724 795 880 965 98

360 530 628 707 779 868 954 98

360 440 518 594 682 810 920 98

FM FACTOR

2.3

1

2.9

0.0

0.0

PILOT PLANT CONVERSION VOL-%

77

83

68

76

89

API K MW S, WT-%


Feed hydrotreating is becoming more common as the quality of FCC feedstocks decreases. Table 5 summarizes the advantages of hydrotreating. The economic question of increased cost for hydrotreating versus increased yields and other benefits must be solved for each plant. TABLE 5 ADVANTAGES OF HYDROTREATING 1.

HIGHER CONVERSION

2.

HIGHER GASOLINE YIELD

3.

HIGHER C3-C4 YIELD

4.

LOWER SULFUR AND METALS

5.

LOWER CLARIFIED OIL YIELD

6.

LOWER COKE MAKE

J Cracking A modification of feed hydrotreating is J cracking. This process treats the light cycle oil stream, which is then recycled back to the riser with the raw feedstock. The hydrotreated LCO can be cracked, increasing the yield of all the lighter products. J cracking partially saturates the di-, tri-, and tetra-aromatics present in LCO. These components then crack instead of forming coke when the stream is recycled. Table 6 compares the product yields for J cracking and normal operation for a hydrotreated feed.


TABLE 6 FCC YIELD COMPARISON FOR J CRACKING FEED: HYDROTREATED 25° API MIDDLE EAST GAS OIL OPERATION NORMAL + J CRACKING CONVERSION, VOL-%

85

95

H2S, WT-%

0.15

0.25

C2 MINUS WT-%

2.77

3.06

C3, VOL-%

12.5

13.7

C4, VOL-%

18.0

19.6

C5- 380°F 90% GASOLINE, VOL-%

67.1

74.3

LCO, VOL-%

10.0

0.0

CLARIFIED OIL, VOL-%

5.0

5.0

COKE, WT-%

5.0

6.1

POTENTIAL C3-C4 ALKYLATE, VOL-%

32.8

34.5

TOTAL 10 PSI RVP GASOLINE, VOL-%

110.2

120.1

RESEARCH OCTANE—FCC GASOLINE

92.5

92.6

RESEARCH OCTANE—TOTAL GASOLINE

93.2

93.3

MOTOR OCTANE—TOTAL GASOLINE

85.3

85.2

YIELDS

NOTE: RVP IS REID VAPOR PRESSURE


OXYGEN AND NITROGEN Oxygen and nitrogen pose different problems than sulfur. Oxygen may be present in the feedstock chemically bonded to a hydrocarbon, or it may simply be absorbed by the oil as it sits in storage. Dissolved oxygen, if not stripped, may cause fouling of heat exchangers when the temperature approaches 400°F (200°C). Another source of oxygen is the instrument purges or the entrained oxygen carried into the reactor with the regenerated catalyst. This is more of a problem with the complete CO combustion units than with the traditional plants, because regenerator oxygen levels are high. Most of the oxygen in the reactor will be quickly converted to water, carbon oxides, phenols, cresols, or acids. Other reactions, including the problems of free oxygen, are discussed later under Product Treating. A severely contaminated feedstock may contain 4000 ppm nitrogen, a good feed less than 500-600 ppm. Much of the nitrogen will be converted to ammonia, which can cause plugging problems in the main column overhead. Cyanides are also formed, these contribute to blistering and corrosion in the gas concentration section. Wash water to the main column overhead and gas plant is used to minimize these problems. Some of the nitrogen will stay with the catalyst as a constituent of coke. It burns off in the regenerator to give nitrogen oxides and small amounts of ammonia. These problems are discussed in Section XIII, Environmental. High nitrogen levels are detrimental to catalyst activity, since the basic nature of the ammonia formed tends to deactivate the acid sites on the catalyst. This deactivation is reversible and catalyst activity will be restored with low nitrogen feed. Difficulties with nitrogen and oxygen are normally not severe enough to justify hydrotreating for only these poisons. But they rarely occur alone; a bad feedstock will usually have high concentrations of sulfur or metals in addition to the oxygen and nitrogen.


The amount of poisons removed again depends on what species are present, and the severity of treatment. Moderate hydrotreating will remove 10-50% of the total nitrogen. A high severity operation may run close to 100% removal. Treatment for oxygen gives similar results. PROCESSING OF RESIDUAL FEEDSTOCKS There are several problems and penalties associated with processing residual feedstocks. There is a loss of selectivity in liquid product yield since the feed generally contains less hydrogen than the gas oil fraction. The increased boiling range brings more contaminants into the unit, both as coke precursors and as metals. The coke precursors will cause a higher delta coke on catalyst and will result in a hotter regenerator and less conversion for the coke yield generated. The higher metals will cause more light gas production as will the higher regenerator temperatures. Catalyst activity will be effected by the high metals and by the additional deactivation from the high regenerator temperature. The higher coke yield will mean thruput will have to be reduced to stay within the coke burning capacity of the unit.


The following table from a commercial unit illustrates what happens as the feed is changed from VGO to Resid.

Commercial Operations Changing from Gas Oil to Residue Processing Feed

A

B

C

D

Composition, vol Ratio VGO / Atm. Residue

100 / 0

38 / 62

25 / 75

0 / 100

Conradson Carbon, wt-%

0.16

2.25

2.96

3.95

Metals (Ni+V), wt-ppm

1.0

9.6

9.0

6.3

Conversion, lv-%

86.2

82.2

79.7

76.5

Gasoline Yield, lv-%

62.6

60.3

59.5

57.4

Coke, wt-%

5.6

7.2

7.1

7.5

Commercial Performance

As the contamination in the feed increased as a reflection of the poorer feed quality, the coke make required to maintain conversion increased. However, by careful selection of the crudes purchased, the metal content of the resid was controlled so that their effect on the operation was limited and the additional coke burning requirement was the major change. Note that the gasoline yield was reduced by almost 5 LV% due to the change in feed.


This same dependency on feed quality is evident in the RCC unit as well. The following tables show the performance of the unit on three different quality feeds:

RFCC Unit Feedstock Variation

°API Sulfur, wt-% Nitrogen, wt-% Conradson Carbon, wt-% Metals Nickel, wt-ppm Vanadium, wt-ppm

Low Carbon Residue 22.8 0.9 0.12 4.8 8 17

Intermediate Carbon Residue 21.3 1.1 0.14 6.0 13 31

High Carbon Residue 19.2 1.2 0.19 7.9 17 52

Commercial RFCC Performance on Different Residues

Dry Gas, wt-% C3’s + C4’s, lv-% Gasoline (430°F EP), lv-% Light Cycle Oil (630°F EP), lv-% Clarified Oil, lv-% Coke, wt-% Total C5+ Liquid, lv-% RON Gasoline

Low Carbon Residue 3.4 25.2 59.1

Intermediate Carbon Residue 3.2 24.7 56.6

High Carbon Residue 4.0 23.9 55.6

15.0

14.2

15.0

7.5 8.4 106.4 91.9

10.2 9.1 105.7 93.2

10.9 10.8 105.4 93.3


Figure 1 illustrates the relative loss in conversion as a result of an increase of contaminants in the FCC feed, as expressed in terms of the Ramsbottom carbon. It is assumed that the proper catalyst makeup policies have been followed, so as not to operate with heavily contaminated catalyst (for example, total metals on catalyst of less than 5000 ppm). The loss in conversion is, in part, attributable to the deterioration of the feed quality, but the main cause is the decrease in catalyst circulation resulting from the higher regenerator temperature experienced by the unit.


Figure 2 summarizes the effect that contaminated feed has on the unit's regenerator temperature.


These plots also summarize the experiences of most refiners, who found that the units designed for VGO could not accommodate more than about 3.5 to 4.0 wt-% Conradson carbon in the feedstock due to excessively high regenerator temperatures. The high temperatures also resulted in severe catalyst deactivation and poor unit performance. This fact, i.e., that residual stocks deposit more carbonaceous material during the cracking cycle than can be effectively used to fuel the conversions reactions, led to the development of alternative technologies directed at: • •

Limiting the heat release during regeneration Seeking external removal of heat

The first approach led to the development of the two-stage regeneration system that is currently practiced in the Ashland RCC unit and the second to the development of the catalyst cooler, also incorporated in the design of this unit. Table 5 summarizes the relevant properties of an atmospheric resid that will be used to illustrate the benefits that can be obtained by varying the heat removal in a commercial unit. Case 1 found in Table 6 represents a very low coke make operation where the regenerator temperature has been allowed to equilibrate at a very high level, in the order of 1480°F. The yield structure is poor and the gas make extremely high. The subsequent cases illustrate the benefits that can be derived by manipulating the heat balance forcing the unit to produce more coke. Even though the unit is producing an additional amount of coke, the yield pattern improves. The data in Table 6 is plotted graphically in Figure 3.



TABLE 5 RCC FEEDSTOCK FOR COKE YIELD VS. CONVERSION STUDY API Conradson Carbon Metals, wt-ppm UOP K Source

20.8 4.8 20.0 12.1 Atmospheric Resid

TABLE 6 RCC ESTIMATED YIELDS VS. UNIT COKE YIELD Case Coke Yield, wt-% Conversion, LV-% Dry Gas, wt-% Gasoline, LV-% LCO, LV-% Slurry, LV-% Total Liquid, LV-% Heat Removal

1 7.3 68.3 7.2 46.0 16.9 14.8 101.6 No

2 8.57 77.2 4.2 56.3 13.4 9.4 106.6 No

3

4

9.73 80.9 3.4 58.5 11.4 7.7 107.6 Yes

11.10 83.0 3.3 59.1 9.7 6.4 107.5 Yes

Feed: Table 5 Gasoline, 380°F at 90% Pt. LCO, 600°F at 90% Pt. Installing a variable heat removal catalyst cooler enables the refiner to adjust the heat removal required for a given feedstock to produce the optimum yield pattern. Even though the coke yield has increased, the resulting yield pattern at the higher coke make is more favorable than that which can be achieved at the lower coke makes. These benefits are achieved at relatively low regenerator temperatures, of less than 1350°F, for good catalyst activity maintenance.


Commercial experience indicates that operations at regenerated catalyst temperatures greater than 1350-1400°F offer very little advantage, and, in fact, result in poor yields and very high gas production. Units currently processing some resid which are forced to operate at high regenerator temperatures are ideally suited for the installation of a UOP catalyst cooler. In addition to the conversion and selectivity benefits obtained by operating at the lower catalyst temperatures, considerable savings can be obtained by the reduction in catalyst makeup. This is due to the reduction in the hydrothermal deactivation of the catalyst as a result of the lower regenerator temperature. If it is desired to keep the unit coke make constant, another interesting application of this technology is illustrated in the following case. Table 7 shows the savings possible in a unit processing resid. As can be seen in the table, the heat balance has been adjusted to keep the coke make constant to satisfy the unit constraints. The reduction in regenerator temperature results in a lower catalyst addition rate, since the hydrothermal deactivation of the unit's catalyst has been reduced. For a 20,000 BPD unit, this catalyst savings amounts to approximately 1.6 million dollars per annum. TABLE 7 CATALYST COOLER ADDITION FOR MORE PROFITABLE OPERATION AT CONSTANT COKE MAKE Feed API Conradson Carbon, wt-% Feed Temperature, °F Reactor Temperature, °F Regenerator Temperature, °F Conversion, LV-% Coke Yield, wt-% Catalyst Addition Heat Removal Catalyst/Oil Ratio

No Cat Cooler

With Cat Cooler 27.0 3.0

Base Base 1410 85 Base Base No Base

Base + 160 Base 1350 85 Base 0.5 Base 2500 Btu/lb coke Base


The preceding discussions have illustrated the benefits that can be derived with the installation of a catalyst cooler. This flexible technology permits: •

Adjustments in the unit conversion to achieve the optimal yield pattern for a given feedstock

•

Enhancement of the unit's capability to process more contaminated feedstocks

•

A more profitable unit operation while keeping the unit coke production constant.


REGENERATION SECTION The main purpose of the regeneration zone is to oxidize the coke on spent catalyst and reestablish the catalyst activity of the fluidized particles prior to being returned to the reactor riser. The heat from this combustion of coke in turn provides the energy to satisfy the process requirements. Thus, the regeneration section has a very fundamental significance. Over the years this section has undergone fundamental design and mechanical improvements. One of the first big movements occurred during the 1970's, with the advent of total combustion. UOP moved away from the conventional bubbling or turbulent-bed regenerator to a fast-fluidized, high-efficiency style combustor. Since then this design has become well established as shown in Figure 8, and more than 45 units in commercial operation. Contrasting a modern high efficiency combustor design with a typical bed style configuration of the past, as in Figure 6, the major regenerator improvements are aimed towards: •

Enhanced and controlled coke burning kinetics

•

Reduced catalyst inventory

•

Narrow catalyst residence time distributions

•

Ease of start-up and routine operability

•

Uniform radial carbon and temperature profiles

•

Limited afterburn and uniform temperature distribution at cyclones

•

Additional heat balance flexibility from Total CO combustion Dense phase catalyst cooling

•

Particulate, power and waste heat recovery


These benefits are principally derived from the combination of a more homogeneous gas and solids contacting regime and reduced yet controlled particle residence time distributions. The following sections will discuss in detail the fundamental regenerator fluidization regimes, mixing characteristics, combustor hydraulics and coke burning kinetics, and how they relate to commercial performance and unit optimization.

FCC REGENERATOR FLUIDIZATION REGIMES In commercial FCC regenerator designs, various fluidization regimes exist. The schematics shown in Figure 4 serve to illustrate these various regimes pictorially.

FIGURE 4 FLUIDIZATION REGIMES A) Bubbling Fluidized Bed

(B)

Turbulent Fluidized Bed (C)

Fast Fluidized Bed

Transport Riser Reactor

Bed Density

(D)

Catalyst Flux

Gas Velocity

UOP 3106-6


Bubbling Bed (Umb - 1.0 ft/s Superficial Velocity) The bubbling bed regime ranges from the minimum bubble velocity Umb (which typically for FCC type A particles is from 0.02-0.1 ft/s), up to about 1.0 ft/sec superficial velocity. Here three distinct yet interchanging gas phases exist: the bubble phase, the emulsion phase, and the gas phase inside the catalyst pores. These three phases all flow at various relative velocities. Discrete bubbles of gas flowing through the bed produce abrupt pressure fluctuations at the bed surface. The relative fluctuations are determined by the bubble frequency or superficial gas velocity Ug, as shown in Figure 5.

FIGURE 5 BUBBLING TO TURBULENT BED TRANSITION


Figure 6 represents a typical bubbling bed regenerator in which there is limited solids entrainment and transport through the freeboard region. Most of the larger particles that are entrained are returned to the bed through the two stage cyclone diplegs. There exists a sharp, distinct catalyst bed level. The distance between the primary cyclone inlet horn and the surface of the bed should be greater than the transport disengaging height (TDH).

FIGURE 6 BUBBLING BED REGENERATOR


Turbulent Bed (1.0-3.5 ft/sec Superficial Velocity) With higher gas velocities, the distinct bubble phase disappears and the bulk of the gas flows as described by Yerushalmi "in voids which continually coalesce and split tracing tortuous passages as they rise through the bed"1. The upper bed surface is considerably more diffuse with reduced pressure fluctuations and substantially higher entrainment of solids into the freeboard region (Figure 5 in Fluidization). Because of the requirement for higher coke burning capacity and improved contacting efficiency, the vast majority of commercial regenerators are operating in the turbulent bed regime. In this regime the ultimate regeneration capacity is set by the sharp increase in solids entrainment as velocity increases, see Figure 7, and by the cyclone separation efficiency and dipleg hydraulics.

FIGURE 7 MAXIMUM DILUTE PHASE ENTRAINMENT IN VERTICAL GAS-SOLIDS UPFLOW

1. “Further Studies of the Regimes of Fluidization,” Powder Technology.


Fast Fluidized Bed (3 -10 ft/sec Superficial Velocity) The regime now extends into a complex transport phase where there is a sharp increase in the rate of solids carryover as the transport velocity is approached. In the absence of any solid recycle, the bed would rapidly disappear. Beyond this velocity, catalyst fed to the base of the regenerator traverses it in fully entrained transport flow with the voidage or density of the resulting suspension being dependent not only on velocity of the gas but also on the solids flow rate (flux = lb/s/ft2). If the solids rate is low, dilute-phase flow will result. If on the other hand, solids are fed to the regenerator at a sufficiently high rate, for example by recirculating solids carried-over back to the combustor, then it is possible to maintain a relatively large solids concentration referred to as the fast-fluidized bed. The transport velocity may therefore be regarded as the boundary which divides vertical gas-solids flow regimes into two groups. Below the boundary lies the bubbling, turbulent fluidized bed. Above lies the transport regime which, depending on the solids flux, encompasses a wide range of states from dilute-phase flow to the fast fluidized bed. Due to cluster formation, for FCC catalyst the transport velocity is approximately 20 times the terminal velocity of a single 50µ FCC particle.


COMBUSTOR HYDRAULICS AND MIXING CHARACTERISTICS Figure 8 shows a typical high-efficiency style regenerator. The combustor section can be operated either above or below the fluid cracking catalyst transport velocity. The resulting catalyst/air suspension in the combustor depends not only on the gas velocity but also on the solids flow rate (lb/sec/ft2). Adjustment in the quantity of catalyst being externally recirculated (via slide valve control) can therefore be used to control the catalyst inventory in the combustor for various air rates.

FIGURE 8 HIGH EFFICIENCY REGENERATOR


Figure 9 represents typical combustor hydraulics for various catalyst loadings and superficial gas velocities. The combustor density is measured across the entire combustor height and the solids loading W (lb/sec/ft2) is the summation of both the spent catalyst circulation and the combustor external recirculation. The gas superficial velocity (A) lies well below the catalyst transport velocity, but at a low solids flux (region X-Y), dilute phase flow exists. At condition (Y), the solids flux is sufficient to choke the system and any further increase in solids loading (region Y-Z) results in a substantial density (inventory/residence time) increase.

FIGURE 9 COMBUSTOR OPERATION A

Combustor Density, lb/ft.3

Combustor Catalyst Inventory Is: I = Density x V

Gas Velocity, ft/s

(Z)

B

Where V Is The Combustor Volume, ft.3

E

C

G

F

D (Y)

(Y)

(X)

Catalyst Loading (W) lb / s/ft2

At the higher gas velocities (B) and (C), choking takes place at much higher solids flux which results in a less abrupt change in combustor density with further


increases in flux. Such a region is the fast-fluidized regime. Eventually at higher velocities like (D), true transport flow will be achieved where there is no solids flux that can choke the system. This relationship shows how flexible the system is in adjusting the combustor hydraulics (inventory/residence time) for various operating conditions such as temperature, pressure, superficial gas velocity, catalyst circulation and carbon concentration. Consider the operating condition 'E' at a superficial velocity (B). If the superficial velocity is now increased to (C) either by a change in pressure or combustor air flow rate, the combustor inventory will decrease to condition 'F' at constant solids loading. However, if necessary, the combustor inventory may be reestablished at condition 'G' via an increased solids flux (external recirculation slide valve). It is important to keep in mind that any adjustment in combustor inventory results in a change to the upper regenerator level (surge inventory). Since the externally recycled solids are at final regeneration temperature, this will set the pre-combustion temperature of the combined spent catalyst, recirculated catalyst and combustion air streams. Figure 10 shows a typical response of the precombustion temperature for various external catalyst recirculation rates. The control of the quantity of solids being recycled to the combustor therefore sets both the precombustion temperature and inventory (residence time) required for complete combustion with limited afterburn and low carbon on regenerated catalyst (<0.05 Wt-%).


FIGURE 10 CATALYST PRE-COMBUSTION TEMPERATURE

Solids Mixing Characteristics The bubble or gas induced solids mixing characteristics in the various fluidized bed regimes have received considerable attention over the years. Although the bubble induced vertical mixing rate of solids is extremely high, the radial mixing characteristics are relatively poor (since bubbles flow vertically). Since some commercial bubbling/turbulent bed regenerators exceed 50 feet in diameter and have relatively low L/D bed ratios (<0.2), severe gas/solid (carbon/oxygen) distribution problems can be encountered. This results in:


1. 2. 3.

Thermal gradients, both inter-particle and within the bed. Residence time distributions and short circuiting. Afterburn in the dilute phase and cyclones.

Singer investigated catalyst mixing patterns in various commercial catalytic cracking units (reactor/regenerator/stripper sections) via the use of certain radioactive isotopes2. The measured distributions indicate that although the dense beds in the regenerator approach good mixing, there are substantial deviations from perfect mixing attributable to catalyst by-passing and regions of relatively immobile catalyst. Over the years, in order to improve the radial mixing characteristics and residence time distributions (short circuiting) of the turbulent/bubbling bed systems, considerable attention has been placed on: •

Dual diameter vessels

•

Fluidization bed length/diameter ratios

•

Air distribution, grid pressure drop and plugging patterns

•

Spent catalyst addition and withdrawal points, tangential swirls, deflector plates, baffles and entry elevation within the fluidized bed

•

Cyclone dipleg discharge orientation

2. “Catalyst Mixing Patterns in Commercial Catalytic Cracking,” I&EC, 49.

Most of these intrinsic mixing problems were eliminated with the high efficiency style combustor. Figure 11 compares the theoretical particle residence time distributions of a modern high-efficiency style combustor with that of a conventional bubbling bed. The basic difference in response curves is due to both the fluidization regime (fast-fluidized) and the solids entry/exit configuration. The combustor regenerator more closely approaches a plug flow reactor system where it is impossible for a spent catalyst particle to leave the regenerator without passing through the dilute phase combustion zone.


FIGURE 11 THEORETICAL RESIDENCE TIME DISTRIBUTIONS

Regarding radial/axial mixing, it is still highly desirable for the lower (5 feet) combustor section to be operated with a pseudo-dense phase acceleration zone. The relatively high slip and low voidage enhances the effective fast-fluidized bed thermal conductivity, allowing rapid heat transfer between the hot recycled and relatively cold spent catalyst particles up to the pre-combustion temperature. This produces uniform radial temperatures in the lower combustion zone.


COKE BURNING KINETICS AND OPTIMIZATION The coke deposited on the catalyst consists primarily of carbon and hydrogen along with relatively small amounts of sulfur, nitrogen and metals. This discussion will focus on the carbon and hydrogen species. The amount of hydrogen in coke has typically decreased over the years to levels of 6-7 wt-% with the use of high conversion, zeolite catalysts and modern stripper designs. Catalyst regeneration, as a carbon removal process, is widely accepted as being first order with respect to carbon concentration and oxygen partial pressure:  dC / dt  KCPo2

Where: K = Rate constant, hr-1 atm-1 C = Wt-% carbon on catalyst PO2 = oxygen partial pressure, atm The reaction rate constant will be dependent on temperature and can be represented via an Arrhenius type equation:

 dC / dt  Koe

 E CPo2 RT

(1)

where: ∆E = R = T =

Activation Energy Gas Constant Temperature, °R

This relationship for the rate of carbon burning from fluid catalytic cracking catalysts was found experimentally to hold over a wide range of temperatures with diffusional limitation not controlling. A similar expression can be used for the hydrogen content of the coke, with the oxidation of hydrogen proceeding more rapidly than that of carbon:


 dH / dt  K oe

E HPo2 RT

(2)

Where: Ko  Ko H = Wt-% hydrogen on catalyst When discussing coke combustion rates and kinetics, it is important to differentiate between net coke yield and coke concentration (analogous to heat and temperature). The net coke yield is set by the unit enthalpy balance with little direct impact on the rate of coke combustion of individual particles. Delta Coke

This is a dependent variable, which is difficult to accurately predict due to its dependency on feedstock quality, processing conditions and catalyst formulation. This term also sets the final regenerator temperature via:

TRegen  TRx  C / CpHcomb  H

FG

air

 Hloss

With a few simplifying assumptions, such as a plug flow combustor and total combustion of coke to CO2, the differential equations (1) and (2) can be solved numerically in order to gain insight into the impact and optimization of such process variables as: • • • •

Temperature Catalyst recirculation Oxygen partial pressure Carbon and hydrogen concentration

on the overall coke combustion rates.


Pre-combustion Mix Temperature and Catalyst Recirculation

As discussed in a previous section, controlled catalyst recirculation to the combustor serves to: • •

increase residence time via combustor hydraulics raise pre-combustion temperatures (Figure 10)

Pre-combustion temperature is the mix temperature of the cold spent catalyst, hot recycled catalyst, and combustion air: Tmix 

0.275F CCR TRegen  0.275 CCRTRx  0.273BTB 0.273B  0.275F CCR  0.275 CCR

Where: B F CCR TB 0.275 0.273

= = = =

lb air/hr Cat Recirculation, wt ratio of CCR Cat Circulation Rate, lb/hr Blower Temp, °F

= specific heat of catalyst, BTU/lb/°F = specific heat of air, BTU/lb/°F

The table below shows the substantial impact of recirculated catalyst on the combustion rates via the increased initiation temperature Tmix. Recycle

Coke Concentration

Mix Temperature

Regen Time

Zero

0.8 to 0.05 Wt-%

913

121 sec's

1.5 CCR

0.8 to 0.05 Wt-%

1161

44 sec's

For identical conditions the overall combustion time is reduced by a factor of 3. In reality this would be reduced further by additional internal recirculation (solids slip).


Regenerator Temperature and Coke Concentration

These two parameters are strongly interdependent since the final regenerated catalyst temperature is set by: TRegen = TRx + ∆C/Cp [∆HComb - ∆HAir - ∆Hlosses]

Where: ∆C = Cspent - Cregen With Cregen --> 0 for most operations The impact of temperature alone on the rate of carbon combustion is exponential and quite dramatic: Temperature Relative combustion rate

1100°F

1200°F

1300°F

1400°F

1

2.5

6

18

And slows dramatically with carbon concentration on catalyst: Carbon Reduction Relative time required For combustion

1.0 to 0.9 Wt-%

0.15 to 0.05 Wt-%

1

10

At constant temperature, oxygen partial pressure, and delta coke reduction.


Oxygen Partial Pressure

This parameter is set by the combination of regenerator operating pressure, excess air, and oxygen concentration in the supply air. The relationship between the rate of combustion and oxygen partial pressure is linear. Figures 12 and 13 show the relative effect of total pressure and excess oxygen on the regeneration time at various temperatures.

FIGURE 12 RELATIVE EFFECT OF EXCESS OXYGEN CONCENTRATION ON REGENERATION TIME AS A FUNCTION OF TEMPERATURE 1. 2. 3. 4.

Total pressure assumed constant at 30 psig Plug flow regeneration system Carbon reduced from 1.10 to 0.3 wt-% on catalyst Excess O2 change made by adjusting main air rate

Regenerator Time, sec

100

5% Excess O2

2% Excess O2 10 1120

1140

1160

1180

1200

1220

Temperature, oF

1240

1260

1280


FIGURE 13 Relative Effect of Total Pressure on Regeneration Time Requirement at Various Temperatures

Regeneration Time, Sec

1. Plug flow assumed. 2. Carbon level reduced from 1.10 TO 0.3 wt-%

150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0

0

1150F 1200F 1250F

10

20

30

40

Total Pressure, psia

50


COMBINED PROCESS VARIABLES

With some simplifying assumptions, all the process variables and their interactions can be mathematically modeled via equations (1) & (2) in order to achieve additional understanding and optimization of the combustor performance. As an example of the model's use, Figure 14 and Table 8 show the calculated time dependent responses of the numerous products of combustion for the following set of typical combustor operating conditions:

Typical Process Conditions

Regenerator pressure lb air/lb coke Hydrogen in coke Coke on spent catalyst Recycle catalyst temperature Blower discharge temperature Catalyst recycle Reactor temperature % oxygen in air CO2/(CO2+CO)

30.0 psig 14.18 6.0 wt-% 0.8 wt-% 1344°F 320°F 1.5 X’s 980°F 21 mol% 1.0

Calculated Time Dependent Variables

Particle Temperature, °F Carbon Concentration, wt-% Hydrogen Concentration, wt-% Flue Gas Oxygen Concentration, mol% Assumptions:

1. Complete combustion 2. Plug flow regeneration


FIGURE 14 COMBUSTOR COMPOSITION PROFILES 25

1500

1450

20 1400

10

1300

1250

W t- % 1200

Ca rbo nW

5

Flue Ga s t- %

O2 mol -%

1150

0

1100 0

1

2

3

4

5

10

15

20

25

30

Time (seconds)

35

40

45

50

55

60

Temperature

o Temp, F

en og dr Hy

Composition

1350 15


TABLE 8 FCC COMBUSTOR MODEL CALCULATED TIME-DEPENDENT VARIABLES

Time

Temp, °F

Carbon wt-%

0 1 2 3 4 5 10 15 20 25 30 35 40 45 50 55 60

1160 1177 1182 1206 1218 1230 1272 1295 1310 1319 1325 1330 1333 1335 1336 1338 1339

0.752 0.708 0.663 0.616 0.574 0.532 0.358 0.246 0.174 0.126 0.097 0.076 0.060 0.046 0.039 0.032 0.026

Hydrogen wt-%

0.046 0.036 0.028 0.022 0.016 0.012 0.002 0.001 -------------------------------------------------------

Flue Gas O2 mol%

21.0 18.3 17.7 16.2 14.8 13.7 8.5 7.0 5.5 4.6 3.9 3.5 3.2 2.9 2.7 2.6 2.5

Total Combustion and CO Promoter

In the absence of a CO combustor promoter, large variations in CO2/CO ratios are observed. At the catalyst surface it is believed that the ratio of CO2/CO is an intrinsic function of the temperature at the burning site ("Arthur's ratio"). However, the CO exiting the burning site may be further oxidized to CO2 at a rate dependent on temperature, CO, O2, and H2O partial pressures, active metals on the catalyst, carbon/oxygen distributions within the fluidized bed, and even the catalyst presence.


This burning of CO in the dilute phase known as "afterburning" can produce large flue gas temperature increases above those in the dense phase. This is due to the relative heats of combustion: C

CO ≈ 3,960 BTU/lb

C

CO2 ≈ 14,150 BTU/lb

Heat Combustion, BTU / lb Coke

Heat of Combustion vs. CO2/CO Ratio in Flue Gas 17000

Total Combustion (16,500 BTU / lbCoke)

16000 15000 14000 13000 12000 11000 1

2

3

4

5

6

7

CO2/CO mol Ratio As the amount of CO combustion is increased to 100%, the air required for combustion also increases. This is partially offset by the improved utilization of the heat of combustion, but full CO combustion will require a larger blower than a partial CO combustion operation if the catalyst is regenerated reasonably clean.


Combustion Air Requirement vs. CO2/CO in the Flue Gas Air/Coke (wt-Ratio)

15

Complete Combustion with 2% Oxygen in Flue Gas

14 13

Complete Combustion with No Oxygen in Flue Gas

12 11 10 1

2

3

4

5

6

7

CO2/CO mol Ratio

In the absence of catalyst, a flue gas stream of 3/1 CO2/CO mol composition undergoing complete combustion to CO2 could see a temperature increase > 600°F. (Note: this temperature increase is moderated due to the sensible heat of any entrained catalyst.) In the 1970's, in order to exploit the higher activity, more coke selective zeolite catalysts, operations with high temperature, once through total combustion were utilized for the processing of conventional high quality vacuum gas oils. Table 9 shows a comparison between old and modern operations.


TABLE 9 CATALYST FORMULATION AND PROCESS DESIGN MODIFICATIONS

Components H2S C2 C3

C4 C5+ Gasoline LCO CLO Coke

Amorphous Catalyst Wt-% LV-% 0.5 4.5

6.75 11.95 45.4 12.8 8.7 9.4 100

54.4 12.4 7.6

Process Conditions Bed cracking = 935°F CFR = 1.3 Partial burn Capacity = Base RONC = 92.5

RE-Y Zeolite Wt-% LV-% 0.4 3.75

6.45 11.45 50.2 12.55 9.4 5.8 100

59.8 12.0 8.0

Process Conditions Riser cracking = 980°F CFR = 1.0 Total combustion Capacity = 1.7 x Base RONC = 91.8

From the unit enthalpy balance: 100

Hfeed

+ HRx + Hdiluents + Hrecycle BTU/Lb feed = wt% coke on feed BTU/Lb coke Hcomb - Hair - Hlosses

The total combustion, no recycle operation resulted in substantially lower coke yields (9.4 > 5.8 Wt-%), higher liquid volume yields, and capacity increases (1.7) at equivalent conversion. These changes resulted from several major process changes, described below:


∆Hcomb: ∆Hrecycle:

9055 14,150 BTU/lb coke Zero BTU/lb feed

With the combustor operation, complete CO combustion (< 50 ppm CO in flue gas) can be achieved thermally, without promoter at temperatures greater than 1270°F and 2 mol % excess oxygen in the flue gas. However, Mobil discovered that certain Group VIII metals, particularly platinum, could be used at very low levels (1-3 wt ppm) to effectively catalyze the combustion of CO to CO2 either as an integral part of the fresh FCC catalyst or as a separate additive. In the combustor operation these additives are also frequently used but at lower concentrations than the conventional bubbling bed regenerators. The optimization of the combustor operation is simply the optimization of coke burning kinetics, with a slight twist over conventional bubbling bed systems because of the fast-fluidization regime. The combustor operation is normally quite stable, and with only a little attention from the operator, optimal conditions can be maintained without difficulty. The focus of where to begin lies in only two areas: carbon on regenerated catalyst and afterburning control. If these are both within acceptable values, no further optimization is required. To adjust these values, we need to examine the areas of control for the combustor. Temperature

The primary control point for optimization is the combustor temperature. Although the other factors are important, the temperature can be considered the one truly independent variable for the operator.


The catalyst recirculation slide valve controls the combustor temperature by adjusting the amount of hot catalyst recycled to the vessel. It is normally operated on TRC control. With adequate excess oxygen, combustor density and normal velocity, an upper combustor temperature of 1275°F should be sufficient. This temperature may need to be increased for the following reasons: •

Dilute temperatures are too high due to afterburn

•

Combustor density is low due to high air rates

•

Flue gas oxygen is low due to blower limits

•

CRC is high due to above situations

The ability to increase temperature in the combustor is, however, limited. Once the recirculation slide valve is full open, no further increase in temperature can be gained with hot recycle catalyst. The effect of other factors, such as the use of CO promoter or increased ∆coke operation, can lead to additional temperature increase. It should be noted that a flow-through catalyst cooler can limit the effectiveness of the recirculation catalyst to raise combustor temperature. The cooled catalyst flow will require additional hot catalyst recycle to maintain desired temperatures. Somewhat compensating is the high ∆coke operation that calls for the catalyst cooler in the first place. However, if it becomes necessary, the cooled catalyst slide valve can be adjusted to help optimize the combustor temperature. Density

Density and velocity together determine the residence time of the catalyst within the combustor. It is desirable to maintain at least 6 lb/ft3 density in the combustor and if possible, 9 - 10 lb/ft3 should be a normal operating target.


The combustor density is a dependent variable, since the temperature controls the recirculation slide valve, unit capacity determines the velocity based on air rate, and spent catalyst flow is dependent on the catalyst/oil ratio. However, where possible the density (residence time) should be increased for the following situations: •

CRC is high due to insufficient combustion time

•

Dilute temperatures are too high due to afterburn

Increased density will promote better thermal mixing and will increase residence time to help resolve these conditions. Also keep in mind that as density increases, the combustor catalyst inventory increases and the upper regenerator level will decrease. Velocity

Combustor velocity is a dependent variable determined by the amount of air needed to complete the combustion and thus is not a variable available to the operator to optimize other combustor conditions. However, excessive levels of flue gas oxygen (>2.5%) provide little benefit and only serve to increase combustor velocity unnecessarily. A good range for combustor velocity is 4 to 5.5 ft/s. The lower end of the range is generally better when there is a choice. When velocity exceeds 5.5 ft/s, the unit can become combustion limited and increased afterburning may be observed. At high combustor velocities due to capacity demands, CO promoter can be a valuable addition. Although velocities in excess of 6 ft/s have been observed commercially, they have been run in a promoted operation. Combustor velocity may be too high if the following conditions are observed: •

Afterburn increases despite other efforts to minimize

•

CRC increases even with high combustor temperature


Excess Oxygen

The air rate and regenerator pressure will determine the oxygen partial pressure in the combustor. Although higher levels of excess oxygen via increased air rate are beneficial to the combustion kinetics, they are counterproductive to the unit economics. Velocity will increase as air rate is increased, negating some of the advantages of higher oxygen partial pressure. Thus, flue gas oxygen should be maintained at or below 2 mol%. There are certain situations when an increase in oxygen beyond normal flue gas levels may be warranted: •

Velocity is low and combustor temperature cannot be increased further

•

CRC is high despite other efforts to minimize

One element that can be considered to increase oxygen partial pressure in the combustor without increasing air rate is the use of pure oxygen injection. In certain situations this may be economically feasible. CO Combustion Promoter

If all other conditions in the combustor are optimized, the use of promoter will not be required. However, it is perhaps inevitable that the FCC unit will be pushed to the limits such that the combustor conditions will generally exceed optimum values. In this case, the use of promoter can provide an additional measure of safety for controlling afterburn and avoiding excessive dilute phase temperatures. Particularly when maximum unit capacity is required beyond design levels, the use of promoter can be beneficial. The following conditions might suggest that promoter should be considered: •

Velocity is high, density is low and recirculation catalyst is maximized

•

Dilute temperatures are high due to the previous conditions

•

Afterburn is high despite high combustor temperatures


Maximum Capacity

Since many units operate on the basis of pushing the FCC capacity to the limit, it is appropriate to discuss what will happen in the combustor. Prior comments on the operating variables have addressed some of the concerns of higher capacity operation, but an overall perspective is needed. In general, pushing capacity to the maximum will result in a combustor operating at high velocity, low density and low catalyst residence time. Directionally, this will tend to increase afterburn and carbon on regenerated catalyst. From earlier data shown regarding coke burning kinetics, carbon is reduced to levels around 0.15 wt% very quickly, but to get to 0.05 wt% requires substantial additional time. In a maximum capacity operation, there may not be sufficient time in the combustor to reach carbon levels of 0.05 wt% or below. It is important to point out that this is not all bad, and in fact may be economically advantageous. With the use of promoter to maintain control of after burning, it is possible to allow the CRC to increase as a trade off for more feed capacity. Although effective catalyst activity is reduced, the reduction may be small enough to justify the higher capacity operation. Optimization Summary

All of the above variables need to be thoroughly understood, especially how they interact with each other. Application of this knowledge properly will result in arriving at the best operation for each particular unit. Optimization should be considered as a continuous effort, since what is optimal in one set of circumstances may not be in another. As a final note, it should be emphasized that UOP is always interested in obtaining feedback on the operation and experiences of these units from the refiner. Our Technical Service department is always available to provide assistance or consultation as needed. Only through working together can we hope to continue to improve our designs and their performance capability.


MAIN COLUMN BOTTOMS AND SLURRY SETTLER

INTRODUCTION

The main column bottoms system on an FCC presents some unusual mechanical and operating problems. The vapors entering the column are superheated and contain catalyst fines which may cause erosion or plugging. Heavy oil cascading down the disc and doughnut (side-to-side on some units) trays cool and condense the heavy vapors so that they can be fractionated. The catalyst fines are washed down to the bottom of the column by this cascading stream.

COKE

Many units have some coke buildup in the reactor, vapor line, and bottom of the main column. The amount in the reactor and vapor line can be minimized by good insulation. The coke buildup in the main column is influenced by three factors: 1.

Hydrocarbon characteristics

2.

Temperature

3.

Residence Time

Some hydrocarbons have a greater tendency to thermally crack and produce coke than others. The Conradson or Ramsbottom carbon residue tests may be used to get some idea of this tendency, but it is difficult to compare one stock with another. The amount of catalytic cracking that has taken place in the reactor will influence the coke production of the hydrocarbons in the main column. The operation of the FCC unit and of the upstream units that produce FCC feed are controlled by other, more important variables than the main column bottoms coke make. Temperature and residence time control are used to minimize the bottoms coke problem.


The maximum allowable temperature in the bottom of the tower is usually given as 700°F (370°C). Experience with a particular feed may raise or lower this somewhat, but it is better to run lower to prevent coke problems. Many refiners use 680-700°F (360-370°C) as the normal operating point. The major control on the bottoms temperature is by the composition of the material, but the quench line may also be used to sub-cool the bottoms material. This line returns cool oil directly to the bottoms level, instead of to the disc and doughnut trays. It is generally used at high throughput, when there is adequate flow to the discs and doughnuts, and the heat input to the column is high. Residence time refers only to the time spent in the column, not in the entire slurry system. The oil begins to cool as soon as it leaves the tower. There may be some minor coking in the slurry settler and associated piping, but usually this is not enough to affect plant operations. If the oil is in the tower too long, at too high a temperature, serious coking can plug the bottom. Removal is difficult, because chipping away with hammers usually leads to lining damage. Other methods, such as chemical cleaning, are usually ineffective. At low charge rates, the bottoms circulation through the exchangers must be reduced to heat balance the tower. The oil cascading down the discs and doughnuts decreases to a point where the trays run dry. Because the hot reactor vapors are not cooled, they can cause warping and distortion of both the disc and doughnut trays and those above as this large volume of vapors tries to pass up the column. Catalyst fines will be carried up into the HCO and LCO circuits, which can be quite serious, because these areas are not designed for fines removal. Product specifications will be adversely affected. The solution to this problem and that of long residence times at low circulation rates is to use the minimum flow line which bypasses the exchangers and returns oil directly to the column, over the disc and doughnut trays. The flow over these trays should be at least 50 gpm/ft tower ID (37.3 m3/hr/meter tower ID). This line is usually used on startup, when the charge rate is low and the tower is cool. As


charge rate increases, the flow rate through the minimum line is decreased to force oil through the heat exchange circuits. There should always be a small amount of oil flowing through the line, so that the oil does not set up. The same is true for the quench line to the bottom of the fractionator. Some units have used a large impingement baffle to break up the reactor vapors as they enter the column. Many of these baffles proved to be a good starting point for coke, and the baffles were removed. Inevitably, there is some coke buildup; pieces may break off with thermal shock or other stresses. If a chunk of coke could enter the bottom line, it could cause plugging or pump damage. A coke trap made of Type 405 or 410 stainless steel is used at the bottom of the column to keep these large chunks out of the line. Smaller pieces of coke pass through it and are caught by the pump suction screens. These can be cleaned on stream, while the larger chunks of coke might prove difficult to remove from the line.

CATALYST CARRYOVER

The amount of catalyst carryover from the reactor will depend on unit design, cyclone efficiency, catalyst type, and unit throughput. Unusual problems such as high reactor level, cyclone failure through cracks or plugged diplegs, or pressure surges can increase the catalyst carryover to unacceptable levels. A conventional unit of older design (single stage cyclones) should lose less than 0.4 LB/bbl charge over to the main column, with most of this returning with the slurry recycle. A new unit with a riser "Tee" and high efficiency cyclones should lose less than 0.05 LB/bbl. It is not practical to actually measure the catalyst content of the main column inlet vapors. The catalyst content of the circulating bottom stream is, however, a good indication of the catalyst losses. A general assumption may be made that very little catalyst is entrained up into the HCO or LCO products. If this is not the case, the bottoms circulation over the disc and doughnut trays should be increased as much as possible, within the limits imposed by other operating variables. The lower part of


the tower should be inspected carefully at the next turnaround to determine if the problem is here. The main column bottoms pump discharge manifold is constructed to direct the catalyst fines and coke particles out of the circulating system. The lines to the various exchangers come off the top of the manifold. The line to the slurry settler will come off the end of the manifold from the bottom. If there is no settler, the reactor recycle and bottoms product lines will both come off the bottom of the manifold. This obviously will not achieve a complete removal of catalyst fines or coke, but it will help prevent a buildup in the main column bottoms circulating system.

SLURRY SETTLER

The slurry settler process flow is shown in Figure 16 in Process Control section. Main column bottoms enter the settler through a tangential nozzle; this gives it a swirling motion that promotes a more even distribution of the heavy oil as it moves up towards the outlet. The catalyst fines settle out and are carried back to the reactor. The carrying medium will depend upon the operation. 1.

Low activity catalyst with large amounts of recycle:

This would be typical of older units that require the high recycle rates to get desired conversion. The recycle consists of main column bottoms and HCO, which is cooler and has a higher flowing specific gravity than the bottoms material. The HCO is injected into the upper diluent point and flows down with some of the bottoms and most of the catalyst that enters the settler. A plant operating in this mode would have a CFR of 1.2 to 2.0.


2.

Higher activity catalyst with little or no recycle:

The use of the new higher activity catalysts and better reactor design enable the refiner to crack most of the feed on the first try. The heavy oil product make is lower; it is also a more refractory material which tends to go to coke and dry gas when recycled to the reactor. The carrying material used in this case may be HCO, but for many units, raw oil is the better choice. The bottom diluent injection point is used to minimize the amount of raw oil that will go up into the settler. If the raw oil thermally cracks or goes out with the CSO, it may cause problems with the CSO product specifications. To help prevent this, the flow back to the reactor should always be higher than the diluent flow to the settler.

SETTLER OVERHEAD

The amount of CSO that comes off the top of the settler will also depend upon the operation. The decreased heavy oil make of the newer units can lead to a buildup of catalyst in the main column because there is less main column bottoms leaving the tower to take it out. This refers to oil leaving the system, not to the bottoms circulation streams. The concentration of fines in the tower may build up enough to cause serious erosion and plugging problems. The most effective solution is to use a return line from the top of the settler to the main column disc and doughnut trays. The flow to the settler can then be increased by the amount of oil which is returned to the main column relatively free of catalyst. Typical settler flow rates are shown in Table 11. For this table, it was assumed that catalyst is coming overhead from the reactor at a rate of 100 LB/day. The catalyst can only leave the system through the outlet to the settler; therefore, the concentration will be equal to the amount of catalyst (100 LB/day), divided by the flow to the settler.


TABLE 11 SLURRY SETTLER FLOW CASE A

CASE B

CASE C

10,000

10,000

10,000

Main Column Bottoms To Slurry Settler, BPD

1500

500

1500

Main Column Bottoms Catalyst Concentration, lb/bbl

0.067

0.20

0.067

500

500

500

CSO Product, BPD

1,000

500

500

Recycle to Reactor, BPD

1,000

500

500

0

0

1,000

Feed to FCC, BPD

Diluent, BPD

CSO Return to Main Column, BPD

Case A would be an operation with lower activity catalyst and a higher CSO product make. Case B would be a conversion to higher activity operation, decreasing the amount of recycle and the CSO product. Neither Case A or B uses a return line from the top of the settler to the main column. Case C is similar to B in that a higher activity catalyst is used, with a small amount of main column bottoms produced. The use of the return line from the top of the settler back to the main column allows the flow to the settler to be increased to 1500 BPD. A limit on flow to the settler is the velocity of fines settling. The liquid velocity should never exceed 30 BPD/ft2 (50 m3/d/m2) of settler cross sectional area. Above this, there may be problems with catalyst carry over into the overhead stream.


FCC UNITS WITHOUT SLURRY SETTLER

Some of the new units have incorporated a reactor cyclone configuration which decreases the amount of catalyst carryover to the main column. The reactor riser ends with a pair of outlet arms, and a two stage cyclone system on the reactor outlet eliminates most of the catalyst in the overhead vapors. With this system, some refiners have elected not to use a slurry settler. The recycle to the reactor and the bottoms product come directly from the main column bottoms pump discharge manifold.

PLUGGING

There are occasional problems with plugging in the lines or exchangers of catalyst bearing streams. If there is a plant upset which causes large amounts of catalyst to go overhead, such as a sudden dip in column pressure, immediate action should be taken to remove the extra catalyst from the system. Increased slurry recycle and clarified oil product would be the two most important steps. These should be continued until the laboratory confirms that the BS & W content of the bottoms stream is back to normal. In the event of a major upset that completely fills the bottoms of the column with catalyst, care must be taken to avoid further complications. Catalyst holds heat fairly well and conducts it poorly, so the cool-down period may be on the order of several days. The thermocouples in the column will be well insulated by the catalyst close to the wall, so the readings will probably be lower than the actual temperature of most of the catalyst. Introduction of cold oil or wash water will produce severe pressure surges that may damage the column internals. The oil soaked catalyst is both a fire and breathing hazard. The shutdown for cleanout can be estimated at one week, a good argument against hasty measures that could lead to excessive catalyst carryover to the column.


MAIN COLUMN BOTTOMS EXCHANGERS

To prevent catalyst plugging or erosion in the exchangers, UOP calls for the following velocities in the tubes: 1.

Straight tubes: maximum velocity 8.0 ft/sec, minimum 4.00 ft/s.

2.

U-tubes: maximum velocity 8.00 ft/sec, minimum 4.00 ft/s.

In general, the optimum velocity is 8.00 ft/s. Straight tube construction is recommended. It is important to think of these numbers when changing raw oil charge or exchanger flow rates. Plugged exchangers are difficult to clean. A catalyst bearing stream is never routed through the shell side of an exchanger because the catalyst fines will settle to the bottom of the exchanger. There will be a progressive loss of heat transfer area as more and more tubes are covered by the fines.

MAIN COLUMN BOTTOMS HEAT REMOVAL

The main column bottoms circulation rate is adjusted to control the column's heat removal requirement. The bottoms stream generally exchanges heat with the raw oil feed and is utilized in the production of superheated steam in the steam generators. A reduction or increase in the bottoms heat removal must be compensated by an increase or reduction in heat removal in another section. Provided no other changes are made, the overhead reflux rate will compensate for any changes in bottoms heat removal. As illustrated in Table 12, a decrease in the bottoms stream heat removal results in an increase in reflux rate. The bottoms stream heat removal should be adjusted to minimize the reflux rate and maintain good product distillations.


TABLE 12

Feed Rate, BPD

17550

17550

Overhead reflux rate, BPD Net ovhd light gas yield, MMSCFH Net ovhd liquid yield, BPD Ovhd heat removal, MM-BTU/hr

10488 13.81 7973 42.81

11581 13.81 7993 45.24

Net heavy naphtha yield, BPD Circ. heavy naphtha heat removal MM-Btu/HR

2782

2782

12.32

12.32

1790

1790

10.17

10.17

967

967

31.0

28.57

Net LCO yield, BPD Circ. LCO heat removal, MM-BTU/hr Net bottoms yield, BPD Circ. bottoms heat removal, MM-BTU/hr

GASOLINE/DISTILLATE PRODUCTION

The main column draw temperatures are dependent on the stream's composition and vary with changes in draw rate. As the gasoline product draw is reduced, liquid will drop down the column to the LCO draw tray and require an increase in LCO product draw. The reduced gasoline draw rate will result in a lighter gasoline product having a lower ASTM distillation end point and a lower draw tray temperature. The LCO product also becomes lighter because of light material dropping to the LCO draw tray. The LCO initial boiling point temperature will decrease with a slight decrease in the draw tray temperature. Table 13 illustrates the changes which result to the main column and product streams as the LCO yield is increased by reducing the gasoline endpoint and yield.


TABLE 13

Gasoline: API Product rate, BPD 90% BP temp., °F RONC MC overhead temp., °F

57.9 15800 372 90.7 311

57.2 16200 381 90.6 316

56.3 16600 392 90.4 323

Light cycle oil API Product rate, BPD Flash point, F 10% BP temp., °F 90% BP temp., °F End point, F MC draw temp., °F

20.1 6000 198 462 624 664 482

19.8 5600 204 471 624 664 485

19.3 5200 212 482 624 664 499

Slurry °API Product rate, BPD

6.2 2800

6.2 2800

6.2 2800

GASOLINE CUT PROPERTIES

Gasoline or any other liquid stream can be broken up into numerous cuts, each having distinct properties. Examining the cuts which comprise a typical gasoline sample will show how overall product quality can be improved.


A FCCU gasoline sample from a high severity operation was broken up into nine cuts, each having a narrow TBP range as illustrated in Table 14. The properties of the individual cuts are shown in Figures 6 and 7. TABLE 14 Cut No.

1 2 3 4 5 6 7 8 9

Cumulative Vol -% 20 30 40 50 60 70 80 90 100

TBP Fraction, °F 75 - 93 93 - 145 145 - 181 181 - 210 210 - 250 250 - 286 250 - 286 286 - 334 334 - 387 387+


FIGURE 15: FCC GASOLINE CUT PROPERTIES Avg. BoilingPoint Temp, °F

400

300

200

100

0 1

2

3

4

5

6

7

8

9

6

7

8

9

6

7

8

9

Gasoline Cut 100

API Gravity

80

60

40

20 1

2

3

4

5

Gasoline Cut 100

RONC

96

92

88

84

80 1

2

3

4

5

Gasoline Cut


FIGURE 16: FCCU GASOLINE CUT PROPERTIES 1.2 1

Sulfur, wt %

0.8 0.6 0.4 0.2 0 1

2

3

4

5 Gasoline Cut

6

7

8

9

5

6

7

8

9

100

Bromine

Number

90 80 70 60 50 40 30 20 10 0 1

2

80

3

4

Gasoline Cut

Paraffin/Naptha

70

Aromatic

Olefin

Liquid Vol %

60 50 40 30 20 10 0 1

2

3

4

5

Gasoline Cut

6

7

8

9


The sulfur content of gasoline increases sharply in the last 387+ TBP fraction as indicated in the Wt-% SULFUR graph. The overall sulfur content of gasoline hence could be reduced by lowering the gasoline end point temperature. The RONC graph of the various cuts indicates a reduction in octane at CUT 4 (181-210 TBP) and CUT 9 (387+ TBP). A slight increase in the overall gasoline octane can be obtained by reducing the end point temperature of the gasoline. If it were possible to remove CUT 4, gasoline octane could be further increased.


ROUTINE PROCESS VARIABLE CONTROL REACTOR REGENERATOR SECTION

This tabulation of process conditions is intended to assist the operator in selecting the optimum operating conditions for different operations. It may be noted here that process units rarely operate at their design conditions. Variable

Operating Conditions

Raw oil charge rate

As desired.

Raw oil temperature

To balance coke yield, conversion, and RON requirements.

Slurry recycle rate

Normally at minimum.

HCO or raw oil to slurry settler

Equal to or slightly less than total flow of slurry recycle to reactor, or until clarified oil gravity changes.

Heavy recycle rate

Heavy recycle rate can be varied to adjust conversion product yields or increase coke yield.

Combined feed temperature

Normally as high as possible provided neither reactor nor regenerator temperature is excessive.

Reactor temperature

As necessary to obtain desired conversion and RON.

Reactor pressure

Equals fractionation column receiver pressure plus fractionation pressure drop.

Reactor level (Riser cracking operation)

To cover top stripping grid.

Reactor level (Bed cracking operation)

Minimum level needed to achieve desired conversion.

Emergency steam to riser

Used to initiate catalyst circulation on startup and to avoid plugging the riser on an emergency shutdown. Normally not used.


Variable


LCO or gasoline to riser

Used to control regenerator temperature when the unit is "behind in burning".

Stripping steam

Just enough to strip catalyst. This value can be arrived at by observing the effect of decreasing the stripping steam on regenerator temperature. 1.5-2 lb/1000 lb catalyst circulation is typical.

Steam to feed nozzle

Normally adjusted to maximize feed distributor pressure drop within feed pump hydraulic constraint. Usually 1-2 wt% of feed.

Regenerator air rate

As necessary to control regenerator temperature spread or to give good control using automatic snort. On total CO burning units to obtain 1-4% oxygen in flue gas.

Regenerator dense phase temperature

Normally cracking conditions are varied to optimize regenerator dense phase temperature. Too cold and catalyst will not be well regenerated. Too hot and reaction with oil will be thermal with resultant loss of gasoline and increase in gas.

Regenerator dilute phase or flue gas temperature (on conventional partial CO burning units)

The difference between regenerator dense, dilute and flue gas temperatures is an indication of the amount of excess oxygen present, and is the criterion by which the air rate is varied.

Regenerator pressure (conventional)

Equals the reactor pressure plus the reactor regenerator differential pressure.

Regenerator pressure (with flue gas power recovery)

Regenerator pressure controlled, reactor-regenerator differential allowed to swing within reasonable limits.

Regenerator level

The catalyst level can vary from about 35" to 80" of H2O with the unit inventory and regenerator velocity.

Reactor-regenerator differential pressure

Varied to obtain stable slide valve differentials and minimum utility consumption.


Variable


Slide valve differential pressure

Dependent on vessel differential and catalyst condition. Over-rides are normally set at 1-2 psi.

Steam to spray and torch nozzles

Flow will be almost zero except when sprays are in service.

Torch oil rate

Used on startup and to control afterburning on partial CO combustion units.

Torch steam pressure

Only used when necessary to atomize torch oil and normally 5-10 psi higher than regenerator pressure.


TABLE 15 MAIN COLUMN Variable


Main column bottoms temperature

Adjust clarified oil yield and quench to hold long enough to prevent coking.

Circulating slurry rate

As necessary to control fractionating column heat removal requirements. To minimize top reflux to obtain minimum gap (25-40°F) on gasoline between 90% and E.P.

Slurry return temperature

Of little interest. Dependent on cleanliness of exchangers, number in service and circulation rate.

Clarified slurry yield, or main column bottoms yield if no slurry settler

As necessary to control fractionating column bottoms temperature and level.

Main column bottoms BS&W

Adjust flow to slurry settler, clarified oil return to main column, and slurry recycle. If no slurry settler, adjust bottoms recycle and bottoms product.

Heavy recycle oil circulation

Rate is set as desired to transfer heat to various reboilers.

Heavy recycle deck temperature

Depends on distillation range on heavy recycle and on tower pressure.

Light cycle oil yield

Depends on charge rate and conversion, and is varied to maintain desired properties of light cycle oil. Also used to control bottoms level.

Flush oil

As required to keep catalyst out of instruments. Normally 1500-2000 BPD, but this will vary between different units.


Variable


Light cycle oil stripping steam

Enough to meet flash point specifications.

Light cycle oil deck temperature

Depends on distillation range of light cat gas oil and on tower pressure.

Unstabilized gasoline top reflux rate

Is varied to control tower top temperature and depends on amount of heat removed lower in column.

Circulating top reflux temperature

Depends on water temperature and flow rates.

Fractionator column top temperature

Varied to control endpoint of unstabilized gasoline.

Overhead receiver pressure

Can be varied as discussed.

Overhead receiver temperature

Should always be as cold as is economically possible.

Unstabilized gasoline yield

Depends on charge rate and conversion.

Wet gas molecular weight

Minor variations in density will be due to changes in receiver conditions, but major changes will be due to increased hydrogen production. At low densities, the compressor will have an increased tendency to surge.

Wet gas flow

Dependent on compressor speed but must be adequate to handle production plus spillback for control. Must be above minimum to keep out of surge.


TABLE 16 GAS CONCENTRATION SECTION Variable


Wet gas compressor Variable speed

Run at minimum governor until spillbacks close.

Fixed speed centrifugal

Butterfly valve opens away from limiting stop as spillbacks close.

Fixed speed reciprocating

On spillback control.

Wet gas spillbacks High pressure separator Temperature Pressure Primary absorber Top and intercoolers temperatures

Vary as needed to hold main column overhead receiver and interstage pressure. Used to keep centrifugal machine out of surge. 80°F-100°F (27°-38°C) Rides on primary absorber backpressure. Less than 100°F (38°C)

Intercooler flow rates

As needed for good absorption efficiency.

Pressure

Rides on sponge absorber backpressure.

Sponge absorber Top temperature

As cool as economically practical.

Lean oil flow rate

As needed for good absorption efficiency. Do not flood tower.

Pressure

Controlled to give good absorption efficiency and hold correct backpressure on HPS.


Variable


Stripper Overhead vapor flow rate

Controlled by heat input to column to give sufficient C2- and H2S stripping.

Heat input to column

Controlled to give proper overhead vapor rates.

Pressure

Rides on high pressure separator backpressure.

Debutanizer Top temperature

Controls reflux to give desired RVP of gasoline.

Top reflux temperature

Should be as cold as economically possible.

Reboiler heat

As required to give good fractionation.

Pressure

Varied as needed for good fractionation, but must be high enough to allow condensation of C3-C4 stream by air fin fans or cooling water.

157048 Process Calculations Page 1

PROCESS CALCULATIONS INTRODUCTION Throughout the years the Fluid Catalytic Cracking process has been a very versatile and flexible tool for the refiner, and has become the basic conversion step in the modern refinery. This process has survived and prospered because of its ability to handle the many changes in catalyst, operating conditions, and feedstocks that have occurred over the years. The FCC Unit produces large volumes of high octane gasoline, olefinic LPG, fuel oil (LCO and MCB), fuel gas, steam, and electricity. The yields are mainly determined by process variables (i.e. feedstock, operating conditions, mechanical features, and type of catalyst). Process variables have varying degrees of interdependence and may change frequently producing changes in the yield structure of the products. A performance test conducted at least once a week is recommended to evaluate the effect of process variables on yield. The tests can be used to chart a history of the unit and to find conclusions at different operating conditions. The performance test provides accurate yield structures at a particular set of operating conditions and provides a base point for further testing. The Performance Test normally includes a heat balance, material balance, and a pressure survey. In those cases where more information is desired, a Mechanical Evaluation Test is recommended. The refiner can use this test to assess the potential of the unit and determine possible bottlenecks. This section explains how to accomplish an acceptable Heat and Material balance and how to do some of the most important calculations in the FCC Unit. An FCC Performance Test Procedure is available on request from the Technical Service Department. This procedure explains in detail how to conduct a performance test in the FCC Unit.


MATERIAL BALANCE A material balance on an FCC Unit is done by drawing an envelope around the unit in a manner that flow rates are known for all streams. This envelope includes the Reactor, Regenerator, Main Column, and Gas Concentration sections. Normally, the Gas Concentration Unit includes the primary absorber, sponge absorber, stripper column, and debutanizer column.

1.

Data

Flow rates, flowing temperatures and laboratory analyses are required for each stream. Pressure is also needed for the gas streams. The following table shows the information needed to do a material balance: INPUT DATA FOR HEAT AND MATERIAL BALANCE Stream

Flow Temp. Pressure

API

Distillation

Feed

Yes

Yes

Yes

D-1160

Air*

Yes

Yes

MCB

Yes

Yes

Yes

D-1160

Yes

LCO

Yes

Yes

Yes

D-86

Yes

Gasoline

Yes

Yes

Yes

D-86

LPG

Yes

Yes

Sponge Gas

Yes

Yes

Meter Factor Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Flue Gas Notes

GC

*Ambient temperature and relative humidity of air are needed. Reactor, Regenerator, and combined feed temperature are needed for Heat Balance

The method for including extraneous streams is straight forward as long as flow rates and analyses are known. These additional streams coming into the Main Column and Gas Concentration Unit are subtracted from the product streams.


2.

Liquid Streams

Calculate corrected liquid flow rates using the following equation: Q = K R (Gf)1/2 / Gb Where:

Q = flow rate K = flow meter constant ("K" factor) R = chart reading Gb = base gravity @ 60°F Gf = gravity at flowing temperature

The flowing gravity Gf is calculated using the following equation: Gf = Gb x VCF Where:

VCF = Volume correction factor VCF = EXP [ (-ßo) ∆T(1+0.8 ßo ∆T) ]

0.9545

Where: Go = density at 60°F in kg/m3 T = observed temperature in °F ∆T= T - 60 ßo = coefficient of thermal expansion at 60°F, (1/°F)

926 173 113

The set of correlations for the coefficient of thermal expansion based on API Data Tables are: ßo = (Ko + K1 Go + K2 Go² )/ Go² Where:

0.00039779


Product Crude Oils

°API Range 0 – 100

Ko

K1

K2

341.0957

0

0

0.2438

0

Gasoline

52 – 85

192.4571

Gasoline/Jet

48 – 52

1489.0670

0

-0.0018684

Jet Fuels

37 – 48

330.3010

0

0

Fuel Oils

0 – 37

103.8720

0.2701

0

0.3488

0

Lubricating Oils

-10 – 45

0

For FCC Raw Oil Feedstock (VGO); Ko = 341.0957, K1 = 0, K2 = 0 The following rounding is applied to the input and output of all routines. Temperature: Density: VCF:

0.1°F 0.1 °API or 0.5 kg/m3 five significant figures for computation

Example: Raw Oil Charge K = 3,380 in BPSD R = 8.9 Gb = 0.9260 VCF = 0.9545 Gf = 0.9260 x 0.9545 = 0.8839 Q = 3,380 x 8.9 x (0.8839)1/2 / 0.9260= 30,542 BPSD Q = (BPSD) x 5.614583 ft3/bbl x (Gb x 62.3635 lb/ft3 H2O)/24 hr/day Q = 412,615 lb/hr Similar calculations are done for all C5+ liquid streams. LPG streams are treated as follows:


Estimation of VCF for LPG The following equation approximates VCFs from API Tables #33 and #34:





VCF   10  A*Gb B   T  60   1.0 Where: T Gb A B

= = = =

Flowing Temperature, °F Specific Gravity @ 60°F, in g/ml 2.64641798 1.40583481


3.

Gas Streams

Calculate corrected gas and air flow rates as follows: Q = K x R x (Pf/(Tf x SG))1/2 Where:

Q = flow rate K = flow meter constant ("K" factor) R = flow meter reading Pf = Pressure at flowing conditions (absolute) Tf = Temperature at flowing conditions (absolute) SG = specific gravity of gas = MWgas/MWair (1.0 for air)

Example: Sponge Gas Q = K x R x [(Pf/(Tf x SG))]1/2 Where:

K = 91,848 scfh R = 6.5 chart reading Pf = 173 psig + 14.7 = 187.7 psia Tf = 113°F + 460 = 573°R SG = 0.7054 from Sponge gas calculations MW = 18.9 from Sponge gas calculations

Q = 91,848 x 6.5 x [(187.7)/(573 x 0.7054)]1/2 = 406,837 scfh M = Mass Flow = scfh x MW/379.67 = 20,261 lb/hr Where:

379.67 is the conversion factor for scf / mol

Similar calculations are done for all gas, vapor, and air streams.


4.

Calculate Coke

The coke make is calculated from the Heat Balance. (Refer to Heat Balance calculation section.)

5.

Calculate the as Produced Yields

A product yield is defined as the product rate divided by the raw oil rate. The volume percent of each product stream is: Vol-% (A) = (A, bpsd)100/Fresh Feed, bpsd The weight percent is: Wt-% (A) = (A, lb/hr)100/Fresh Feed, lb/hr Where:

A = any product stream


6.

Weight and Liquid Volume Recoveries

Once the weight and the volume flows are known for each stream, the Weight recovery and the Liquid volume recovery can be calculated. Proper data analysis requires that the Weight recovery must be 100.0  2.0 wt-%. Errors outside this range are significant and cast doubts on the validity of the test data. The Sponge Gas used to calculate the Weight recovery should not include the inert gases (N2, O2, CO, CO2). In addition, the as-produced Liquid Volume recovery does not include the C3+ from the sponge gas. Weight % Recovery = (Products, lb/hr x 100)/(Fresh Feed + Extraneous Feeds) lb/hr Liquid Volume % Recovery = (Products, bpsd x 100)/(Fresh Feed + Extraneous Feeds)bpsd

7.

Conversion

Conversion is defined as the volume percentage of raw oil converted to gasoline and lighter components. This is calculated as: Conversion, Vol% =

Feed - LCO - HCO - MCB  100 Feed

This conversion is called ‘as-produced’ or ‘apparent’ conversion because is not corrected for cut-points. The ‘corrected’ or ‘true’ conversion is calculated using the same equation after the gasoline and LCO yields are corrected for cut-points. 8.

Gasoline and LCO Yield Adjustments

It is important to correct the gasoline and LCO yields on a constant cut-point basis. It is inaccurate to compare gasoline yields at different 90% or EP temperatures. The gasoline yield must also be adjusted by removing the C4's and adding the C5's and C6's from the LPG and Sponge Gas streams. The procedures to adjust the liquid yields and to calculate the C4's in the gasoline are attached in this section.


9.

Gasoline Selectivity

The gasoline selectivity is the corrected gasoline yield divided by the true conversion: Gasoline Selectivity =

10.

(Corrected Gasoline Yield)  100 True Conversion

LPG and Sponge Gas Calculations

These procedures use the mol percentage from the GC analysis, the molecular weight, and the specific gravity to calculate the flow rate of each component. Also, the stream specific gravity and molecular weight are calculated. The procedure is attached in this section.

11.

C3 and C4 Recovery

The C3 and C4 recovery indicate how the Gas concentration is performing. The C3 recovery is calculated as:

C3 Recovery, Vol% =

C3 in LPG  100 (C3 in LPG + C3 in Fuel Gas)

The C4 recovery is calculated as: C4 Recovery, Vol% =

C4 in LPG  100 (C4 in LPG  C4 in Fuel Gas  C4 in Gasoline)


FCC Unit Material Balance Refiner: _______________________________

Location: _______________________

TI - Tag

FE - Tag

Gravity

Vol.Corr.

Flowing

Meter

Temp. °F

Readings

Gb@60°F

Factor

Gravity Gf

"K"

1

Fresh Feed (FF)

TI—10

2

Main Col Bottoms (MCB)

TI-20

lb/Hr

Vol%

Wt%

8.90

0.9260

0.9545

0.8839

3,380

30,538

412,601

4.80

1.0412

0.8806

0.9169

418

1,843

28,005

6.04

6.79

9.00

0.9200

0.9838

0.9051

700

6,515

87,450

21.33

21.19

7.60

0.7599

0.9489

0.7211

2,118

17,987

199,434

58.90

48.34

7.20

0.5612

0.9653

0.5418

632

5,964

48,833

19.53

11.84

0

0.00

0.00

FRC-20 462

TI-30

Light Cycle Oil (LCO)

BPSD

FRC-10 173

3

Date: __________

FC-30 101

4

Gasoline (DebBt)

TI-40

FRC-40

5

LPG

TI-50

6

LPG (ELPG) Extraneous Feed

TI-

7

Coke

See attached Heat Balance calculation sheet.

148 FRC-50 87 FRC-

0

25,187

6.10

6,589

18,821

4.56

6,996

20,887

0

0

PE-Tag Press, psig 8

TI-60

Sponge Gas (SGas)

FC-60 113

8

Gas (HGas) Extraneous Feed

TI-

9a

Air to Regenerator (Dry Back)

TI-70

9b

9

1.00

5.00

1.00

W/o Inerts

0.00

44.5

46,000

81,721

374,251

56

325

520

2,382

56

635

1,260

5,771

83,502

382,405

PI-72

FIC-74 230

1,518

PI-70 6.83

FIC-72

TI-74

Air to Distributor 2

W/o Inerts 173

PRC-

FIC-70

230 9c

0.6519

FC-

TI-72

Air to Cat. Cooler

PRC-60 6.50

415

SCFM

PI-74 6.20

1.00

Total Air to Regen (Wet Basis)

As produced Calculations

10

Weight Rec Inert Free

= [MCB lb/hr+LCO lb/hr+DebBt lb/hr+LPG lb/hr+SGas lb/hr+Coke lb/hr]*100/ [FF lb/hr+ELPG lb/hr+EGas lb/hr] =

=

98.82

Wt%

11

Liquid Vol. Recovery

= [MCB bpsd+LCO bpsd+DebBt bpsd+LPG bpsd]*100/[FF bpsd+ELPG bpsd]

=

105.8

Vol%

12

Conversion

= [FF bpsd-MCB bpsd-LCO bpsd]*100/[FF bpsd]

=

72.6

Vol%

For Liquids

:

BPSD = Units K [SQRT(Gf)]/Gb

:

lb/hr = [BPSD (Gb) (5.6146ft3/bbl) (62.3689lb/ft3)]/(24h/d) = (14.591) ( BPSD) (SG)

For Gases or Air

:

SCFM = K Units SQRT{psia/(°R*Gb)}

LV%

= (stream, bpsd)(100)/(FF, bpsd)

lb/hr = (scfm MW 60min/hr)/379.5 Wt% = (stream, lb/hr)(100)/(FF, lb/hr)

uop1292rc


FCC Unit Material Balance Continuation Refiner: _______________________________

Location: ____________________

Yields Adjustment for Gasoline @ 380°F - 90% and LCO @600°F - 90%. IBP, °F 10%, °F 70%, °F 90%, °F EP, °F Gasoline LCO

319

460

MCB

524

710

569

Date: _____________

°API

BPSD

lb/hr

367

410

54.71

17,987

618

653

22.30

6,515

87,450

4.40

1,843

28,005

26,346

314,889

Totals

199,434

Gasoline Gasoline Factor (F1)

= [(380-T90)/(EP-T90)](1/9)+1

= [ ( 380 – 367 )/(410-367)]

(1/9) + 1=

LCO Factor (F2)

= [(380-IBP)/(T10-IBP)](1/9)

=

(1/9) =

cGasoline, BPSD

= (Gasoline, bpsd)(F1) + (LCO,bpsd)(F2)

[ ( 380 - 319 )/(460-319)]

=

18,587

1.034 F1 0.05 F2

cBPSD

60.9

cLV%

c °API

= API + 37.5(LV% - cLV%) / LV%

=

52.13

c °API

0.771

cSG

cGasoline, lb/hr

= cBPSD x cSG x 14.591

=

208,974

c lb/hr

50.6

cWt%

lb/hr

BPSD

C4's in Gasoline

=

3,390

380

C5's + C6's in LPG

=

423

46

C5's + C6's in SGas

=

248

26

C5's + C6's in Extr Feeds

=

0

0

Gasoline Yield Adjustment for C4's, C5's, & C6's.

13 Corrected Gasoline, BPSD

= cGasoline-C4's+C5's+C6's Extr(C5+C6)

=

18,587

corrBPSD

60.9

corrLV%

14 Corrected Gasoline, lb/hr

= cGasoline-C4's+C5's+C6's-Extr(C5+C6)

=

208,974

corrlb/hr

50.6

corrWt%

15 Corrected °API

= ((141.5*BPSD*14.591)/(lb/h*24))-131.5

=

52.13

corr°API

LCO LCO Factor (F3)

= [(600-T70)/(T90-T70)](0.2)+0.7 =

[ (600-569)/ (618-569)] (0.2) +0.7 =

0.827 F3

MCB Factor (F4)

= [( 600-IBP )/( T10-IBP )]( 0.1 ) =

0.041 F4

Gasoline Factor (F5)

= Gasoline - cGasoline

[ ( 600 524 ( 17,680) -

=

16 Corrected LCO, BPSD

= [(LCO bpsd)F3 + (MCB bpsd)F4 + F5]/0.9

17 Corrected LCO °API 18 Corrected LCO, lb/hr

)/ (710-524)]

(0.1) =

(18,587) =

-907 F5

=

5,059

corrBPSD

16.6

corrLV%

= API + 5( LV%-corrLV% )/LV%

=

23.42

corr°API

0.913

corrSG

= corrBPSD x corrSG x 14.591

=

67,416

corrlb/hr

16.3

corrWt%

=

2,392

corrBPSD

7.8

corrLV%

=

35,780

corrlb/hr

8.7

corrWt%

MCB 19 Corrected MCB, BPSD

= Total C5+ Liq. Yield - Corr Gasoline - Corr LCO = = 17,689+6,515 +1,843-18,587 -5,059

20 Corrected MCB, lb/hr

= Total C5+ Wt. Yield - Corr Gasoline - Corr LCO = = 196,715+

87,450+

28,005-208, 974-67, 416

21 Corrected °API

= [( 141.5 * BPSD * 5.6146 * 62.3689 )/( lb/h * 24 )] - 131.5

=

6.5

corr°API

22

= [FF bpsd - corrMCB bpsd - corrLCO bpsd]100/[FF bpsd]

=

75.6

corrVol%

True Conversion

*Note that Total C5+ yield = cGasoline + LCO + MCB

API = (141.5/SG) - 131.5

Note- The Factor equations are valid only if the as produce 90% temperatures are between 360-400°F for gasoline and 580-620°F for LCO. The Factor Equations were developed from the 90% plus 10% Method. Used this method if the 90% Temp. are not in the specified range.


Sponge Gas Calculations Example Refiner: ______________________________Location: _____________________________ Date: _______________ A

B

C=AxB

D

E

g/100mol

Ib/hr

sp gr

0.04 0.00 1.47 4.31 5.82

1.3 0.0 64.7 120.7 186.7

14 0 716 1,335 2065

28.29 2.57 28.21 15.83 15.55 0.48 1.48 0.59 0.25 0.24 0.25 0.12 0.06 0.00 0.00 0.00 0.00 0.26 94.18 100.00

57.0 87.6 452.6 476.0 436.2 21.2 62.3 34.3 14.5 13.5 14.0 6.7 3.4 0.0 0.0 0.0 0.0 22.4 1,702 1,888

Component

MW

mole%

O2 CO CO2 N2 Total Inertes

32.00 28.01 44.01 28.01 (t)

H2 H2S C1 C2 C2= C3 C3= iC4 nC4 1-C4= i-C4= t-C4= c-C4= 1,3-C4== i-C5 n-C5 C5= C6+ Total Products TOTAL

2.02 34.08 16.04 30.07 28.05 44.10 42.08 58.12 58.12 56.11 56.11 56.11 56.11 54.09 72.15 72.15 70.14 86.18 (T)

MW = TC/TB = 18.9 SG = MW/28.966 = 0.6519 Mole % Inerts=tB= 5.82 Wt% Inerts=100tD/TD= 9.89 Inerts = tD = 2,065 Ib/hr

F bpsd (scfh)

(168) (0) (6,170) (18,091) (24,429)

631 969 5,006 5,265 4,825 234 689 379 161 149 155 74 37 0 0 0 0 248 18,821 20,887

0.5077 0.5220 0.5631 0.5844 0.6013 0.6004 0.6100 0.6271 0.6272 0.6248 0.6312 0.6496 0.6640

Pres. Psig = 173 Temp. F = 113 Meter Units = 6.5 Meter K = 91,100

(118,745) (10,787) (118,409) (66,445) (65,270) 31.6 90.4 46.2 18.8 17.0 17.7 8.4 4.1 0.0 0.0 0.0 0.0 25.6 259.8 (404,086)

=14.7=psia + 460=R

187.7 573.0

Gas with Inerts: V = Vol. Flow = KxUnitsxSQRT{psia/(R*SG)} = 419,742 scfh M = Mass Flow = scfh x MW/379.5 = 20,887 Ib/hr Inert-free Gas:

Mass = (M – Inerts) = M(100-wt%Inerts)/100 = Vol. Flow=(V)-(tF)=V-(V x M%Inerts/Tmoles) =

bpsd = [Ib/hr x24]/[ SGx5.614583ft3/bblx62.3689lb/ft3]

18,821 Ib/hr 395,313 scfh

scfh = V x Mol%/TB

D = C x M/TC = C x Total Mass Flow/(Total gr/100 mol)


Debutanizer Overhead (LPG) Calculation Example: Refiner: ______________________________Location: _____________________________ Date: _______________ A Component

MW

B mole%

C=AxE

D

g/100mol

SG

E=C/D

F=MxCftC G=VxE/tE

cc/lOOmol

lb/hr

H2S

34.08

0.0

0.0

0.7871

0

0

C2

30.07

0.0

0.0

0.3563

0

0

C2=

28.05

0.0

0.0

0.3680

0

0

C3

44.10

12.1

533.6

0.5077

1,051

5,216

704

C3=

42.08

36.2

1523.3

0.5220

2,918

14,891

1,955

iC4

58.12

10.7

621.9

0.5631

1,104

6,080

740

nC4

58.12

3.8

220.9

0.5844

378

2,159

253

1-C4=

56.11

8.5

476.9

0.6013

793

4,662

531

i-C4=

56.11

12.4

695.7

0.6004

1,159

6,801

776

t-C4=

56.11

9.2

516.2

0.6100

846

5,046

567

c-C4=

56.17

6.2

347.9

0.6271

555

3,401

372

1,3-C4==

54.09

0.3

16.2

0.6272

26

159

17

i-C5

72.15

0.4

28.9

0.6248

46

282

31

n-C5

72.15

0.2

14.4

0.6312

23

141

15

C5=

70.14

0.0

0.0

0.6496

0

0

0

C6+

86.18

0.0

0.0

0.6640

100.0

4,996

TOTAL (t)

0

0

8,899

48,838

bpsd

0 5,962

LPG S.G. =

tC/tE = 0.5614

Temperature, 0F

=

87

LPG MW =

tC/tB = 50.0

Vol. Corr. Factor

=

0.9653

Flow Gravity, Gf

=

0.5419

Meter Units Meter K

= =

7.2 631.5

V=Vol.Flow=Units x K x [SQRT(Gf)]/SG

=

5,962 bpsd

M = Mass Flow = (BPSD x SG x 5.6146ft3/bb1 x 62.3689lb/ft3)/24hr/d = =

48,838 lbs/hr


Propane and Butane Recovery Calculation Example: (1) (2) Sponge Gas Component C3 C3= Total C3’s (t)

bpsd 32 90 122

234 689 923

bpsd 704 1,956 2,660

lb/hr 5,215 14,888 20,103

(5) (6) Stabilized Gasoline (Debut. Bottoms) bpsd lb/hr 0 0 0 0 0 0

iC4 nC4 1-C4= i-C4= t-C4= c-C4= 1,3-C4= = Total C4’s (T)

46 19 17 18 8 4 0 112

379 161 149 155 74 37 0 956

740 253 531 776 567 372 17 3,257

6,078 2,159 4,661 6,800 5,045 3,400 159 28,301

0 47 34 34 112 44 109 380

lb/hr

(3) (4) Debutanizer Overhead

0 399 299 299 997 399 997 3,390

C3 Recovery = t(4) * 100 / [t(2) + t(4)] =

95.6

wt-%

C4 Recovery = [T(4) + T(6)] * 100 / [T(2) + T(4) + T(6)] =

97.1

wt-%


Yields Adjustment Composite 90% Plus 10% Method

This method uses the ASTM distillation of the liquid products to create a composite curve and correct the yields to any specified 90% temperatures. Procedure

1. Using the ASTM distillation and a straight line interpolation equation, calculate the LV% distilled every 20F for all the product streams. %x=[(Tx-Ta)/(Tb-Ta)](%b-%a)+%a

 Straight line interpolation equation

%a < %x < %b Ta < Tx < Tb 2. Calculate the composite volume and percent every 20F using the following equations: Cumulative BPD @ Tx = BPD = [(%Gasoline) (BPSD Gasoline) + (% LCO) (bpsd LCO) + (% MCB) (bpsd MCB)}/100 Cumulative LV% @ Tx = 100 BPD/Total BPD 3. Calculate Corrected Gasoline 90% @ 380F or Specified 90% Temperature: Corrected Gasoline (Composite Yield @ 380F)/(0.9) 4. Calculate Corrected LCO 90% @ 600F or Specified 90% Temperature: Corrected LCO = (Composite Yield @ 600F)-(Corr Gasoline)/(0.9) 5. Calculate Corrected MCB by difference: Corrected MCB = Total Liquid Yield – Correct Gasoline – Correct LCO


Data: (%) IBP 10% 30% 50% 70% 90% EP BPSD

Gasoline (°F) 94 124 165 222 294 367 410

LCO (°F) 319 460 503 533 569 618 653

MCB (°F) 524 710 759* 807* 876* 1100* 1200*

Total

17,680

6,515

2,227

26,422

Note - *These temperatures are not needed. %x = ((Tx-Ta)/(Tb-Ta))*(%b-%a)+%a %a < %x < %b Ta < Tx < Tb i,e, = ((100-94)/(124-94))*(10-0) +0 = 2.5 BPD = [(%Gasoline)(BPSD Gasoline) + (% LCO)(bpsd LCO)+(% MCB)(bpsd MCB)} / 100 %

= 100 BPD / Total BPD


Data: Tx, °F 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 367 380 400 410 420 440 460 480 500 520 540 560 580 600 618 620 640 660 680 700

Gasoline

LCO

MCB

%x 0.0 2.0 8.7 17.8 27.6 35.3 42.3 49.3 55.0 60.6 66.1 71.6 77.1 82.6 88.1 90.0 93.0 97.7 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

%x

%x

0.0 0.1 1.5 2.9 3.4 4.3 5.7 6.5 7.2 8.6 10.0 19.3 28.6 37.9 53.9 65.0 74.5 82.7 90.0 90.6 96.3 100.0 100.0 100.0

0.0 0.9 1.9 3.0 4.1 5.1 5.2 6.2 7.3 8.4 9.5

Composite BPD* 0 354 1,532 3,148 4,873 6,235 7,475 8,716 9,724 10,706 11,688 12,667 13,640 14,701 15,726 16,134 16,728 17,643 18,100 18,147 18,239 18,332 18,938 19,544 20,150 21,210 21,958 22,600 23,156 23,656 23,696 24,092 24,358 24,382 24,406

Percent 0 1.34 5.80 11.91 18.44 23.60 28.29 32.99 36.80 40.52 44.24 47.94 51.62 55.64 59.66 61.06 63.31 66.77 68.51 68.68 69.03 69.38 71.67 73.97 76.26 80.27 83.10 85.54 87.64 89.53 89.68 91.18 92.19 92.28 92.37

Corrected Gasoline = =

(Composite Yield @ 380F) / (0.9) (16,728 / 0.9) = 18,587 BPD

Corrected LCO

= =

((Composite Yield @ 600F) - (Corrected Gasoline)) / (0.9) (23,156 – 18,587) / 0.9 = 5,077 BPD

Corrected MCB

= =

(Total Liquid Yield – Corrected Gasoline – Corrected LCO) (26,422 – 18,587 – 5,077) = 2,759 BPD


REACTOR AND REGENERATOR HEAT BALANCE

Burning coke in the regenerator provides all the heat necessary for the operation of the unit. Yet, roughly 30-40% of the heat generated by the combustion of coke exits the regenerator in the form of hot flue gas. The remainder is absorbed by the regenerated catalyst which carries it to the reactor riser where it is used to vaporize and heat up the combined feed to the desired cracking temperature. The amounts of energy associated with the unit's operation are determined from a catalyst section heat balance. The energy balance equation at steady state may be written as: Energy in + Energy produced = Energy out + Energy consumed

(1)

Regenerator Energy Balance

Energy in

= Energy (air + spent catalyst + coke)

Energy produced

= Combustion heat of coke

Energy out

= Energy (flue gas + regenerated catalyst + removed + radiation losses)

Energy consumed

= 0

If the Regenerator temperature is the reference temperature then, -∆H Air - ∆H Spent Catalyst - ∆H Coke + ∆H Combustion of coke = ∆Hremoved + ∆Hradiation losses

or:

∆H Spent Catalyst = ∆H Combustion of Coke - ∆H Coke - ∆H Air - ∆Hremoved ∆Hlosses (2)


Reactor Energy Balance

Energy in

= Energy (feed + regenerated catalyst + diluents)

Energy produced

= 0

Energy out

= Energy (reactor vapors + spent catalyst + radiation losses)

Energy consumed = Heat of reaction If the Reactor temperature is the reference base temperature, then -∆Hfeed - ∆Hdiluents + ∆Hregenerated catalyst = ∆Hradiation losses + Heat of Reaction or ∆Hregenerated catalyst = ∆Hfeed + ∆Hdiluents + ∆Hradiation losses + Heat of Reaction

(3)

The enthalpy change for the spent and regenerated catalyst is given by ∆Hspent catalyst = mass flow Cp (Rg Temp - Rx Temp)

(4)

∆Hregenerated catalyst = mass flow Cp (Rx Temp - Rg Temp)

(5)

At steady conditions, ∆Hspent catalyst + ∆Hregenerated catalyst = 0

(6)


Regenerator Heat Balance Flue Gas Spent Cataly st Coke Ra diati on Los ses H Combustion of Coke He at Remov al

Regenerated Catal ys t Air

Reactor Heat Balance Reactor Vapors Regenerated Catalyst

H Reaction

Spent Catalyst Coke Feed

Diluents

Radiation Losses


Combining equations (2), (3) and (6) ∆H Combustion of coke = ∆H Air + ∆H Coke + ∆Hremoved + ∆Hregen.rad. losses + ∆Hfeed + ∆Hdiluents + ∆Hrx rad. losses + Heat of Reaction

(7)

The equation (7) demonstrates that all the energy in the Reactor-Regenerator system is provided by the combustion of coke. The radiation loss term in this equation is not a major item, but since vessel insulation is not perfect, some radiation losses do occur. The term ∆Hremoved refers to the heat duty of catalyst cooler(s). The heat of reaction is the energy required to convert the feed to products via the catalytic reaction mechanism. The heat produced by the combustion of coke, equation (7), can be calculated from the coke product rate and the mode of the regeneration operation. If all the CO were burned to CO2 in the regenerator, more heat would be available per pound of carbon than when the unit runs in this normal partial combustion mode. The heat liberated by carbon combustion to CO2 is 14,150 BTU/lb (7,860 kcal/kg or 32,910 kj/kg) of carbon whereas heat for combustion to CO is only 3,960 BTU/lb (2,200 kcal/kg or 9,210 kj/kg). Conversion Factors: kcal/kg  1.8 = BTU/lb BTU/lb  2.326 = kJ/kg kcal/kg  4.187 = kJ/kg The heat of reaction is endothermic. Energy is consumed by the reaction which breaks the heavy hydrocarbon molecules into smaller, light hydrocarbon products. The heat of reaction must be calculated from the energy balance using equation (7). The most important value that can be calculated from the energy balance is the catalyst/oil weight ratio. This ratio is important because it is a major factor in hydrocarbon conversion and coke lay-down.


The following sample outlines the energy balance calculation method: 1.

Data Required

This heat balance is for a 30,565 BPSD feed case with the FCC Unit in total combustion mode. The process conditions are: Temperatures:

Pressures:

Reactor Combined Feed Lift Gas Lift Steam Feed Steam Stripping Steam

970°F 375°F 100°F 380°F 380°F 380°F

Regenerator Regenerated Catalyst Flue Gas Average Hottest in Rg Air Blower Discharge HP Boiler Feed Water Catalyst Cooler Steam

1371°F 1368°F 1375°F 399°F 220°F 463°F

Catalyst Cooler Steam

452 psig


Flow Rates:

*

Fresh Feed (No Recycle) Lift Gas Lift Steam Feed Steam Stripping Steam Total Air to Regenerator* Catalyst Cooler Steam Catalyst Cooler Blowdown

3,250 lb/hr 12,900 lb/hr 1,800 lb/hr 5,000 lb/hr 382,405 lb/hr 56,033 lb/hr 6,952 lb/hr

Total air includes: air to combustor, air to upper regenerator, and air to catalyst cooler.

Flue Gas Composition, mol% (by GC Method)

CO CO2 O2 + Ar N2 SO2 NO2

2.

412,923 lb/hr

= = = = = =

0.0 15.50 3.47 81.03 0.0 0.0

Flue Gas Composition Adjustment

Unlike an Orsat analysis, a GC analysis includes Argon with the Oxygen. The first step is to adjust the flue gas O2 content for Argon (Ar). The Ar content is assumed to be 1.2% of the nitrogen, therefore, Ar = (0.012) (81.03) = 0.97 mol%.


The corrected analysis is now: CO CO2 O2

= = = = = =

N2 + Ar SO2 NO2

3.

0 15.50 3.47 - 0.97 = 2.5

81.03 + 0.97 = 82.0 0 0

Combustion Air Corrected to a Dry Basis

A psychometric chart is used to determine the moisture content of the regeneration air. At atmospheric conditions of 62°F and a relative humidity of 97%, the moisture content is: Moisture Content =

0.01152 lb H2O lb dry air

Wet Air = 380,200 lb/hr Dry Air = 380, 200 lb/hr wet air 

1 lb dry air (1 + 0.01152) lb wet air

= 375, 870 lb/hr

Water in Air = 380,200 lb/hr - 375,870 lb/hr = 4,330 lb/hr

4.

Calculate Flue Gas Rate

The flue gas rate can be calculated from the regeneration air rate. These two streams are related by the inert N2 + Ar content which remains constant through the catalyst regeneration. From a Nitrogen balance, Since, moles =

Weight Molecular Weight


Then Dry Air = (375,870 lb/hr)/(28.966 MW) = 12,976 lb mol/hr mol/hr (N2 + Ar) in dry air = mol/hr (N2 + Ar) in flue gas 79 mol inerts lb mol FG 82 mol inerts 12, 976 lb mol  =  100 mol air hr 100 mol FG hr

Flue Gas (FG) = 12,501 lb mol/hr

5.

Calculate the Carbon (C) Content of Coke

The carbon (C) content of the coke is calculated from the flue gas composition. One mol of C is burned for each mole of CO or CO2produced. C + O2 + H2 + S + N = CO + CO2 + H2O + SO2 + NO2 + O2 C=

0 mol CO + 15.503 mol CO2 1 mol C 12, 501 lb mol   mol CO/CO2 hr FG 100 mol FG

C = 1,938 lb mol/hr of carbon

6.

Calculate the Hydrogen Content of Coke

The hydrogen (H2) content of the coke must be calculated from an O2 balance: O2 in regeneration air = excess O2 in flue gas + + O2 reacted to CO (0.5 mol O2/mol CO) + O2 reacted to CO2 (1 mol O2/mol CO2) + O2 reacted to H2O (0.5 mol O2/mol H2O) + O2 reacted to SO2 (1 mol O2/mol SO2) + O2 reacted to NO2 (1 mol O2/mol NO2)


Where: O2 in regen. air =

12, 976 lb mol dry air 21 mol O 2 2,725 lb mol  = of O2 hr hr 100 mol air

Excess O2 in FG =

O2 reacted to CO =

12, 501 lb mol FG 2.5 mol O2 312 lb mol  = of O2 hr hr 100 mol FG 12, 501 lb mol FG 0 mol CO 0.5 mol O2   hr 100 mol FG mol CO

= 0 lb mol/hr O2 O2 reacted to CO2 =

12, 501 lb mol FG 15.5 mol CO 2 1 mol O2   hr 100 mol FG mol CO2

= 1,938 lb mol/hr of O2 O2 reacted to SO2 =

12, 501 lb mol FG 0 mol SO2 1 mol O2   hr 100 mol FG mol SO2

= 0 lb mol/hr of O2 O2 reacted to NO2 =

12, 501 lb mol FG 0 mol NO2 1 mol O2   hr 100 mol FG mol NO 2

= 0 lb mol/hr of O2 O2 reacted to H2O (by difference) is: O2 Reacted to H2O = 2,725 - 312 - 0 - 1,938 - 0 - 0 lb mol/hr O2 = 475 lb mol/hr of O2 The hydrogen burned by oxygen in the regenerator is: H2 burned by O2 =

475 lb mol 2 mol H2 950 lb mol  = hr O2 mol O2 hr H2


7.

Calculate Coke from Carbon and Hydrogen

The mass of coke combusted to CO + CO2 + H2O is: from carbon =

1, 938 lb mol 12.01 lb C 23, 275 lb  = hr C lb mol C hr C

from hydrogen =

950 lb mol 2.016 lb H 1, 915 lb  = hr H2 lb mol H2 hr H

Total = 23,275 + 1,915 = 25,190 lb/hr coke

8.

Calculate Coke Yield Percent

The quantity of coke produced from the fresh feed is:

Coke Yield =

Coke, lb/hr100 FF,lb/hr 

Coke Yield =

25, 190 lb/hr coke 412, 923 lb/hr raw oil

9.

 100 = 6.10 wt - % coke

Calculate Hydrogen in Coke

The H2 content of the coke is: H2 in Coke =

H2 in Coke 

H2 ,

lb/hr(100) Coke, lb/hr

1,915 lb/hr H 25,190 lb/hr coke

 100 = 7.6 wt - % hydrogen


10. Calculate the Air/Coke Ratio

Air to Coke =

(Air, lb/hr)(100) Coke, lb/hr

Air to Coke =

375,870 lb/hr dry air lb of dry air = 14.92 25, 190 lb/hr coke lb of coke

11. Calculate the Heat of Combustion of Coke

Combustion heats are calculated based on the average hottest temperature in the regenerator. The dense, dilute, cyclones, and flue gas average temperatures are calculated and the hottest is used as basis. The average hottest temperature is 1375°F.

Hc (2C + O2

2CO) = 46,216 + 1.47 (1375°F) = 48,237

= (0

BTU lb  mole

BTU  lb  mole  of O2 reacted to CO) (2) 48, 237  lb  mole hr

= 0 BTU/hr

Hc (C + O2

CO2) = 169,135 + 0.5 (1375°F) = 169,822

= (1,938

BTU lb  mole

BTU  lb  mole  of O2 reacted to CO2) (1) 169, 822  lb  mole  hr

= 329.12  106 BTU/hr


Hc (2H2 + O2

2H2O) = 104,546 + 1.585 (1375°F) = 106,725

= (475

BTU lb  mole

BTU  lb  mole  of O2 reacted to H2O) (2) 106, 725  lb  mole  hr

= 101.389  106 BTU/hr ∆HCombustion of Coke = (0 + 329.12 + 101.39)  106 = 430.51  106 BTU/hr Using as basis 1 lb of coke ∆H combustion of coke =

430.51  10 6 BTU/hr 25, 190 lb/hr coke

= 17,090 BTU/lb coke This heat of combustion must be corrected for the coke’s hydrogen content according to the equation Correction

= 1133 - 134.6 (wt-% H) = 1133 - 134.6 (7.6) = +110 BTU/lb coke

The net heat of combustion of coke is: ∆HCombustion = 17,090 + 110 BTU/lb coke = 17,200 BTU/lb coke


REGENERATOR HEAT BALANCE Basis: 1 lb of coke 12. Heat Consumed to Heat Up the Regeneration Air

Since ∆H = mass  Cp  ∆T Air is heated from the main air blower discharge temperature of 399°F to the average hottest temperature of 1375°F at an average specific heat of 0.26 BTU/lb °F. 375, 870 lb/hr air 0.26 BTU HAir =  (1375- 399F)  = 3, 787 BTU/lb coke 25,190 lb/hr coke lb F

13. Heat Consumed to Heat up the Regeneration Air Water Vapor Water vapor is heated from 399 to 1375°F at an average specific heat of 0.5 BTU/lb °F. 0.5 BTU 4, 330 lb/hr H2 O HH O    (1375 - 399F) = 83.9 BTU/lb coke 2 lb F 25, 190 lb/hr coke

14. Heat Consumed to Heat Up the Coke Coke is heated from the reactor temperature of 970°F to the average hottest temperature of 1375°F at an average specific heat of 0.4 BTU/lb °F. ∆HCoke = (1375-970°F)  0.4 BTU/lb °F = 162 BTU/lb coke


15. Heat Consumed to Generate Steam in the Catalyst Cooler Enthalpies for water and steam are obtained from steam tables. Water in at 220°F Steam out at 463°F

= (188 BTU/lb) (56,033 + 6,952 lb/hr) = 11.841  106 BTU/hr = (1,205 BTU/lb) (56,033 lb/hr) = 67.52  106 BTU/hr

Blowdown out at 463°F = (441 BTU/lb) (6,952 lb/hr) = 3.066  106 BTU/hr Catalyst cooler duty = (67.52 + 3.066 - 11.841)  106 BTU/hr = 58.75  106 BTU/hr Or

∆HRemoved =

58.75  106 BTU/hr = 2, 332 BTU/lb of coke 25,190 lb/hr of Coke

16. Regenerator Heat Balance Using a typical regenerator heat loss rate of 250 BTU/lb coke, the heat consumed to heat up the catalyst is: RgHeat = (∆HComb Coke) - ∆HCoke - ∆HAir - ∆HH2O - ∆HLoss - ∆HRemoved Hregen = 17,200 - 3,787 - 84 - 162 - 2,332 - 250 = 10,585 BTU/lb Coke


17. Calculate the Catalyst Circulation Rate The catalyst is heated from the reactor temperature of 970°F to the regenerated catalyst temperature of 1371°F at an average specific heat of 0.275 BTU/lb °F. Since Q = m Cp ∆T then m = Q/Cp ∆T (Coke lb/hr)(Rg Heat BTU/lb Coke) (0.275 BTU/lb F) (Cat T - RXT)

CCR =

CCR =

25,190 lb/hr coke  10,585 BTU/lb coke  40, 299 lb/min 0.275 BTU/lb F  (1371 - 970F)  60 min/hr

or CCR =

40, 299 lb/min = 20.15 ton/min 2,000 lb/ton

18. Calculate the Catalyst/Oil Ratio C/O =

CCR lb/hr FF lb/hr

C/O =

40, 299 lb/min catalyst  60 min/hr = 5.86 wt/wt 412,923 lb/hr fresh feed

19. Calculate the Regenerator Efficiency Rg Eff =

Rg Heat  100 HCombustion of Coke

=

(10, 585 BTU/lb of Coke) (100) 17, 200 BTU/lb of Coke

= 62% (This number will be higher without catalyst cooler)


20. Calculate the Delta Coke Wt-%: Coke

(100) (Coke, lb/hr) 25, 190  100 = = 1.04 wt% Cat. Circ. lb/hr 40, 299  60

REACTOR HEAT CALCULATIONS 1 lb of fresh feed is used as basis in the following calculations.

21. Heat Consumed to Heat and Vaporize the Combined Feed Enthalpies for the raw oil feed are obtained by using the equation as discussed after this section. The UOP K to use for entering the enthalpy table is calculated from UOP Method 375 as discussed in the following section. UOP K Factor is a function of °API and Engler distillation. A high UOP K value of 12.5 indicates a more paraffinic (saturated chain) hydrocarbon, while a lower value of 11.2 occurs for a more aromatic (unsaturated cyclic) stock. Higher UOP K paraffinic feeds crack easier yielding higher conversion at a given reactor temperature. Raw Oil: UOP K = 11.8

°API Gravity = 21.3

At the 375°F riser inlet temperature,

Hraw oil

= 252 BTU/lb

At the 970°F reactor temperature,

Hvapor

= 760 BTU/lb

∆Hraw oil = 412,923 lb/hr  (760-2891 BTU/lb) = 194.487  106 BTU/hr The heat required to heat up the combined feed is the ∆Hraw oil multiplied by the Combined Feed Ratio (CFR), where, CFR =

(raw oil + recycle) lb/hr = 1.0 raw oil lb/hr

So, ∆Hcomb feed = 194.487  106 BTU/hr


Hcomb feed =

194.487  106 BTU/hr = 471 BTU/lb raw oil 412, 923 lb/hr raw oil

22. Heat Consumed to Heat Up the Lift Gas: The lift gas is heated from 110°F to 970°F at an assumed average specific heat of 0.5 BTU/lb °F. Hlift gas =

3, 250 lb/hr lift gas  (970 - 110F)  0.5 BTU/lb F = 3.4 BTU/lb raw oil 412, 923 lb/hr raw oil

23. Heat Consumed to Heat Up Lift Steam, Feed Steam and Stripping Steam Steam is heated up from the header temperature of 380°F to the reactor temperature at 970°F at an average specific heat of 0.485 BTU/lb °F.  Hsteam =

(5, 000 + 12, 900 + 1, 800) lb/hr  (970 - 380 F)  0.485 412, 923 lb/hr raw oil

= 13.8 BTU/lb raw oil

24. Heat of Inert Gas Carried from Regenerator to Reactor by Regenerated Catalyst The inerts gas can be calculated from the sponge gas stream and the procedure is at the end of this section. If this number is unknown, use 0.007% of fresh feed. Use an average specific heat of 0.275 BTU/lb °F. ∆Hinerts = (inerts wt%) (Cp) (RxT - RgT) = 0.007  0.275  (970 - 1371) = -0.8 BTU/lb raw oil


25. Reactor Heat Balance The total heat consumed in the reactor equals the sum of the heats consumed for the combined feed, lift gas, all steam, the reactor losses, plus the heat of reaction. Using a typical reactor heat loss rate of 2 BTU/lb raw oil, the heat balance is: Rx Heat = ∆Hcomb feed + ∆Hlift gas + ∆Hsteam + ∆Hinerts + ∆Hloss + ∆HRxN Hreactor = (471 + 3.4 + 13.8 - 0.8 + 2) BTU/lb raw oil + ∆Hreaction Hreactor = 489.4 BTU/lb raw oil + ∆Hreaction

26. Overall Heat Balance The heat consumed in the reactor is supplied by the hot catalyst circulated to the riser. At steady state, the heat consumed in the reactor must balance the heat produced in the regenerator. The reactor heat is based on a per lb of fresh feed basis while that of the regenerator is on a per lb of coke basis. These two can be equated using the raw oil to coke weight fraction to determine the heat of reaction: lb coke   Hregenerator BTU / lb coke  = Hreactor [BTU/lb raw oil] lb raw oil  

10, 585 BTU 25, 190 lb/hr coke  = 489.4 BTU/lb raw oil + ∆Hreaction lb coke 412, 923 lb/hr raw oil ∆Hreaction = 119 BTU/lb raw oil


FCC Unit Heat Balance Example Refiner: Location: Combustion Air Correction to a Dry Basis (a) Ambient Temperature = 62 °F (b) Relative Humidity = 97 % (c) Sat. Vapor Pressure = 10^[6.40375-(3165.36/(°F+392.565))] = 0.275586177 psia (d) lb H2O/lb dry air = [0.00622*(b)*(c)]/[14.7-0.01*(b)*(c)] = 0.011520532 lb/lb (e) Total Air to Regen = 380,200 lb/hr (f) Total Dry Air = (e)/[1+(d)] = 375,870lb/hr (h) H2O in Air = (e) – (f) = 4,330 lb/hr

Date: Flue Gas Composition Adjustment: GC, mol-% Corrected for Ar* (i) CO = 0.00 CO = 0.00 15.50 CO2 = 15.50 (j) CO2 = 3.47 O2 - (0.012*N2) = 2.50 (k) O2 = 81.03 N2 + O2 - Cor O2 = 82.00 (L) N2 = 0.00 SO2 = 0.00 (m) SO2 = 0.00 (n) NO2 = 0.00 NO2 = Total 100.00 100.00 *Correction required only for GC not for Orsat analysis.

Temperatures: Rg=Regenerator, Rx=Reactor (o) Flue Gas Line = 1368 °F (p) Rg Avg Cyclone Outlet = 1368 °F (q) Rg Avg Dilute = 1375 °F = 1371 °F (r) Rg Avg Dense (RgT) (s) Avg Hottest Rg Temp = 1375 °F Oxygen Balance: (y) (L) * (2 * 21 ) / 79 – 2(n) - 2(m) - 2(k) - 2(j) - (i) (z) 2.016*(y) =12.01[ (i) + (j) ] + 32.06*(m) + 46.01*(n) (1a) Hydrogen = 2.016 * (y) * 100 / (z) (1b) Air = 28.966 * 100 * (L) / [ 79 * (z) ] (1c) Coke = Dry Air / [ Air / Coke ] = (f) / (1b) (1d) Fresh Feed S.G. = 0.926 (1e) Coke Yield = (1c) * 100 / (1d)

(t) (u) (v) (w)

= = = = = = =

Air to Rg Rx Temp Hot Rg T – Air T Rg Dense T – Rx T

7.6 201.5 7.6 14.9 25,187 412,923 6.10

= = = =

399 970 976 401

°F °F °F °F

mol H2 / 100 mol Flue Gas lb coke / 100 mol Flue Gas wt-% Hydrogen in Coke Air/Coke, lb/lb lb/hr of Coke lb/hr Feed wt-% Coke (of Fresh Feed)

Combustion of Coke Basis: Average Hottest Regenerator Temperature: (1f) Hc(CO) = 46,216 + 1.47 * (s) = 48,237 BTU/lb-mol (1g) Hc(CO2) = 169,135 + 0.5 * (s) = 169,823 BTU/lb-mol (1h) Hc(H2O) = 104,546 + 1.585 * (s) = 106,725 BTU/lb-mol H Comb = [ (1f) * (i) + (1g) * (j) + (1h) * (y) ] / (z) = 17,091 BTU/lb-mol (1i) (1j) Correction Factor = 1133 – 134.64 * (1a) = 109 BTU/lb-mol H Combustion Coke = (1i) + (1j) = 17,200 BTU/lb-mol (1k) Regenerator Heat Balance Basis: 1 lb of Coke (1m) H coke = 0.4 BTU/lb-mol °F * (w) H air = (1b) * 0.26 BTU/lb °F * (v) (1n) H H2O = (h) / (1c) * 0.485 BTU/lb °F * (v) (1o) H Radiant Losses (1p) (1q) Cat Cooler Heat Duty H removed = Cooler Duty / Coke = (1q) * 10^6 / (1c) (1r) (1s) Rg Heat = (1k) – (1m) – (1n) – (1o) – (1p) – (1r) (1t) Rg Eff = Rg Heat * 100 / HcombCoke = (1s) * 100 / (1k) (1u) Cat/Oil = (1s) * (1e) / 100 * (0.275 BTU/lb °F) * (w) (1v) Cat Circ = (Cat/Oil) * (FF) / 60 = (1u) * (1d) / 60 min/hr) Coke = 100 * Coke / (60 * Cat Circ) = 100 * (1c) / 60 * (1v) (1w)

= = = = = = = = = = =

160 3,787 81 250 58.7 2,331 10,590 61.6 5.86 40,313 1.04

BTU/lb coke BTU/lb coke BTU/lb coke BTU/lb coke MM-BTU/hr (calculated separately) BTU/lb coke BTU/lb coke % Regenerator Efficiency Catalyst-to-Oil Ratio Catalyst Circulation, lb/min Coke, wt-%


FCC Unit Heat Balance Example Continuation Refiner: Location:

Date:

Reactor Heat Calculations: Basis: 1 lb of Coke (2a) Rx Temp (RxT) (2b) Combined Feed Temp. (CFT) (2c) Fresh Feed (FF) (2d) FF Enthalpy @ CFT (H @ CFT)* (2e) FF Enthalpy @ RxT (H @ RxT)I (2f) Recycle (Recy) (2g) Recy Enthalpy @ CFT (E @ CFT)* (2h) Recy Enthalpy @ RxT (E @ RxT)* *See Following Pages for Enthalpy Calculation Method

= = = = = = = =

970 375 412,923 289 760 0 0 0

°F °F lb/hr BTU/lb BTU/lb lb/hr BTU/lb BTU/lb

(2i) (2j) (2k) (2m) (2n) (2o) (2p) (2q) (2r) (2s)

Inerts from Rg with catalyst (Inerts) Inerts Specific Heat (Cp) Lift Gas (LGas) Lift Gas Temp (LGasT) Lift Gas Specific Heat (LGas Cp) Steam Temp (StmT) Steam Specific Heat (Stm Cp) Stripping Steam Lift Steam Feed Steam

= = = = = = = = = =

2,748 0.275 3,250 110 0.5 380 0.485 5,000 12,900 1,800

lb/hr BTU/lb °F lb/hr °F BTU/lb °F °F BTU/lb °F lb/hr lb/hr lb/hr

(2t) (2u) (2v) (2w) (2y) (2z)

Heat consumed by FF = (FF) * (H @ RxT – H @ CFT) / (FF) Heat consumed by Recycle = (Recycle) * (E @ CFT – E @ RxT) / FF Heat Consumed by Steam = (Total Stm) * (Stm Cp) * (RxT-StmT) / FF Heat Consumed by Lift Gas = (LGas) * (LGas Cp) * (RxT-StmT) / FF Heat from Inerts = (Inerts) * (Inerts Cp) * (RxT-RgT) / FF Reactor Heat Loss

= = = = = =

471 0 13.7 3.4 -0.7 2

BTU/lb FF BTU/lb FF BTU/lb FF BTU/lb FF BTU/lb FF BTU/lb FF

(3a)

Rx Heat = (H FF) + (H Recy) + (H Stm) + (H Lgas) + (H Inrts) + (H Loss) + (H Rxn) = 489.3

BTU/lb FF + H Reaction

(Rg Heat BTU/lb Coke) * (Coke lb/hr) / (FF lb/hr) = Rx Heat = BTU/lb + H Rxn (3b)

H Rxn = [(Rg Heat BTU/lb Coke) * (Coke lb/hr) / (FF lb/hr)] – BTU/lb =

157

H Rxn BTU/ lb FF


uop K Factor from °API and Engler Distillation Data:

ASTM Distillation: Vol% Temperature, °F 10 660 30 781 50 887 70 1015 90 1075

Specific Gravity:

0.9258

1. Calculate the volumetric average boiling point as the average of the 10, 30, 50, 70 and 90 vol-% temperatures. VABP = (T10% + T30% + T50% + T70% + T90%) / 5 VABP = (660 + 781 + 887 + 1015 + 1075) / 5 =

883.6

2. Calculate the Engler slope as °F per percent (°F/%) by subtracting the 10 vol-% temperature from the 90 vol-% temperature, and dividing the difference by 80. Slope = (T90% – T10%)/80

=

5.1875

3. Calculate the Cubic Average Boiling Point (CABP): CABP = VABP * A + B Where:

A = (0.000297 * Slope + 0.001438)*Slope + 1.0 A = 1.01545 B = (-0.581 * Slope – 1.339)*Slope B = -22.5809 4. Calculate UOP K: UOP K 

3

CABP  459.69 SG

= 11.89


Enthalpy of Heavy Petroleum Fractions The Following equations can be used to calculate the Enthalpies for the feed and recycle streams on the FCC unit. Liquid Enthalpy Equation:

Equation source is API Procedure 4.7.B4

A1 =

(-1171.26 + (23.722 + 24.907 * SG) * UOP K)

A1 =

A1 + (1149.82 – 46.535 * UOP K) / SG

A1=

A1 / 1,000

A2 =

(1 + 0.82463 * UOP K) * (56.086 – 13.817 / SG) / 1,000,000

A3 =

- (1 + 0.82463 * UOP K) * (9.6757 – 2.3653 / SG) / 1E+09

The enthalpy of liquid FCC feedstock at the riser inlet conditions is: Hin = A1 * (TE – 259.67) + A2 * (TE² – 259.67²) + A3 * (TE³ – 259.67³) Where:

TE

=

Combined Feed Temperature, (°F + 459.67)

SG

=

Fresh Feed Specific Gravity

UOP K

=

Fresh Feed UOP K

Vapor Enthalpy Equation: Equation source is a curve fit from UOP PD Chart PD-189 F1 =

3.0186E-04 * SG + 3.9975E-06 * UOP K * (UOP K – 13.8584)

F2 =

0.67036000 * SG + 0.00675130 * UOP K * (UOP K – 24.7770)

F3 =

85.52390000 * SG – 4.73260000 * UOP K * (UOP K – 21.9249) – 459.6742

Enthalpy of feed in fully vaporized condition is: Hout = F1 * (T²) – F2 * T + F3 Where: T

=

Reactor temperature, °F


Heat of Combustion of Coke, BTU/lb-Mole Table: Temperature, °F

77

1,100

1,200

1,250

1,300

1,350

1,400

CO

47,565

47,847

47,980

48,050

48,123

48,199

48,274

CO2

169,332

169,677

169,735

169,760

169,784

169,808

169,835

H2O Vapor

104,129

106,279

106,448

106,529

106,610

106,687

106,765

Equations: Hc(CO) = Hc(CO2) = Hc(H2O) = Where:

46,216 + 1.47 * (T) 169,135 + 0.5 * (T) 104,546 + 1.585 * (T)

T is in °F


MECHANICAL EVALUATION The Mechanical Evaluation Test should cover all facets of the FCC Unit, including the reactor, regenerator, main column, and the gas concentration unit. All major pieces of equipment should be part of this test, including vessels, pumps, compressors, heat exchangers, and piping hydraulics. The procedure for this test is long and involved, but the information can be very useful for the Refiner and for UOP. If the Refiner wants to revamp the unit, it is important to determine maximum throughput and actual equipment limitations. Most of the information will be collected only once, although parts of it (such as exchanger surveys) can be repeated to follow fouling or other potential problems. The lists and data sheets included in this section can be used as guidelines in collecting the required data, although the Refiner may have to modify certain parts for his particular unit. It is important to finish collecting the data within as short a time as possible. A single survey is generally satisfactory and it is no necessary to use long term average data. The Unit should be operating smoothly to get realistic and good quality data. Label the data collected and prepare a report in an orderly fashion.


GENERAL INFORMATION LIST This list is only a guideline. Please modify or expand as required. 1.

Ambient Air Conditions a. Temperature b. Relative Humidity c. Barometric Pressure d. Wind Velocity and Direction (show on rough plot plan)

2.

General Description of Unit a. Process Flow Diagram, including flow meter and control valve locations b. Plot Plan showing Layout of Vessel and Equipment c. Operational Mode (partial or complete CO Combustion)

3.

Units used (USA, Imperial, Metric) and Standard Conditions (0°C, 760 mm; 60°F, 14.7 psia)

4.

Limiting Factors a. Environmental Constraints (CO emissions, special specifications) b. Utility Limitation (shortage of steam or electricity, etc.)

5.

Performance Data a. Accurate Flow and Weight Balance including sample analyses

product


HYDRAULIC AND PROCESS SURVEY LIST 1.

Single gauge pressure survey of every point in reactor-regenerator circuit, including air into regenerator and flue gas out to point of discharge.

2.

Slide valve positions and hydraulic oil pressure at each valve.

3.

Air blower suction and discharge pressures, total and net (to regenerator) flow rates, relative humidity and temperature of air, with manufacturer’s data and performance curve for comparison.

4.

Electrical or steam consumption for blower driver.

5.

Power recovery units should add flue gas temperatures and pressures around expander, butterfly valve positions, electrical power consumed or generated and single gauge pressure survey of third stage separator.

6.

For electrostatic precipitator, or other flue gas treaters, take temperatures in and out, power consumption, and amount and size distribution of particulates removed.

7.

Complete flue gas sample before and after flue gas treater.

8.

Catalyst consumption and losses.

9.

Single gauge pressure survey of main column and gas concentration section (use one low pressure and one high pressure gauge, depending on location, for better accuracy). Include feed flow rate and temperature, reflux flow rate and temperature, reboiler heat input, overhead temperature and pressure and enough other data to calculate a heat and weight balance around the column.

10.

Samples of main column overhead receiver gas and liquid; and gas, hydrocarbons, and water samples from high pressure receiver for phase equilibrium studies.


11.

Compressor suction and discharge pressures, flow rates, composition, and temperatures, with manufacturer’s performance curve for comparison. Include amounts and compositions of liquids drained from knockout drums.

12.

Compressor driver type and power consumption.

13.

All pump suction and discharge pressures, flow rates, liquid compositions or boiling ranges, and power consumption of driver, with manufacturer’s performance data for comparison.

14.

Data on streams not usually measured, such as LCO to the sponge absorber, including flow rates, temperatures, composition or boiling range, and single gauge pressure survey of circuit.

15.

Pressures, temperatures, and flow rates of flushing oil to instruments and pump seals and glands.

16.

Utility consumption/product data: Steam (all pressures) Air (plant and instrument) Purges to instruments, packing glands, and expansion joints (specify air, steam, nitrogen, or fuel gas), with single gauge pressure survey of utility lines at purge points). Cooling water Boiler feed water Steam condensate Utility water Treating chemicals for boiler feed water (type and amount) Inhibitor and anti-corrosion chemicals used


EXCHANGERS INFORMATION LIST 1.

Flow through exchangers on both sides (gas and liquid), composition or boiling range, and mass flow.

2.

Temperatures in and out of both sides, also between shells and bundles.

3.

Pressures in and out of both sides, also between shells and bundles.

4.

Any material bypassed around exchangers (give rough sketch).

5.

If air coolers: air temperatures in and out, air velocity out, motor amps, note any belt slippage, variable pitch position, louver position, etc.

6.

In preparing data, submit overall heat transfer coefficient and specifics on exchangers.


FIRED HEATERS INFORMATION LIST 1.

Process flow (volume and mass, composition, molecular weight and boiling range).

2.

Single gauge pressure survey for both process and fuel system.

3.

Fuel type (gas or oil) and analysis (composition, sulfur, gravity, etc.), pressure and temperature of fuel at heater.

4.

Fuel consumption.

5.

Steam or air pressure for fuel oil atomization.

6.

Temperatures throughout the heater, such as firebox, convection points, stack, air preheat, and all process points.

7.

Draft in firebox and stack.

8.

Design information: type of furnace, materials of construction, and number, layout and materials of tubes; including dimensions of furnace.

9.

Burner data: rating, design, number. Note any unusual problems such as plugged or inoperative burners.

10.

Refiner should obtain sufficient data to calculate heat flux from both process and fire side, heat release, heater efficiency and steam balance.



Reactor-Regenerator Pressure Survey Refiner: Location: To Main Column

Date:

J

Time: By:

I 19 18 17 16 15 14 12

13

Steam Generator

Orifice Chamber

Flue Gas SV

11

20

H

10 ESP

Flue Gas to Stack

9

G 8

Slide Valves % Open Regenerated Recirculating Spent Flue Gas A Flue Gas B

6

7

F

Cat Cooler Slidevalve

E

22

Feed

5 D

4

Process Flow

C

DFAH

Feed Rate Air Rate

Main Air Blower

2

A

B

3

1 Atmospheric Air

Pressure Survey: _____________Units of Pressure 1 2 3 4 5 6 7 8

11 12 13 14 15 16 17 18

A B C D E F G H




MAIN COLUMN SUMMARY – BOTTOMS

page ____________________________ date ____________________________ Item No.: _____________________________ by ____________________________ Service: __________________________________________________________________ Type of Operation: _________________________________________________________ No. of Trays: _________________ Reflux Ratio:_________________________________ Type of Trays: _____________________________________________________________ Main Column Bottoms

Circulating HCO LCO

Circ.

Quench

CSO

Other

Mass Flow

_______

_______

_______

_______

_______ ________

Temperature Out Return Pressure

_______ _______ _______

_______ _______ _______

_______ _______ _______

_______ _______ _______

_______ ________ _______ ________ _______ ________

Distillation IBP 5% 10% 20% 30% 40% 50% 60% 70% 80% 90% 95% EP API or S.G.

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

BS & W

_______

_______

_______

_______

_______ ________

Steam to Stripper

_______

_______

_______

_______

_______ ________

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

(Sketch system showing flows, P, T, Q on separate page) __________________________ Weight balance _______________________ Heat balance _________________________ Deviations from UOP Specifications: ___________________________________________ ________________________________________________________________________


MAIN COLUMN SUMMARY — CYCLE OIL PRODUCTS AND OVHD.

page _______________________________ date _______________________________ Item No.: ____________________________ by _______________________________ Service: _________________________________________________________________ Type of Operation: _________________________________________________________ No. of Trays: ________________ Reflux Ratio: ________________________________ Type of Trays: ____________________________________________________________ HCO Product

LCO Product

Naphtha Product

Reflux

Net Ovhd. Liquid

Ovhd. Gas

Mass Flow,

_______

_______ ________ _______

_______

_______

Temperature

_______

_______ ________ _______

_______

_______

Pressure Composition, ______ % H2 N2 H2S H2O C1 C2 C3/C3= iC4 nC4/C4= iC5 nC5 C6+ Avg. Mol. Wt. Gravity Distillation

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

Steam to Stripper

_______

_______ ________ _______

_______

_______

Flash Point

_______

_______ ________ _______

_______

_______

IBP 5% 10% 20% 30% 40% 50% 60% 70% 80% 90% 95% EP


COLUMN SUMMARY

page _______________________________ date _______________________________ Item No.: _____________________________ by _______________________________ Service: __________________________________________________________________ Type of Operation: _________________________________________________________ No. of Trays: _________________ Reflux Ratio:_________________________________ Type of Trays: _____________________________________________________________ Net Off Ovhd. Feed Reflux Gas Btms. Liquid Other Mass Flow

_______

_______

_______

_______

_______ ________

Temperature Pressure

_______ _______

_______ _______

_______ _______

_______ _______

_______ ________ _______ ________

Composition, ______ % _______ _______ H2 N2 _______ H2S _______ H2O _______ C1 _______ _______ C2 C3 _______ iC4 _______ _______ nC4 iC5 _______ _______ nC5 C6+ _______ Avg. Mol. Wt. _______ Gravity _______ Distillation _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

IBP 5% 10% 20% 30% 40% 50% 60% 70% 80% 90% 95% EP

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

(Sketch system showing flows, P, T, Q on separate page) __________________________ Weight balance _______________________ Heat balance _________________________ Deviations from UOP Specifications: ___________________________________________


ABSORBER SUMMARY

page _______________________________ date _______________________________ Item No.: ____________________________ by _______________________________ Service: _________________________________________________________________ Type of Operation: _________________________________________________________ No. of Trays: ________________ Reflux Ratio: ________________________________ Type of Trays: ____________________________________________________________ Gas In

Liquid In

Gas Out

Liquid Out

Pumparound Upper Lower

Mass Fl Temperature

_______ _______

_______ ________ _______ _______ ________ _______

_______ _______

_______ _______

Pressure

_______

_______ ________ _______

_______

_______

Composition, ______ % H2 N2 H2S H2O C1 C2 C3 iC4 nC4 iC5 nC5 C6+ Avg. Mol. Wt. Gravity Distillation, ° _______ IBP 5% 10% 20% 30% 40% 50% 60% 70% 80% 90% 95% EP

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________ ________

_______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______ _______

(Sketch system showing flows, P, T, Q on separate page) __________________________ Weight balance ______________________ Heat balance ________________________ Deviations from UOP Specifications: ___________________________________________


CENTRIFUGAL COMPRESSOR DATA page _______________________________ date _______________________________ Item No.: _____________________________ by _______________________________ Service: __________________________________________________________________ Manufacturer: _____________________________________________________________ Type, Model: ______________________________________________________________ No. of Stages: _____________________________________________________________ OPERATING CONDITIONS/PERFORMANCE Flow Rate: ____________ Suction Pressure: ____________ psig Discharge Pressure: ____________ psig Differential Head: ____________ Polytropic : ____________ Operating Speed: ____________ rpm

Suction Temperature: Discharge Temperature: Power: MW:

________ ________ ________ ________

°F °F hp

Type of Seal: _________________________________________ Lube/Seal Oil System: ________________________________________ Buffer Gas: (yes/no) Buffer Gas Rate: ______________ SCFH Automatic Surge Control: (yes/no) DRIVER Motor Manufacturer: _________________________________________ Rating: ____________________ Service Factor: ______________ Insulation Class: ____________________ Voltage/phase/cycle: Turbine Manufacturer:_________________________________________ Speed: ______________ Steam Supply: _______ psig Steam Rate: ______________ Steam Exhaust: _______ psig Gear Manufacturer: Rating: Type:

______ °F ______ °F

_________________________________________ ____________________ Service Factor: _____________ ____________________ Power Loss: _____________

Deviations from UOP Specification: ____________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________


RECIPROCATING COMPRESSOR DATA page ___________________________ date ___________________________ by ___________________________

Item No.: ______________________________ Service: _______________________________ Manufacturer: ___________________________ Cylinder Lubrication: ____________ Type, Model: ___________________________ Clearance Pockets: (yes/no) No. of Stages, No. of Cylinders: ___________ Sparing Description: ____________ OPERATING CONDITIONS/PERFORMANCE Flow Rate: ____________ Suction Temperature: _________ °F Suction Pressure: ____________ psig Discharge Temperature: _________ °F Discharge Pressure: ____________ psig HP/stage: _________ hp MW: ____________

Operating Speed: ____________ rpm Cylinder Diameters: _________ Piston Speed: ____________ ft/s # of Suction/Discharge Valves: _________ Actual Rod Loadings, T/C: ________________________________________ lbf Max Allowable Rod Loadings, T/C: ________________________________________ lbf DRIVER Motor Manufacturer: Rating: Insulation Class:

________________________________________ ____________________ Service Factor: _____________ ____________________ Voltage/phase/cycle:

Turbine Manufacturer: ________________________________________ Speed: _______________ Steam Supply: _______ psig Steam Rate: _______________ Steam Exhaust: _______ psig Gear Manufacturer: Rating: Type:

______ °F ______ °F

________________________________________ ____________________ Service Factor: _____________ ____________________ Power Loss: _____________

Deviations from UOP Specification: ___________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________


CONTROL VALVE SUMMARY page ___________________________ date ___________________________ Item No.: ______________________________

by ___________________________

Service: __________________________________________________________________ Description of Valve: _____________________

Design CV: ______________________

Mfgr. and Catalog No.: ______________________________________________________ Positioner? _______________________________________________________________

Actual

Design

Percent open (valve position)

____________

Flow rate:

______________________

____________

__________

Upstream pressure:

______________________

____________

__________

Downstream pressure: ______________________

____________

__________

Flowing temperature:

____________

__________

______________________

Deviations from UOP Specification: ____________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________


AIR FIN COOLER SURVEY

page _______________________________ date _______________________________ Item No.: ____________________________ by _______________________________ Service: _________________________________________________________________ Manufacturer: _____________________________________________________________ Type, Model: _____________________________________________________________ No. of Bundles: _______________________ No. of Passes: _____________________ No. of Tubes per Pass: _________________ Fans/bundle: ______________________ Tube Size _______________ ID x _______________ Gauge x _____________ Length Piping Geometry: ______________________ Type*: ____________________________ Overall Heat Transfer Coefficient: _____________________________________________ Inlet Outlet Air

In Out No. fans on__________________________ Louver position_______________________ Mass flow Q (calc.) Composition, ____ % H2 N2 H2S H2O C1 C2 C3 iC4 nC4 iC5 nC5 C6+ Avg. Mol. Wt. Relative Humidity

Pressure ______________ ______________

Temperature _____________ _____________

______________ _____________ ______________ _____________ Pitch control ________________________ Air ______________ ______________

______________

Process _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________ _____________


Gravity Distillation, ° ______ IBP 10% 30% 50% 70% 90% EP

_____________ _____________ _____________ _____________ _____________ _____________ _____________

Deviations from UOP Specification: ____________________________________________ ________________________________________________________________________ ________________________________________________________________________ *Include sketch of piping geometry if different from UOP standard practice types.


FLOW METER SUMMARY page ___________________________ date ___________________________ Item No.: ______________________________

by ___________________________

Service: _________________________________________________________________ Type of Fluid: ___________________________

Normal Units of Flow: ______________

______________________________________ Type of Meter: ____________________________________________________________

Meter Reading: ___________________________________________________________ Pressure

________________

Temperature

________________

Sp. Gr.**

________________

Meter Factor

________________

Corrected Flow Rate

________________

Mass Flow Rate

________________

Avg. mol. wt.

________________

Molar Flow Rate

________________

**Sketch piping layout, showing distances in nominal pipe IDs.


HEAT EXCHANGER SURVEY

page _______________________________ date _______________________________ Item No.: _____________________________ by _______________________________ Service: __________________________________________________________________ Manufacturer: _____________________________________________________________ Type, Model: ______________________________________________________________ No. of Bundles: ____________________________________________________________ No. of Passes/Bundle: __________________ Tubes per Pass: ____________________ Tube Size ______________ ID x _______________ Gauge x ______________ Length Heat Exchange Surface Area/Bundle: __________________________________________ Piping Geometry (sketch if necessary): _________________________________________ Length of Service: __________________________________________________________ Design Heat Transfer Coefficient: ______________________________________________ Shell Side

Inlet

Stream A

Outlet Tube Side

Inlet Outlet

B

Q (calc.) Shell side Q (calc.) Tube side

______________ ______________

Composition, ______ % H2 N2 H2S H2O C1 C2 C3 iC4 nC4 iC5 nC5 C6+ Mass Flow Avg. Mol. Wt.

A ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________

Pressure ______________

Temperature _____________

______________

_____________

______________ ______________

_____________ _____________

B ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________ ______________


Gravity Distillation, ° _______ IBP 10% 30% 50% 70% 90% EP

______________ ______________ ______________ ______________ ______________ ______________ ______________

______________ ______________ ______________ ______________ ______________ ______________ ______________

Deviations from UOP Specification: ___________________________________________ ________________________________________________________________________ ________________________________________________________________________


HEATER SURVEY page _______________________________ date _______________________________ Item No.: _____________________________ by _______________________________ Service: __________________________________________________________________ Manufacturer: _____________________________________________________________ Type, Model: ______________________________________________________________ No. of Passes:_________________________ Tubes per Pass: ____________________ Tube Size ______________ ID x _______________ Wall x ________________ Length Geometry (Process): ________________________________________________________ Geometry (Flue Gas): _______________________________________________________

Radiant

Inlet

Stream A

Outlet

Pressure ______________


______________

_____________

Convection I

Inlet Outlet

B

______________ ______________

_____________ _____________

Convection II

Inlet Outlet

C

______________ ______________

_____________ _____________

Convection III

Inlet Outlet

D

______________ ______________

_____________ _____________

Fuel Gas

E

______________

_____________

Fuel Oil

F

______________

_____________

Flue Gas Under Convection I

G

______________

_____________

Flue Gas Under Convection II

H

______________

_____________

Flue Gas Under Convection III

I

______________

_____________

Flue Gas Under Stack Damper

J

______________

_____________

Flue Gas Above Floor

K

______________

_____________


HEATER SURVEY page _______________________________ date _______________________________ by _______________________________

Stream A B Mass Flow, ________ _____ _____ Composition, ______ % _____ _____ H2 _____ _____ _____ _____ N2 _____ _____ O2 CO _____ _____ _____ _____ CO2 _____ _____ H2S _____ _____ SO2 _____ _____ C1 _____ _____ C2 _____ _____ C3 _____ _____ iC4 _____ _____ nC4 _____ _____ iC5 _____ _____ nC5 _____ _____ C6-205°C (400°F) 205°C (400°F)+ _____ _____ Avg. Mol. Wt. _____ _____ Gravity _____ _____ Viscosity _____ _____ Total Sulfur, _______ _____ _____ Metals, ___________ _____ _____ Q (calc.) Absorbed _____ _____ Q (calc.) Released Heater Gross Efficiency Excess Air, % Tube Skin Temps:,° _____ Burner Pressure ______________________

C _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____

D _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ ____

E _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____

F _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____ _____

_____

_____

G,H, I,J,K ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______ ______

% of Rating ____________________

Provide sketch showing piping and controls for process piping. Deviations from UOP Specification: ___________________________________________ ________________________________________________________________________ ________________________________________________________________________


CENTRIFUGAL PUMP SURVEY page _______________________________ date _______________________________ Item No.: _____________________________ by _______________________________ Service: __________________________________________________________________ Manufacturer: _____________________________________________________________ Type, Model: ______________________________________________________________ No., Size and Style (Mfgrs. Designation) ________________________________________ ________________________________________________________________________ Suction

Pressure ______________

Discharge

______________


Other Information Rated Flow (STP) _____________ Seal Type? Single, Tandem, Double, Bellow Sp. Gr. _____________ Spillback? Yes/No Viscosity _____________ NPSHR? _________________________ Static Suction Head _____________ Suction Specific Speed: ________________ Speed _____________ Differential Head (flowing condition) _________________________________________ Driver Type: ___________________________________________________________ Manufacturer: ___________________________________________________________ No., Size, Rating and Style (Mfgrs. designation): __________________________________ Rating: _________________ Insulation Class: _________________ Service Factor: _________________ Voltage/Phase/Cycle: _________________ Motor: Power consumption Speed Turbine: Steam consumption Steam supply

______________ ______________ ______________

Steam exhaust Speed

Pressure ______________

Temperature ______________

______________

______________

______________

Supply copy of Mfgrs. pump curve and plot operating point. Deviations from UOP Specification: ____________________________________________ ________________________________________________________________________ ________________________________________________________________________



SUPPLEMENTAL CALCULATIONS REGENERATOR VELOCITIES The following procedure shows how to calculate the superficial velocity in the combustor, upper regenerator, and cyclones. This section presents two methods to calculate the velocities in the regenerator. The first method is more precise but requires more information and it is more laborious than the second method.

Method A 1.

Required Information

The FCC Unit is in total combustion mode for this case and with no catalyst cooler. The process conditions are: Temperatures:

Average Dense Average Dilute Average Cyclones Air to Regenerator Ambient Relative Humidity

1371°F 1375°F 1375°F 399°F 62°F 97%

Pressures:

Regenerator Combustor Cyclones

32 psig 34 psig 31 psig

Areas:

Combustor Cross sectional Regenerator Cross sectional First Stage Cyclones Second Stage Cyclones

300 ft2 452 ft2 24.5 ft2 21.3 ft2


Flow Rates:

Flue Gas:

Air to Regenerator 83,615 scfm = (scfm x 28.76 lb/mol x 60 min/hr)/(379.5 scf/mol) = 380,200 lb/hr CO CO2 O2 N2 + Ar SO2 NO2

2.

= = = = = =

0 15.50 2.5 82.0 0 0

Combustion Air Correction to a Dry Basis

A psychometric chart is used to determine the moisture content of the air. At atmospheric conditions of 62°F and a relative humidity of 97%, the moisture content is: Moisture Content =

0.01152 lb H2O lb dry air

Wet Air = 380,200 lb/hr Dry Air = 380, 200 lb/hr wet air 

1 lb dry air (1 + 0.01152) lb wet air

Water in Air = 380,200 lb/hr - 375,870 lb/hr = 4,330 lb/hr

= 375, 870 lb/hr


3.

Calculate Flue Gas Rate

The flue gas rate can be calculated from the regenerator air rate. These two streams are related by the inert N2 + Ar content which remains constant through the catalyst regeneration. Since, moles =

Weight Molecular Weight

then, Water in Air = (4,330 lb/hr)/(18 MW) = 241 mol/hr Dry Air = (375,870 lb/hr)/(28.966 lb/mol) = 12,976 lb mol/hr mol/hr (N2 + Ar) in dry air = mol/hr (N2 + Ar) in flue gas 79 mol inerts lb mol FG 82 mol inerts 12, 976 lb mol  =  100 mol air hr 100 mol FG hr

Flue Gas (FG) = 12,501 lb mol/hr

4.

Calculate the Water Produced by the Hydrogen Content of Coke

The overall reaction occurring in the regenerator is: C + H2 + S + N + O2 = CO + SO2 + NO2 + H2O + O2


The water produced by the hydrogen (H2) content of the coke can be calculated from an O2 balance: O2 in regeneration air = excess O2 in flue gas + + O2 reacted to CO (0.5 mol O2/mol CO) + O2 reacted to CO2 (1 mol O2/mol CO2) + O2 reacted to H2O (0.5 mol O2/mol H2O) + O2 reacted to SO2 (1 mol O2/mol SO2) + O2 reacted to NO2 1 mol O2/mol NO2 where: O2 in regen. air =

12, 976 lb mol dry air 21 mol O 2 2,725 lb mol  = of O2 hr hr 100 mol air

Excess O2 in FG =

O2 reacted to CO =

12, 501 lb mol FG 2.5 mol O2 312 lb mol  = of O2 hr hr 100 mol FG

12, 501 lb mol FG 0 mol CO 0.5 mol O2   = 0 lb mol/hr O2 hr 100 mol FG mol CO

O2 reacted to CO2 =

12, 501 lb mol FG 15.5 mol CO 2 1 mol O2   =1,938 lb mol/hr of O2 hr 100 mol FG mol CO2

O2 reacted to SO2 =

12, 501 lb mol FG 0 mol SO2 1 mol O2   = 0 lb mol/hr of O2 hr 100 mol FG mol SO2

O2 reacted to NO2 =

12, 501 lb mol FG 0 mol NO2 1 mol O2   = 0 lb mol/hr of O2 hr 100 mol FG mol NO 2


O2 reacted to H2O (by difference) is: O2 Reacted to H2O = 2,725 - 312 - 0 - 1,938 - 0 - 0 lb mol/hr O2 = 475 lb mol/hr of O2 Since H2 + 1/2O2 = H2O Then The water produced by Hydrogen and Oxygen in the regenerator is: H2 O Produced by O2 =

5.

475 lb mol 2 mol H2O 950 lb mol  = hr O2 mol O2 hr H2

Calculate the Wet Flue Gas Rate

The total moles per hour of wet flue gas are: Wet Flue Gas = Dry Flue Gas + Water from Air + Water from H2 in Coke = 12,502 mol/hr + 241 mol/hr + 949 mol/hr = 13,692 mol/hr The actual cubic feet per second (ACFS) of the flue gas can be calculated by using the Ideal Gas equation of state PV = nRT Where R Prg Pcomb Pcycl Tcomb Tdense Tdilute Tcycl

then

= = = = = = = =

V = nRT/P

10.7 (psia x ft3)/(mol x °R) 32 psig + 14.7 = 46.7 psia 34 psig + 14.7 = 48.7 psia 31 psig + 14.7 = 45.7 psia 1,275°F + 460 = 1,735 °R 1,371°F + 460 = 1,831 °R 1,375°F + 460 = 1,835 °R 1,368°F + 460 = 1,828 °R


ACFS @Tcomb = 13,692 mol/hr x 10.73 x 1,735/(48.7x3600 sec/hr) = 1,454 ft3/sec ACFS @Tdens

= 13,692 mol/hr x 10.73 x 1,831/(46.7x3600 sec/hr) = 1,596 ft3/sec

ACFS @Tdilute = 13,692 mol/hr x 10.73 x 1,835/(46.7x3600 sec/hr) = 1,599 ft3/sec ACFS @Tcyc

6.

= 13,692 mol/hr x 10.73 x 1,828/(45.7x3600 sec/hr) = 1,632 ft3/sec

Calculate Superficial Velocities

Regenerator Superficial Velocity = (ACFS @ Tdens, ft3/sec) / (Rg Cross Sect Area, ft2)

= 3.5 ft/sec

First stage Cyclones Superficial Velocity = (ACFS @ Tdilute ft3/sec) / (Total Inlet Area, ft2)

= 65.2 ft/sec

Second Stage Cyclones Superficial Velocity = (ACFS @ Tcyc, ft3/sec) / (Total Inlet Area, ft2)

= 76.6 ft/sec

Combustor Superficial Velocity = (ACFS @ Tcomb, ft3/sec) / (Comb Cross Sec Area, ft2)

= 4.8 ft/sec


Method B 1.

Required Information

This method does not require the flue gas analysis. The process conditions are: Temperatures: Pressures: Area: Flow Rates:

2.

Average Dense Regenerator Regenerator Cross sectional Air to Regenerator

1371°F 32 psig 452 ft2 83,615 scfm

Calculate the Actual Cubic Feet per Second

The volumetric flue gas rate can be calculated by using the Ideal Gas equation of state PV = nRT

then

R = PV/nT

For air we have

R = P1V1/n1T1

For the Regenerator Air

R = P2V2/n2T2

Combining the last two equations

P1V1/n1T1 = P2V2/n2T2

Or

V2 = P1V1T2 x n2 T1P2 n1

Where: P1 T1 V1 P2 T2

= = = = =

0 pisg + 14.7 60 °F + 460 83,615 scfm/(60 s/m) 32 pisg + 14.7 1,371°F + 460

= = = = =

14.7 psia 520 °R 1,393.6 ft/s 46.7 psia 1,831 °R


n2/n1

=

1.04 Assumed. This factor is due to the combustion of Hydrogen to in the coke to water.

Then V2

=

(14.7 psia)(1,393.5 ft/s)(1,831°R)(1.04) = 1,606 ft3/s (520°R)(46.7 psia)

3.

Calculate Regenerator Superficial Velocity

The superficial velocity is calculated by dividing the volumetric flow rate by the cross sectional area: 1,606 ft3/s = 3.6 ft/s 2 452 ft The molar expansion factor n2/n1 can be approximated if the flue analysis is available by using the following equation:

Rg Velocity

=

n2/n1

=

2 - (79/N2%)

Where N2% is the Nitrogen percent form the flue gas analysis.


REGENERATOR AIR DISTRIBUTOR PRESSURE DROP The pressure drop across the regenerator air distributor can be calculated by the following formula:

P =

q2 C2 A2 (2g) 144

where: ∆P q r C A g

= = = = = =

pressure drop, psi air rate at flowing conditions, ft3/sec density of air at flowing conditions, lb/ft3 orifice coefficient, 0.60-0.80 total cross sectional area of holes, ft2 acceleration due to gravity, 32.2 ft/sec2

The perforated grid air distributor is designed for a pressure differential of about 0.71.2 psi. This will give good air distribution for fluidization without causing catalyst attrition. If the pressure differential is too high, the high velocity can cause attrition. If the pressure differential is too low, less than 0.5 psi, it can cause poor distribution of air and distributor erosion problems.


REACTOR STRIPPER DENSITY The following procedure shows how to calculate the density in Reactor Stripper. The density indicator is a differential type instrument and for this case the range is 0-145 inches of water.

1.

Calculate the Distance Between Taps

Lower Instrument Tap Elevation: 59' 8 7/8" or 59.7396' Upper Instrument Tap Elevation: 74' 4 1/8" or 74.3438' Distance Between Taps: 74' 4 1/8" - 59' 8 7/8" = 14' 7 1/4" or 175.25"

2.

Calculate the Density Instrument Readout: 75% Instrument Span: 0-145 inches H2O 75% x 145 inches H2O = 108.75 inches H2O 108.75 in H 2 O 

1 12 in lb/in 2 144 in 2 lb    = 38.7 3 2 175.25 in ft 27.705 in H 2 O ft ft


Notes: i)

The Spent Catalyst Stripper Density is expected to range from 30 to 45 lb/ft3.

ii)

In cases when the pressure taps are under the stripper baffles the distance between taps should be replaced by the distance between the bottom edges of the baffles.


REACTOR STRIPPER LEVEL This procedure shows how to calculate the level in the reactor. The level controller is a differential type instrument and for this case the range is 0-300 inches of water.

1.

Calculate the Distance Between the Taps

Lower Instrument Tap Elevation: 62' 3 7/8" or 62.3229' Upper Instrument Tap Location: 127' 0" - 3' 3" - 1' 6" = 122' 3" or 122.2500' Distance Between Upper Tap and Lower Tap: 122' 6" - 62' 3 7/8" = 60' 2 1/8" or 60.1771'

2.

Data Required Instrument Readout: Instrument Span:

50% 0-300 inches H2O

Stripper Density of 36.25 lb/ft3 (From Stripper Density Calculation) Assume Reactor Vapor Space Density of 1 lb/ft3 Distance Between Upper Tap and Lower Tap = 60.1771' Elevation of Lower Tap = 62.3229' Normal Reactor Catalyst Level = 84.500' Cyclone Dipleg Outlet = 79.5417' 300 in H2O x 50% = 150 in H2O


3.

Method A - This is rough method. lb/in 2 144 in 2 ft 3 150 in H 2 O    = 21.51 ft of Catalyst 27.705 in H 2 O 36.25 lb ft 2

Distance Relative to Normal Catalyst Level = (62.32 + 21.51) - 84.50 = - 0.67 (i.e. ~ 8" below normal level)

4. Method B - This method is more precise than Method A since consider the reactor vapor density. X + Y = 60.18 ft => Y = 60.18 - X Where:

X = Catalyst height. y = Reactor vapor height from catalyst bed to upper pressure tap. 60.18 ft = distance between pressure taps.

lb lb/in 2 144 in 2 X (36.25 lb)   + (60.18 - X)  1 3 150 in H 2 O  2 3 27.705 in H 2 O ft ft ft

779.64 lb/ft2 = 35.25 X lb/ft3 + 60.18 lb/ft2 Then X = 20.41 ft above Lower Level Tap Distance Relative to Normal Catalyst Level = (62.32 + 20.41) - 84.50 = - 1.77 (i.e. 1' 9 1/4" below normal level)


Notes: i)

The Spent Catalyst Stripper Density is expected to range from 30 to 45 Lbs/Ft3.

ii)

In cases when the lower pressure tap is under the stripper baffle the distance between taps should be replaced by the distance between the bottom edge of the baffle and the upper tap.


REACTOR RISER RESIDENCE TIME Residence time, the time that hydrocarbons spend in the riser, is another design variable utilized in controlling reaction severity. This variable is of particular importance in operation using high activity zeolitic catalyst. Typical residence time for current designs is two to three seconds. Conversion is proportional to residence time in that it increases with prolonged contact of catalyst and feedstock. Gasoline yields increase with residence time up to a point after which over-cracking may occur. This results in a loss of gasoline yields and a significant increase in conversion. The method used to calculate riser residence time is as follows: 

= VR/[(1/3)(VF) + (2/3) (VP)]



= Residence time in seconds

VR = Riser volume, ft3 VF

= Volume of vaporized feed, steam, lift gas, water, and inerts calculated at the average conditions at the point of feed injection, ft3/s.

VP

= Volume of vaporized products, steam, lift gas, water and inerts calculated at the average conditions at the point of feed injection, ft3/s.

It should be noted that prior to the residence time calculation, the average temperature and pressure at the point of feed injection must be estimated to obtain VF and VP.


The riser pressure at the point of feed injection can be approximated by assuming a 5 psig pressure drop, hence: Riser Pressure = Reactor dome pressure + 5 psi The average temperature at the point of feed injection is calculated as:

Tavg 

0.2753C / O TR   CPo To   L / O CPL TL   0.495W / O TW   S / O TS   W / O HWL   H Rx   H OL  0.2753C / O   CPo   0.495W / O   S / O   L / O CPL 

Where: C/O

=

Catalyst to Oil wt. ratio, calculated in reactor-regenerator heat balance section

S/O

=

Steam to Oil wt. ratio

W/O

=

Water to Oil wt. ratio

L/O

=

Lift gas to Oil wt. ratio

TR

=

Regenerator dense bed temperature, °F

TO

=

Oil feed temperature, °F

TL

=

Lift gas feed temperature, °F

TS

=

Steam feed temperature, °F

TW

=

Water feed temperature, °F

CPO

=

Specific heat of vaporized oil feed, Btu/lb/°F


CPL

=

Specific heat of lift gas, Btu/lb/°F

∆HOL

=

Latent heat of vaporization of oil feed at inlet temperature, To, Btu/lb

∆HWL

=

Latent heat of vaporization of water at inlet temperature, Tw, Btu/lb

∆HRX

=

Heat of reaction, Btu/lb, calculated in reactor-regenerator heat balance section

Tavg

=

Average temperature at point of feed injection, °F


HYDROGEN BALANCE This document describes a manual method for calculating the Hydrogen balance for an FCC Unit.

Data required: A normalized to 100% recovery product summary in wt-% A breakdown of the C4- components The distillation and API of each C5+ product The distillation and API of the feed API Technical Data Book Figure 2B1.1 "Characterizing Boiling Points of Petroleum Fractions" Figure 2 - UOP Chart 409B-12 "Hydrogen Content of Liquid Petroleum Hydrocarbons"

Description of the Calculation Method: Step 1. Determine the molecular weight of Hydrogen per molecule for each C4product, e.g. H2S, H2, C1, C2, C2=, etc. See column 3 of the attached example. Step 2. Determine the percentage Hydrogen in each C4- product component by dividing the molecular weight of Hydrogen per molecule by the molecular weight of each component. In the attached example this is column 3 divided by column 4 times 100. The result is given in column 5.

Step 3. Calculate the Volume Average Boiling Point (VABP) for each of the heavier products (C5+ gasoline, LCO, and MCB). See Figure 2B1.1 comments for procedure and definitions.

Step 4. Calculate the Engler Slope for each of the heavier products (C5+ gasoline, LCO, and MCB). See Figure 2B1.1 comments for procedure and definitions.


Step 5. Determine the Mean Average Boiling Point from API Figure 2B1.1. Step 6. Determine the Hydrogen content of the hydrocarbon liquid from Figure 2. See the lower half of column 5 in the attached example. Step 7. Multiply the wt% normalized yield pattern for each component by the percentage Hydrogen in each product component. In the attached example this is column 2 times column 5. The result is given in column 6. Step 8. Calculate the percentage of Feed Hydrogen in each component by dividing the Wt% Hydrogen in each product component by the Wt% H2 in the feed. In the attached example this is the value in column 6 divided by the Wt% H2 in the feed (13%).


Hydrogen Balance Summary 1

2

3

4

5

6

7

Mass Balance

wt-%

#H/Molecule

MW

%H2

H2 wt-%

% Feed H2

Results Feed:

100.00

Values from Figures 2B1.1 & Fig. 2

13.00

100.00

H2S

0.04

2.0158

34.08

0.06

0.0021

0.02

H2

0.23

2.0158

2.02

1.00

0.2280

1.75

C1

0.89

4.0361

16.04

0.25

0.2245

1.73

C2

0.81

6.0474

30.07

0.20

0.1622

1.25

C2=

0.91

4.0316

28.05

0.14

0.1303

1.00

C3

1.30

8.0632

44.10

0.18

0.2375

1.83

C3=

4.64

6.0474

42.08

0.14

0.6662

5.12

IC4

2.69

10.0790

58.12

0.17

0.4665

3.59

NC4

0.63

10.0790

58.12

0.17

0.1094

0.84

C4=

4.88

8.0632

56.11

0.14

0.7007

5.39

C5+ Gasoline

44.55

LCO

27.60

MCB Coke Total Products

13.80

6.15

47.29

Values to the right

11.30

3.12

23.99

5.22

were determined from Chart

10.20

0.53

4.10

5.62

2B1.1 & Fig 2.

4.44

0.25

1.92

39.74

12.98

99.82

100.00

See attached method.

Total

Laboratory Summary for Gasoline Distillation D-86

°F

°R

(°R)^1/3

SpGr

0.7286

IBP

96.8

556.8

---------

API

62.02 214.7

10%

131.9

591.9

8.3962

VABP

20%

145.4

605.4

8.4596

Engler Slope

2.40

30%

163.4

623.4

8.5426

Correction Factor

6.0

40%

182.3

642.3

8.6280

CABP, °R Uncorrected

672.8

50%

203.9

663.9

8.7237

CABP, °F Corrected

206.8

60%

230.9

690.9

8.8404

CABP, °R Corrected

666.8

70%

259.7

719.7

8.9616

UOP K

11.99

80%

291.2

751.2

9.0904

Total Sulfur, wt%

0.0032

90%

323.6

786.3

9.2193

Octane (F1 C)

92

EP

371.3

831.3

---------

RVP @ 378°C, kg/cm

Recovered Volume

99.5

%

0.5

%

Residue Volume

2

39.6


Hydrogen Content of Liquid Petroleum Hydrocarbons



Comments on Figure 2B1.1 Purpose: The various average boiling points which are used to characterize petroleum fractions are correlated in Figure 2B1.1 with the ASTM D86 distillation properties of the fraction. If these boiling points are required for mixtures (or portions of a mixture) for which the composition is known, using the defining equations (2-0.3) through (2-0.7) given in the introduction. Reliability: The reliability is unknown. Notation: The volumetric average boiling point of a petroleum fraction is the weighted average of the ASTM D86 distillation temperatures after 10, 30, 50, 70 and 90 percent by volume have been distilled.

T10  T30  T50  T70  T90  5

. The slope is calculated assuming a linear ASTM D86 distillation curve

 T90  T10   in degrees Fahrenheit per percent distilled.  90  10 

between the 10 and 90 percent points 

The relationships between the various average boiling points given in Figure 2B1.1 for petroleum fractions are analogous to those defined by equations (2-0.3) through (2-0.7) for mixtures of identifiable hydrocarbons. Special Comments: For ASTM D86 distillation temperatures above 475°F, use the following correction for cracking: log D  1.587  0.00473  T (2B1.1-1) Where: D = correction to be added to T, in degrees Fahrenheit T = observed distillation temperature, in degrees Fahrenheit If the available distillation data are not from ASTM Method D86, they must be converted by the methods of Chapter 3 to calculate the volumetric average boiling point. Literature Sources: This figure was developed by Smith and Watson, Ind. Eng. Chem. 29 1408 (1937). Equation (2B1.1-1) was given by S.T.Hadden, Gulf Research and Development Company, Pittsburgh, Pa., private communication (1964). Example: Determine the molal average boiling point, weighted average boiling point, cubic average boiling point, and mean average boiling point of a petroleum fraction having the following ASTM D86 distillation properties: Distillation, percent by volume 10 30 50 70 90 Temperature, degrees Fahrenheit: 149 230 282 325 371

VABP 

149  230  282  325  371  271 F 5

Slope 

371  149  2.78  F % 80

Using Figure 2B1.1, the average boiling points are calculated from the volumetric average boiling point:

MABP  271  30  241 F WABP  271  7  278 F

CABP  271  7  264 F MeABP  271  19  252 F


Calculation of Flow Meter Constant “K” for Liquid Flow:  Gf   Q max  N  S  D 2  Fa  Fc  hm    Gb    Where: Qmax N D d S Fa Fc Gf Gb hm Note:

= = = = = = = = = =

maximum flow rate at base conditions, (@ 60°F) constant based on flow units Process pipe inside diameter orifice diameter discharge coefficient, f(d/D) thermal expansion of plate Reynolds correction factor liquid specific gravity at flowing conditions gas specific gravity at base condition (@ 60°F) maximum differential pressure (design basis)

N, S, Fa, and Fc can be found in L.Spink’s “Principles and Practices of Flow Meter Engineering”

Example for Sponge Gas Meter Flow, bpsd N, for bpd D d B = d/D S Fa Viscosity, cS Re Fc hm Temp, °F Gb Gf

= = = = = = = = = = = = = =

34,000 194.3 7.981 in 5.034 in 0.6307 0.2672 1.001 for type 304 stainless steel plate 10 39,292; Re = 92.235*bpsd / (viscosity*D) 1.015 100 in H2O 173°F 0.9266 0.8854 Gf = Gb*VCF = 0.9260 * 0.9562 = 0.8854

 0.8854   Q max  194.3  0.2672  7.9812  1.001  1.015  100    0 . 9266  

Q max  34,119 BPSD The “K” constant can be calculated by using the following equation:

K

Q max* Gb   3,360 BPSD 10  Gf

Where: 10: maximum units meter reading, MR, then:

 Gf  ; Q  3,360  MR    Gb   

Q  _________ BPSD


Calculation of Flow Meter Constant “K” for Gas Flow: Pf Q max  N  S  D 2  Fa  Fc  Y hm  Tf  Gb  Z  Where: Qmax N D d S Fa Fc Y hm Pf Tf Gb Z Note:

= = = = = = = = = = = = =

maximum flow rate at base conditions, (@ 60°F) constant based on flow units Process pipe inside diameter orifice diameter discharge coefficient, f(d/D) thermal expansion of plate Reynolds correction factor upstream orifice expansion factor maximum differential pressure (design basis) absolute flowing pressure upstream of the orifice, psia absolute flowing temperature, °R gas specific gravity at base condition (@ 60°F) compressibility factor of gas

N, S, Fa, Fc, Y and Z can be found in L.Spink’s “Principles and Practices of Flow Meter Engineering”

Example for Sponge Gas Meter Flow, scfm N D d B = d/D S Fa Viscosity, cP Re Fc Y hm Pf Tf Cp/Cv Gb Z

= = = = = = = = = = = = = = = = =

8,375 = lb/hr = 26,100 128.78 for scfm and psi 6.065 in 3.5834 in 0.5908 0.2284 1.0003, for type 304 stainless steel plate 0.011 2,472,487; Re = 6.32(lb/hr)/(viscosity*D) 0.988 0.9944; Y = 1-(0.41-0.35B^4)(h/2)/(27.67Pf*(p/Cv)) 200 in H2O 187.7 psia 573°R 1.27 0.7054 0.965 f(Tr,Pr) Tr = Tf/Tc Pr = Pf/Pc

Q max  128.78  0.2284  6.065 2  1.0003  0.9881  200 

Q max  10,490scfm The “K” constant can be calculated by using the following equation:

 Gb   Q max    1,539scm K    Tf    10   Pf  Where: 10: maximum units meter reading, MR, then:

Q  1539  MR 

Pf  Gb ; Tf

Q  _________ scfm

187.7 573  0.7054  0.965

157048 Treating Page 1

FEED/PRODUCT TREATING INTRODUCTION FCC feeds contain a number of contaminants that affect yields, product quality, plant emissions and corrosion in the main column and gas concentration unit. These contaminants are handled by a combination of treating either the feed or products as well as unit design. This subject is of increasing importance to refiners with the ever tighter limits on emissions from the plant, especially SOx and NOx and on limits in liquid product sulfur levels. The industry trend towards processing resid feeds which typically have higher concentrations of sulfur, metals and carbon residue makes this issue even more important. In the United States new fuel specifications are will limit the sulfur concentration in both gasoline and high speed diesel fuels to less than 50 wppm. This is an issue critical to the FCC because in a typical refinery gasoline pool 98% of the total gasoline pool sulfur comes from the FCC naphtha even though the FCC naphtha makes up only 30-40% of the pool. Limits on fuel oil sulfur will also require a reduction in the MCB product sulfur. In some areas the limits on SOx emissions will be more restrictive requiring less than 300 ppm in the flue gas. In the coming years these restrictions will likely become even more stringent. FEED TREATING Light and heavy vacuum gas oils are the most common FCC feedstock with an increasing trend towards atmospheric resid. Also, there are economic incentives towards processing lower priced crudes which typically contain higher levels of contaminants. Hydrotreating is the most common and effective method of improving the FCC feed quality. Hydrotreating not only reduces the contaminant concentration but also improves yields. Hydrogen addition to the feed, especially to the large polynuclear aromatics, makes these molecules easier to crack resulting in higher conversion to desired products with less coke and light gas make. Table 1 shows the impact of hydrotreating on both the FCC feed properties and the FCC yields.


Table 1 Feed Hydrotreating Benefits Feed Desulfurization Feed Properties Gravity, ºAPI Sulfur, wt% Nitrogen, wppm Carbon Residue, wt% Metals (Ni + V), wppm Yields, wt% H2S C2LPG Naphtha LCO MCB Coke Conversion, lv% Key Product Properties Naphtha RONC Naphtha MONC LCO Cetane Index Product Sulfur, wppm H2S Naphtha LCO MCB SOx , vppm in flue gas

Untreated

90%

98%

99%

20.5 2.6 880 0.4 5

23.5 0.25 500 0.25 2

24.8 0.06 450 0.1 1

26.0 0.02 400 0.1 <1

1.1 3.3 16.3 48.3 16.7 9.0 5.4 74.3

0.1 3.5 17.6 51.5 15.7 6.6 5.0 77.7

0.0 3.2 18.7 52.5 15.0 5.9 4.7 79.1

0.0 2.8 19.9 53.6 14.0 5.2 4.4 80.8

93.2 80.5 25.7

93.0 80.8 25.7

92.9 81.1 26.4

92.7 81.0 26.5

10,100 3,600 29,700 57,800 2,000

750 230 3,400 11,000 410

190 55 900 3,000 120

95 18 300 1,100 42


As more sulfur is removed the sulfur balance in the FCC shifts towards higher percentage of the total feed sulfur going to the MCB product and SOx in the flue gas. This is because the hardest to remove sulfur is in the heavy aromatic compounds which tend to form coke or remain uncracked. From Table 1 the percentage of sulfur in the feed ending up in the MCB product increases from 20 to 30% and the percentage of feed sulfur ending up in the naphtha decreases from 6.7 to 4.8% as the untreated feed is desulfurized by 98%. Table 1 illustrates a case where hydrotreating is used primarily for reduction in sulfur. In some units, especially those treating resid feeds, the reduction of carbon residue and metals is more important. Hydrotreating can reduce feed metals and carbon residue by 90% or more allowing processing of extremely contaminated resid feeds in the FCC while still achieving good yields. This can also significantly reduce the required fresh catalyst addition rate. Nitrogen is also removed by feed hydrotreating. Some of the nitrogen in the FCC feed is converted to ammonia which can cause salt formation and plugging in the main column overhead. Basic nitrogen in the feed acts a temporary poison to the FCC catalyst. Severely contaminated FCC feeds may contain as much as 4,000 wppm nitrogen while a clean feed may contain less than 500 wppm. Cyanides are also formed from nitrogen in the feed which can cause blistering and corrosion in the gas concentration unit. Wash water and proper control of the main column overhead temperature is usually sufficient to prevent these plugging and corrosion concerns so that these are usually not a significant factor when considering hydrotreating as an option to improve profitability. Oxygen may be present in the feed either chemically bonded in the hydrocarbon or absorbed from storage. Dissolved oxygen may cause fouling of heat exchangers when the temperature approaches 400ºF (200ºC). The best method for preventing this is to blanket the raw oil tanks with either fuel gas or nitrogen. This is most important for cracked feed stocks such as coker gas oil or highly olefinic feeds. Once in the reactor the oxygen will quickly be converted to water, carbon oxides, phenols, creosols or acids.


PRODUCT TREATING

FCC products contain undesirable material which must be either removed or transformed into an inoffensive material if the product is to meet specifications for use. The cycle oil, gasoline, LPG and fuel gas must be non-corrosive, stable in storage and of acceptable odor. The specifications may also limit the allowable amount of contaminants such as sulfur. Feed treating is effective in reducing nearly all of these contaminants but it requires a very high capital investment. The major undesirable constituents found in FCC products are:

1.

Sulfur compounds. In the gasoline and lighter compounds there are particularly hydrogen sulfide, mercaptans, elemental sulfur, and carbonyl sulfide. While the regenerator flue gas is not considered a product of the FCC, reduction of SOx emissions from this stream is of increasing importance to most refiners. SOx control is covered in more detail in the Environmental section of this manual.

2.

Oil soluble or ionic metals, principally copper.

3.

Nitrogen containing compounds, such as pyridine, quinoline, and pyrrole. NOx in the regenerator flue gas is also a common concern.

4.

Oxygenated compounds, including phenols and carbonic acids.

5.

Diolefinic hydrocarbons.

The degree of undesirability is modified and varies between groups of substances as well as within families of compounds, depending upon the concentrations present and the required product specifications. Synergistic and inhibiting relationships are known to exist, as between thiophenols and olefins which intensify gum formation, or between elemental sulfur and hydrogen sulfide which promote copper strip corrosion. The distribution and concentration of objectionable


components in particular product are governed by FCC charge properties, catalyst type and activity, the promotion of thermal cracking reactions, and fractionation schemes. The following sections will discuss the impurities and what may be done to render them harmless. SULFUR There are a variety of different sulfur compounds present in FCC product streams. They vary in their potential damage, but overall rate as the most troublesome impurity. Hydrotreating each product stream would remove them along with other contaminants, but from both a process and cost basis this technique is rarely used for LPG or gasoline treating. Hydrotreating is occasionally used for cycle oil treating. Elemental Sulfur Elemental sulfur may occur naturally, but because it is non-volatile it does not distill to any product above the feed tray. Elemental sulfur formation can be a problem in the gas concentration unit when oxygen in present. The most common source of the oxygen is the wash water injected in the wet gas compressor interstage. It is also possible that during startup catalyst circulation in the FCC oxygen will be entrained into the reactor and enter the gas concentration unit which can cause elemental sulfur formation for a short period after startup. The best solution to minimize elemental sulfur formation is to eliminate all possible sources of oxygen including using steam condensate for the wash water. Elemental sulfur can also be found in the FCC gasoline or LPG products if there is carry over of caustic from the Merox or other treating units. Elemental, sometimes called free sulfur corrodes copper and consequently produces a positive copper strip corrosion test. In the absence of H2S, as little as 5 wt. ppm free sulfur can give a failing copper strip test. If H2S is present, even in very low concentrations, the amount of allowable sulfur is even less. When 0.3 wt. ppm H2S is present, 0.5 wt. ppm elemental sulfur can give a filling copper strip test.


Elemental sulfur has a harmful effect on alkyl lead susceptibility for gasoline octane improvement. Free sulfur will remain as a deposit when LPG is evaporated. Removal of elemental sulfur is very difficult; hydrotreating or redistillation are two costly removal methods. Chemical treatments such as scrubbing with sodium or potassium polysulfide solutions or caustic-sodium sulfide, or reverse doctor treating may be used but may not be completely effective. Certain additives may be helpful in masking the sulfur corrosivity test. They should be used with caution. The best treatment method is to avoid elemental sulfur formation by preventing H2S oxidation. Hydrogen Sulfide Hydrogen sulfide results from the decomposition of sulfur compounds during the cracking reaction and concentrates in the fuel gas and LPG products. It has an obnoxious odor at low concentration, and is very poisonous. It paralyzes the involuntary breathing function, leading to asphyxiation, unconsciousness and death. Hydrogen sulfide in fuel gas will burn in a heater to form SO2. This acid gas may cause corrosion and environmental problems due to its acidity when it dissolves in water. In the presence of oxygen, H2S may oxidize in a product stream to form elemental sulfur. H2S has a deleterious effect on lead susceptibility of FCC gasoline. It promotes peroxide formation and prevents oxidation inhibitors from functioning, thus reducing gasoline stability in storage. It is undesirable in HF acid alkylation feedstocks as one pound of sulfur consumes about five pounds of process acid, and H2S, along with some mercaptans, is sufficiently volatile to stay with the C3 and C4 streams.


Dilute caustic soda or soda ash will remove H2S as well as CO2. The aqueous caustic soda solution must be sufficiently dilute to prevent formation of sodium sulfide crystals as the NaOH is converted to Na2S, and in the case of a dry feed, as the solution dehydrates. Normal caustic concentrations are in the range of 10-15° Be to avoid Na2S crystallization. The other principal treating method for the removal of H2S is regenerative monoethanolamine (MEA) or diethanolamine (DEA) scrubbing. The stream to be treated is contacted countercurrently with the amine to remove H2S, CO2, and COS. The amine is then steam stripped to regenerate it. Although this treatment can reduce H2S concentrations below 6 wt. ppm, it is equilibrium limited and therefore it is normal to provide a dilute batch caustic scrubber following the amine scrubber as a final cleanup and guard in case of an amine unit upset, e.g., when treating LPG. Aside from pollution abatement benefits when operated in tandem with a sulfur recovery process, amine treating usually becomes economically attractive strictly on a chemical consumption basis when H2S levels exceed roughly 1000 ppm. Mercaptans (RSH) Mercaptans are found in all cracked products streams. Some of these may be naturally occurring in the feed, but most mercaptans are formed during the cracking reactions. Decomposition of other sulfur compounds or recombination reactions between H2S and olefins are the major sources. There are two distinct types: alkyl, open chain mercaptans, and aryl, aromatic mercaptans in which the •SH group is linked to a benzene ring structure. The aryl mercaptans are also known as thiophenols and thiocresols. They tend to predominate over alkyl mercaptans in FCC product streams boiling over 300°F (150°C). The lower molecular weight mercaptans have a very obnoxious odor, particularly noticeable due to their low vapor pressure. All mercaptans, like most sulfur compounds, have a deleterious effect on lead susceptibility of motor gasolines. Thiophenols promote gum and sludge formation reactions and cause storage problems. Mercaptans can react with copper, particularly if basic nitrogen is present, to form oil soluble cuprow compounds which can be oxidized to insoluble


copper oxide. The soluble copper can also then react with some oxygenated compounds to form gelatinous copper phenolates. Low molecular weight mercaptans may be extracted to some degree with caustic or be sweetened, in a process which converts the mercaptans to less odiferous disulfides. Mercaptans may be extracted from LPG to very low values, 0.5 wt. ppm in some cases. Extraction of 50-95% of the mercaptans from gasoline is possible, depending on boiling range. However, as mercaptans in FCC gasoline generally represent only 10 to 35% of the total sulfur present, partial gasoline desulfurization by extraction is usually not economical. The majority of FCC gasolines are currently only sweetened to essentially eliminate mercaptans and produce a doctor negative, or sweet, gasoline. Cycle oil is sometimes sweetened, but if low sulfur is desired, the cycle oil must be hydrotreated. Non regenerable caustic scrubbing to extract mercaptans is an old and expensive process which may create or compound waste disposal problems. In many cases it is not very effective. The most successful treating process for either extraction or sweetening is the UOP Merox process. This process uses small amounts of caustic to remove or transform mercaptans. A catalyst and oxygen supplied from atmospheric air are used to regenerate the caustic. Further information on the Merox process may be obtained from UOP. Total gasoline sulfur may also be reduced by undercutting or reducing the endpoint of the FCC gasoline. The sulfur concentration in the FCC gasoline increases rapidly as the boiling range exceeds ~400º (200ºC). By reducing the endpoint of the gasoline so that the heaviest 20% is included in the light cycle oil product the gasoline sulfur content can be reduce by as much as 60%. Unfortunately, there are significant product value losses associated with this for most refiners.


Carbonyl Sulfide (COS) Carbonyl sulfide is produced during the cracking reaction. It boils slightly below propane at -58°F (-50°C). Upon post fractionation the COS concentrates in LPG, propane-propylene, and finally propylene as the distilled cuts are narrowed in boiling range. COS concentrations in the LPG stream may range from 5 to 100 ppm; the amount usually rises with increased feed sulfur, but is very unpredictable. COS, like some other sulfur compounds, should be considered toxic. The principle concern with COS, aside from situations where product total sulfur must be very low or zero, is its tendency to hydrolyze, forming corrosive H2S and CO2. This hydrolysis proceeds slowly but is catalyzed by activated alumina or molecular sieve desiccants. COS also partially reacts with MEA in acid gas scrabbling systems to form high boiling by-products which are not steam regenerable. COS is also a strong poison for polypropylene producers using FCC propylene as a feed. Some polymerization catalysts are sensitive to as little as 5 ppb COS. Carbonyl sulfide can be partially removed from the light product streams by scrubbing with an aqueous DEA solution. Diethanolamine, unlike monoethanolamine, does not chemically react with COS and therefore is steam regenerable. Aqueous DEA scrubbing can usually reduce COS levels to less than 10 ppm. A reduction of COS to levels of one ppm or less can be effected with a batch scrubber using MEA, sodium hydroxide, and water to remove COS by chemical reaction. This system is not regenerable. For units producing polymer grade propylene additional adsorbents or reactive treaters will commonly be used to remove COS down to undetectable levels. Copper Copper is a powerful oxidation catalyst. It is usually picked up from contact with copper or copper alloy surfaces, especially when ammonia or other basic nitrogen and mercaptans are present. Copper catalyzes gum formation by promoting olefin oxidation reactions faster than inhibitor additives can terminate these chain


reactions. Cracked distillates containing copper will form gum and gelatinous precipitates in storage. Copper also contributes to distillate color stability problems and catalyses oxidation of alkyl lead additives, leading to haze in some finished gasolines. Gross copper contamination can be removed by chemical treatment, such as dilute acid scrubbing or clay percolation. If the copper does not exceed 1 mg/liter, it can be rendered harmless with a copper deactivator. Dosage is roughly 10 wt. ppm active agent for each mg/liter. The deactivator, a chelating agent, complexes with the copper so that it cannot take part in any further reactions. Nitrogen Compounds There are usually two types of nitrogen found in gasoline and cycle oil. One type would be neutral compounds such as pyrrole, and the other a basic compound such as pyridine or quinoline. Ammonia is found in the reactor overhead vapors, but if normal main column overhead water injection is maintained, the ammonia is removed at this point. Neutral nitrogen compounds are associated with sediment formation in fuel oils. Basic compounds are typically more harmful. Pyridine has a characteristic odor which is very unpleasant when combined with mercaptans. Both basic and neutral nitrogen compounds are color precursors which affect distillate color stability. Their effect on lead susceptibility and octane number varies between good and bad depending on the particular compound in question. Both types of nitrogen compounds may also promote oil soluble gum formation, but this has not been firmly established at this time. Neutral nitrogen compounds may be removed with either strong acid or strong caustic-methanol solutions. Basic nitrogen compounds could be removed with dilute acid in acid resistant equipment. If acid strength is too high, there may be problems with olefin addition, sulfonation, and other acid catalyzed reactions. Hydrotreating will decompose both types of nitrogen compounds to ammonia. This might be


suitable for cycle oils, but not for gasoline, because aromatics might be saturated, with severe loss in octane number. Oxygen Oxygen compounds such as phenols and cresols are formed during the cracking reaction from oxygenated compounds in the feed and from oxygen entrained with the catalyst. They are generally not considered harmful and may be considered beneficial as some act as oxidation inhibitors. Some of the phenols are suspected of being color precursors. Caustic treating can be used to remove most of the oxygen containing compounds, including most organic acids, when deemed necessary. Diolefins Diolefins, especially conjugated diolefins, are reactive hydrocarbons which quickly enter into gum and sediment forming reactions. These compounds are normally found in significant concentrations only in thermally cracked products. However, when FCC conditions allow cracking to occur in the absence of catalyst, diolefins will be formed in amounts roughly proportional to the degree of thermal cracking present. A good example of this type of compound is butadiene. It stays in the C4 fraction as it is concentrated through fractionation. If this stream is then used as alkylation feed, the butadiene will polymerize to form tar. Higher order diolefins lead easily to gum formation in the gasoline and cycle oil streams. Diolefins can be removed without significant olefins removal by carefully controlled mild hydrotreating or vapor phase clay treatment. Proper dosage of oxidation will nullify their effects. Another solution would be to minimize formation of diolefins by decreasing thermal cracking. This could be done by reducing the reactor temperature, reducing the regenerated catalyst temperature or by mechanical changes minimizes post riser residence time.


SUMMARY – FCC PRODUCT TREATMENT METHODS Contaminant

Concentration

Treatment

Copper

High

Dilute mineral acid Clay percolation

Low (<1 mg/l)

Metal deactivator addition

Elemental sulfur

—

Redistribution Hydrotreating Reverse Doctor Prevent formation by minimizing free oxygen

Hydrogen sulfide, H2S

High (>1000 ppm)

MEA or DEA scrubbing

Low

Caustic soda, MEA or DEA scrubbing

Any

Merox (Regenerative caustic) Extractive for LPG or Sweetening for Gasoline

Mercaptans, R-SH

Total Gasoline Sulfur Carbonyl sulfide, COS

Undercutting of FCC gasoline —

DEA-water scrubbing Caustic-MEA-water scrubbing (Batch operation only) Silica-Alumina Adsorbents PbO Treaters

Basic nitrogen compound

—

Water Wash (NH3) Weak mineral acid

Neutral nitrogen compound

—

Strong acid Strong caustic-methanol solution


Contaminant

Concentration

Treatment

Oxygen compounds, phenols or acids

—

Water Wash Caustic soda

Diolefins

—

Mild hydrotreating Vapor phase clay treating Change FCC operations to minimize formation

NOTE: Hydrotreating may be used to eliminate many of these contaminants. In practice, however, it is used for cycle oil treating in some cases, rarely for LPG or gasoline. Treating of the FCC feed is also effective in minimizing many of these. Regenerative caustic treating, such as the Merox process, is a very common treating method. Batch scrubbing using caustic, or regenerative MEA-DEA scrubbing are also used. This may be in conjunction with a Merox unit or without it, depending on the contaminant and on economic considerations.

157048 Analytical Methods Page 1

ANALYTICAL METHODS Introduction Good analyses of the feed and product streams are essential for control and evaluation of a Fluid Catalytic Cracker. It is extremely difficult to optimize a unit if potential problems are not defined through the laboratory and process variables. The following sections give a typical FCC sampling schedule and a brief outline of some of the more common analytical methods. There are a large number of laboratory tests which may be used. FCC products vary widely, from clarified oil to fuel gas. This in turn leads to markedly different analytical methods, such as six different types of distillations. The product being tested and the type of result desired will determine which test is used. Distillation There are six distillation methods listed in this book. Two of them, UOP 1 and ASTM D 86, cover the lighter fractions, from gasoline to gas oils. The UOP 1 method goes further; it continues past the typical decomposition point of 700°F (371°C) into a thermal cracking of the sample with a dry residue (coke) remaining. Two other methods, UOP 77 and 79, are fractionations in addition to distillations. Either one can be used to separate certain fractions of a product for further analysis. UOP 77 is more commonly used. UOP 79 is a high precision distillation method which is used to determine true boiling points of petroleum fractions. The test requires special equipment and the information obtained from this test is not frequently needed. The two vacuum distillation methods, UOP 76 and ASTM D 1160, are used for heavy material. Reduced pressures, down to 1 mm Hg absolute for D 1160 and 0.3 mm Hg absolute for UOP 76, permit distillation when the temperature used for atmospheric techniques would lead to thermal cracking. Sulfur


The six analytical methods for sulfur analysis can be initially divided into two groups, gases and liquids. The Tutwiler method, UOP 9, gives a quantitative determination of H2S in a gas stream. A second method, UOP 212, measures H2S, mercaptans, and COS in light gases and LPG. The Tutwiler method does not give as detailed a test as does UOP 212, but takes less time. There are four methods given for hydrocarbon liquid analysis. The Doctor Test, UOP 41, gives a qualitative determination of mercaptans and H2S. UOP 163 will give a breakdown of mercaptan and H2S in liquid streams. For total sulfur of lighter oils, the Lamp method, ASTM D 1266, is used. For oils boiling above 350°F (177°C), ASTM D 1552 may be used to determine total sulfur. Octane The Motor octane method, ASTM D 2700, is more severe, i.e., gives a lower rating than does the Research method, ASTM D 2699. The correlation between the two is not exact, so it is generally not easy to predict one octane from another. For similar feedstocks and plant operation, the refiner may be able to make some general predictions from past data. Gas Chromatography To determine the composition of light hydrocarbon gases and LPG streams containing small amounts of C3 and C6 material, either UOP 539 or UOP 709 may be used. Neither of these will separate argon from oxygen, or butene-1 from isobutylene. UOP 709 is easier to run than UOP 539 and requires less elaborate equipment, but UOP 709 does not differentiate between ethane, ethylene and carbon dioxide. For exact determination of fuel gas from an FCC, UOP 539 would be the better method. UOP 725 can be used to determine concentration of C5 and lighter material in FCC gasoline.


Samples for UOP Analysis Most refineries have laboratories for common analyses. Samples may be sent to UOP if the refiner does not have the equipment, manpower, or wishes to check the accuracy at his own lab. Samples which are sent to UOP for analysis sometimes go astray, either in shipping or within UOP. In order to minimize delays caused by lost, misplaced or unidentified samples, please observe the following rules. 1.

Identify each sample with refiner, location, unit, sample, date and technical contact person at UOP responsible for the results. Make certain the sample tag is well secured to the sample and remains legible, even if it is wet with water or hydrocarbon.

2.

It has been found that gasoline or naphtha sample collection in clear bottles, which are left exposed to sunlight (or ultraviolet lighting) either direct or indirect, and whether or not the sample has been treated with oxidation inhibitors, will result in a severe loss of octane rating within a very short time. Every effort must therefore be made to see that only the brown or amber sample bottles are used for daily samples and more important that all samples shipped to UOP for analysis are taken in amber bottles, are kept in a cool dark place until packaged for shipment, and are shipped as soon as possible after sampling.

3.

When shipping samples, mail or phone shipping information (air waybill, flight number, bill of lading, etc.) to UOP.

4.

Send a copy of the request for analysis with the samples; send the original to: UOP Technical Service Department FCC Group 25 E. Algonquin Road Des Plaines, IL 60017 USA


5.

Send samples to: UOP LLC Shipping and Receiving 50 E. Algonquin Road Des Plaines, IL 60017 USA

6.

For larger “Rush” samples being sent to UOP by air, indicate the shipping instructions as: SHIP TO: UOP LLC Chicago, Illinois 60017 USA HOLD AT: O Hare Airport CALL: (847) 391-3043 ON ARRIVAL

7.

All samples shipped from outside the United States require additional paperwork to comply with the U.S. Environmental Protection Agency’s Toxic Substances Control Act (TSCA). The following are some of the guidelines for importing samples into the United States. Always contact your customer service or technical service representative for assistance before shipping any samples to ensure that all regulations are followed.  The use of a freight forwarder (e.g. Burlington Air Express or Emery Worldwide) is preferred for all imported samples.  Overnight couriers (e.g. DHL, Federal Express or UPS) may be used only if the written TSCA certification – authorized, signed and dated by UOP is obtained prior to shipment and physically included with the documentation in the shipped package.  All samples should be routed through Chicago’s O’hare International Airport  All imports must have the following documentation: 

   

Bill of Lading (also known as a Master Airwaybill) which includes vessel/flight information. The Bill of Lading must state: "Customs clearance by Circle International". The Bill of Lading description field must begin with the words “TSCA Certified Chemical Sample.” Pro forma invoice (Attachment 1a and 1b: a blank form and example). TSCA Certification (Attachment 2). Analytical Requisition Form (Attachment 3) Material Safety Data Sheet (MSDS) if the sample material is regulated as hazardous by IATA, IMO or DOT.


 The documentation on the above listed items must contain the following information: On the Proforma Invoice:  Shipper’s name, address, contact name, phone number and fax number.  Importer’s name, address, contact name, phone number and fax number (this must include the UOP contact person within the U.S.). UOP has provided this information except for Contact Person  Consignee/Delivery name, address, contact name, phone number and fax number (this must include the actual delivery location). UOP has provided this information.  Approximate market value (for Customs purposes), for each item, in U.S. Dollars.  Sample descriptions in English.  Packing details – the number and types of containers.  Net weights and Gross weights for each item, in kilograms.  Country of origin. This is the country where the sample was taken. On the TSCA Certification Form:  Sample description in English, UOP has provided this information.  Date the sample is to be shipped, in English. On the International Analytical Requisition Form:  Shipper’s name, address, contact name, phone number and fax number.  Sample shipping information: Carrier and flight information, phone number of the carrier, airway bill no.  Fill in all pertinent sample and analytical request information as required.

 All samples must be prepared according to the hazardous materials shipping regulations of the International Air Transportation Association (IATA), if shipped by air, or International Maritime Organization (IMO) if shipped by sea.  Prior to shipping, the UOP Tech Service Sample Coordinator must be notified of all imports before their arrival in the United States. Please fax a copy of the Pro Forma Invoice, the Bill of Lading, TSCA Certificate, and the Analytical Requisition Form to the UOP Technical Service Sample Coordinator at 847-391-2253. UOP will then ensure proper Customs and TSCA clearance. Failure to send samples with the proper documentation and information will result in a delay of Customs clearance or refusal of the sample.


Attachment 1a Proforma Invoice International Shipper/Exporter (Name & Address) Invoice No: Invoice Date: Terms: Pro-Forma Invoice Reference No:

Contact Name: Contact Phone No.

Fax No.

DESCRIPTION * * * NO CHARGE INVOICE * * * QUANTITY:

GROSS WEIGHT:

MARKET VALUE -DECLARED FOR CUSTOMS CLEARANCE PURPOSES ONLY

BOX(ES) KGS.

COUNTRY OF ORIGIN (where sample was taken) __________________________ MARKS: AS ADDRESSED IMPORTER OF RECORD UOP LLC 25 EAST ALGONQUIN RD. DES PLAINES, IL 60017 ATTN:

CONSIGNEE / DELIVERY ADDRESS: UOP LLC SAMPLE RECEIVING 50 EAST ALGONQUIN RD. DES PLAINES, ILLINOIS 60017-5016 ATTN: UOP Tech Service Sample Coordinator

UOP Contact Name Refinery Representative Signature

US DOLLARS $

DESCRIPTION OF SAMPLE(S):

PACKED IN

AMOUNT

Refinery Representative Name: please print

I HEREBY CERTIFY THAT THIS INVOICE IS TRUE AND CORRECT. ORIGINAL INVOICE


Attachment 1b Example Proforma Invoice International Shipper/Exporter (Name & Address) Mountain View Refining 1000 Mountain View Dr. Boulder City Refining Country Contact Name: Joe R. Engineer Contact Phone No. (12) 3 456 -7890 Fax No. (12) 3-098-7654

Invoice No: Invoice Date: Terms: Pro-Forma Invoice Reference No:

DESCRIPTION

* * NO CHARGE INVOICE * * * QUANTITY: Three 1 liter catalyst samples

AMOUNT

US DOLLARS $

DESCRIPTION OF SAMPLE(S): R-134 CCR Platforming Catalyst Two regenerated samples and one spent sample PACKED IN

1

GROSS WEIGHT:

BOX(ES) 3.5

KGS.

COUNTRY OF ORIGIN (where sample was taken) Refining Country MARKS: AS ADDRESSED IMPORTER OF RECORD UOP LLC 25 EAST ALGONQUIN RD. DES PLAINES, IL 60017 ATTN:

Robert S. UOP

CONSIGNEE / DELIVERY ADDRESS: UOP LLC SAMPLE RECEIVING 50 EAST ALGONQUIN RD. DES PLAINES, ILLINOIS 60017-5016 ATTN: UOP Tech Service Sample Coordinator

UOP Contact Name

Joe R. Engineer Refinery Representative Signature

Joseph R. Engineer Refinery Representative Name: please print

I HEREBY CERTIFY THAT THIS INVOICE IS TRUE AND CORRECT. ORIGINAL INVOICE

MARKET VALUE -DECLARED FOR CUSTOMS CLEARANCE PURPOSES ONLY


Attachment 2a TSCA Certification Form

To: Area Director of Customs Date: Description of Sample: Only one of the following should be selected:

( X ) TSCA Positive Certification I certify that all chemical substances in this shipment comply with all applicable rules or orders under TSCA and that I am not offering a chemical substance for entry in violation of TSCA or any applicable rule or order under TSCA. OR

( ) TSCA Negative Certification I certify that all chemicals in this shipment are not subject to TSCA

Signature (Authorized UOP LLC employee)

Printed Name (Authorized UOP LLC employee)


Attachment 2b TSCA Certification Form

Example Form

To: Area Director of Customs Date: August 23, 1999 Description of Sample: R-134 CCR Platforming Catalyst: Spent Only one of the following should be selected:

( X ) TSCA Positive Certification I certify that all chemical substances in this shipment comply with all applicable rules or orders under TSCA and that I am not offering a chemical substance for entry in violation of TSCA or any applicable rule or order under TSCA. OR

( ) TSCA Negative Certification I certify that all chemicals in this shipment are not subject to TSCA

Signature (Authorized UOP LLC employee)

Angelo P. Furfaro

Printed Name (Authorized UOP LLC employee)


Attachment 3 International Analytical Requisition Form - UOP Technical Service TO: UOP Technical Service Sample Coordinator 391-2253

Phone 847-391-2620 FAX: 847-

UOP Technical Service Contact 391-2253

Phone 847-391-

Fax 847-

Phone

Fax

Customer:

Refinery Address Contact Name

Liquid & Gas Sample(s) Analysis Required:

Description

Check if MSDS is Included

Rush or Standard

Catalyst & Adsorbent Sample(s): Sample Location:

Process or unit: Catalyst Type

Regenerated or Coked Standard

Billing Information: Address:

Attention: Sample Shipping Information: Shipped Via: Phone Number of Carrier:

Airwaybill No.:

Analysis Required

Check if MSDS is Included

Rush or


Minimum Sample Size Analysis

Minimum (cm3)

Heavy Oil (Clarified Oil, Slurry, Raw Oil, Heavy and Light Cycle Oil) API Distillation Viscosity Vacuum Distillation Conradson Carbon Ash (sample to be taken and shipped in a wide mouth sample container) Sediment and Water Sulfur Nitrogen Metals Pour Point Color

100 250 150 300 150 300 150 20 15 100 200 50

Gasoline, LPG API Distillation Hydrocarbon Types by GC C5- by GLC RVP Mercaptan Sulfur Total Sulfur Octane Research Motor Leaded or Clear Catalyst

100 210 150 50 1000 100 100 1000 for each type 1000

NOTE: Individual samples can be taken from one large sample of each product, usually about 1-2 gallons (3.5-7.0 liters).


TYPICAL TEST SCHEDULE

STREAM AND TEST Raw Oil Charge Gravity Viscosity Vacuum Distillation Conradson Carbon Residue BS & W Sulfur Total Nitrogen Metals Content by Wet Ash (May be sent to Outside Lab) Pour Point Basic Nitrogen Heptane Insolubles UOP K

TEST NUMBER

FREQUENCY Normal Startup

D-1298 or D-5002 or D-4052 D-445 D-1160

1/D 1/W 1/D

3/D 3/W 1/D

D-189 or D4530 D-4007 UOP 864 or D-1552 or D-2622 UOP 384 or D-4629 UOP 389 or D-5708

1/D 1/W 1/D 1/W 1/W

1/D 1/W 1/D 1/W 1/D

D-97 UOP 269 UOP 614 UOP 375

1/W Occas Occas 1/D

1/W Occas Occas 1/D

Circulating Main Column Bottoms Gravity D-1298 or D-5002 or D-4052 BS & W D-4007 Ash D-482

Occas Occas Occas

1/D Control 1/D

Main Column Bottoms Clarified Oil Product Gravity D-1298 or D-5002 or D-4052 Viscosity D-445 Vacuum Distillation D-1160 BS & W D-4007 Sulfur UOP 864 or D-1552 Ash D-482

3/D 1/W 1/D 1/D 1/W 1/W

3/D 1/W 1/D Control 1/D 1/D


STREAM AND TEST

TEST NUMBER


Heavy Cycle Oil Product Gravity Distillation Pour Point Flash Point Viscosity Sulfur Cetane Index

D-1298 or D-5002 or D-4052 D-86 D-97 D-93 D-445 UOP 864 D-976

1/D 1/D 1/D 1/D 1/D 1/D 1/D

3/D 3/D 1/D 1/D 1/W 1/D 1/D

Light Cycle Oil Product Gravity Distillation Pour Point Flash Point Viscosity Sulfur Cetane Index

D-1298 or D-4052 D-86 D-97 D-93 D-445 UOP 864 D-976

1/D 1/D 1/D 1/D 1/D 1/D 1/D

3/D 3/D 1/D 1/D 1/W 1/D 1/D

Stripped Heavy Naphtha Product Gravity D-1298 or D-4052 Distillation D-86 H2S and RSH UOP 163 Sulfur UOP 864 Composition (PONA) UOP 777 Research Octane D-2699 Motor Octane D-2700 RVP D-323

1/D 1/D 1/W 1/D 3/W 1/D 1/D 1/D

3/D 3/D 1/W 1/D 1/D 1/D 1/D 1/D

Blended Fuel Oil Product Gravity Distillation Viscosity Sulfur Pour Point Flash Point

D-1298 or D-4052 D-86 D-445 UOP 864 or D-1552 D-97 D-93

1/D 1/D 1/W 1/D 1/D 1/D

3/D 3/D 1/W 1/D 1/D 1/D

Main Column Receiver Gas H2S Composition

UOP 212 UOP 539

Occas Occas

1/D 1/D


STREAM AND TEST

TEST NUMBER


Main Column Receiver Liquid Gravity Distillation GC (C4 and lighter) H2S and RSH

D-1298 or D-4052 D-86 UOP 725 UOP 163

Occas Occas Occas Occas

Occas Occas Occas Occas

Regenerator Flue Gas Composition (Orsat or GC) SOx NOx Particulate

UOP 172 or UOP 539 EPA #6 EPA #7 EPA #5

1/D Occas Occas Occas

3/D Occas Occas Occas

Spent Catalyst Percent Carbon

UOP 703

1/W

Regenerated Catalyst Particle Size Distribution* Surface Area* Pore Volume* Activity* Metals by ICP* Percent Carbon*

UOP 856 UOP 874 UOP 874 D-3907 UOP 546 UOP 703

1/W 1/W 1/W 1/W 1/W 1/D

2/W 2/W 2/W 2/W 2/W Control

Flue Gas to Electrostatic Precipitator Isokinetic Particle Determin. EPA #5

Occas

Occas

Flue Gas from Electrostatic Precipitator Isokinetic Particle Determin. EPA #5

Occas

Occas

*Normally performed by catalyst vendor’s laboratory.


STREAM AND TEST

TEST NUMBER

Main Column Receiver Water Iron, Copper Phenols Cyanides Sulfides Ammonia Total Oils pH Total Dissolved Solids Silica

UOP 314 UOP 262 UOP 682 UOP 683 UOP 740 D-3921 D-1293 D-1126 D-859

BFW and Continuous Blowdown Sodium D-4192 Total Alkalinity D-1067 M Alkalinity D-1067 P Alkalinity D-1067 Chloride (as Cl-) D-512 Silica (as SiO2) D-859 Total Dissolved Solids STD Method 2540C Total Suspended Solids STD Method 2540D pH D-1293 Specific Conductance D-1125 Phosphates D-4327 Oil D-3921 Hydrazine D-1385

FREQUENCY Normal Startup Occas Occas 1/M 1/M Occas Occas 1/D 1/W 1/M

Occas Occas Occas Occas Occas Occas 3/D Occas Occas

1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W

1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W 1/W

Saturated Steam from Catalyst Cooler, Flue Gas Cooler, and Main Column Bottoms Steam Generator Steam Drums Impurities D-2186-C Occas Occas Silica D-859 Occas Occas Sodium D-1428 Occas Occas Lean Gas Product H2S Composition

UOP 212 UOP 539

1/D 1/D

3/D 3/D

Debutanizer Net Overhead Liquid H2S & Mercaptan Sulfur UOP 163 Composition UOP 539 Total Sulfur UOP 923

1/D 1/D 1/D

3/D 3/D 3/D


STREAM AND TEST

TEST NUMBER


Debutanizer Bottoms Product Gravity Distillation H2S & Mercaptan Sulfur RVP Research Octane Motor Octane C4 and lighter Sulfur

D-1298 or D-4052 D-86 UOP 163 D-323 D-2699 D-2700 UOP 725 UOP 864 or UOP 836

1/D 1/D 1/D 1/D 1/D 1/D 1/D 1/D

3/D 3/D 3/D 3/D 3/D 3/D 3/D 3/D

Naphtha Splitter Bottoms Gravity Distillation Research Octane Motor Octane Sulfur

D-1298 or D-4052 D-86 D-2699 D-2700 UOP 864 or UOP 836

1/D 1/D 1/D 1/D 1/D

3/D 3/D 3/D 3/D 3/D

Naphtha Splitter Side Cut Product Gravity D-1298 or D-4052 Distillation D-86 Research Octane D-2699 Motor Octane D-2700 Sulfur UOP 864 or UOP 836

1/D 1/D 1/D 1/D 1/D

3/D 3/D 3/D 3/D 3/D

Naphtha Splitter Overhead Product Gravity D-1298 or D-4052 Distillation D-86 RVP D-323 Research Octane D-2699 Motor Octane D-2700 C4 and lighter UOP 725 Sulfur UOP 864 or UOP 836

1/D 1/D 1/D 1/D 1/D 1/D 1/D

3/D 3/D 3/D 3/D 3/D 3/D 3/D

Depentanizer Overhead Product H2S & Mercaptan Sulfur UOP 163 Composition UOP 539

1/D 1/D

3/D 3/D


STREAM AND TEST

TEST NUMBER


Depentanizer Bottoms Product Gravity Distillation H2S & Mercaptan Sulfur RVP Research Octane Motor Octane C4 and lighter Sulfur

D-1298 or D-4052 D-86 UOP 163 D-323 D-2699 D-2700 UOP 725 UOP 864 or UOP 836

1/D 1/D 1/D 1/D 1/D 1/D 1/D 1/D

High Pressure Receiver Water Iron, Copper Phenols Cyanides Sulfides Ammonia Total Oils pH

UOP 314 UOP 262 UOP 682 UOP 683 UOP 740 D-3921 D-1293

Occas Occas 1/M 1/M Occas Occas 1/D

3/D 3/D 3/D 3/D 3/D 3/D 3/D 3/D Occas Occas Occas Occas Occas Occas 3/D

LABORATORY TEST SCHEDULE FREQUENCY NOMENCLATURE 1/D 3/D 1/W 3/W 2/M N S Occas Control

One determination per day Three determinations per day One determination per week Three determinations per week Two determinations per month During normal operation During startup Determination is done only occasionally Determination is done as frequently as necessary for plant control during startup


OUTLINE OF SELECTED ANALYTICAL METHODS SUBJECT INDEX Test Description Activity Test for Catalyst API Gravity Apparent Bulk Density of Catalyst Ammonia in Refinery Water Ash from Petroleum Products Boilaway (weathering Test) Carbon on Catalyst Carbonyl Sulfide (COS) in Gases Catalyst Loading in Heavy Oil Color – ASTM – Saybolt Conradson Carbon Residue Copper in Water Cyanide in Refinery Water Distillation – of Heavy Oil – of Petroleum – of Petroleum – of Petroleum (Fractionation) – Vacuum – Vacuum Doctor Test Flash Point, Closed Cup Flash Point, Open Cup Fractionation of Petroleum

Number D-3907 D 1298 UOP 254 UOP 740 D 482 UOP 155 UOP 703 UOP 212 UOP 233 D 1500 D 156 D 189 UOP 314 UOP 682 UOP 1 D 86 UOP 77 UOP 79 UOP 76 D 1160 UOP 41 D 93 D 92 UOP 79


Test Description Flue Gas Analysis, (GC) (Orsat) Particulates, SOx, NOx Gas Analysis – GC – GC - Pentenes and lighter in olefinic gasoline Gravity – API Gum – Copper Dish – Existent (Steam Jet) Hydrogen Sulfide (H2S) in Gas – Tutwiler – with Mercaptans Induction Period of Gasoline Iron in Water Isokinetic Particle Determination in Flue Gas Kinematic Viscosity Loss of Ignition of Catalyst Mercaptan Sulfur – Gases – Liquid Hydrocarbons Metals – Trace, in Crackling Catalysts – Trace, in Oils – Trace, in Oils Nitrogen in Heavy Distillate Octane – Motor – Research Particle Size Distribution of Catalyst pH, Iron and Copper in Refinery Water

Number UOP 539 UOP 172 EPA #5,6,7 UOP 539 UOP 709 UOP 725 D 1298 UOP 11 UOP 277 UOP 9 UOP 212 UOP 6 UOP 314 D 3685 D 445 UOP 275 UOP 212 UOP 163 UOP 546 UOP 389 UOP 391 UOP 384 D 2700 D 2699 UOP 422 UOP 314


Test Description Phenols in Petroleum Products Pore Volume and Pore Diameter of Catalyst Pour Point of Petroleum Oils Reid Vapor Pressure Sampling of Petroleum Sediment and Water in Oil Sintering Index of Catalyst Sulfides in Refinery Waters Sulfur in Heavy Oils Sulfur – Doctor Test (H2S and Mercaptans) – H2S in Gases (Tutwiler) – Mercaptan and H2S in Light Distillates – H2S, Mercaptans, and COS in Hydrocarbon Gases – in Heavy Distillates – Total, in Light Distillates Total Sulfur – Lamp – Quartz Tube Surface area, Pore Volume, and Pore Diameter of Catalyst UOP Characterization Factor – K Vacuum Distillation Viscosity – Kinematic Water and Sediment in Oil Weathering Test (Boilaway)

Number UOP 262 UOP 425 D 97 D 323 UOP 516 or D 270 D 4007 UOP 424 UOP 683 D 1552 UOP 41 UOP 9 UOP 163 UOP 212 D 1552 D 1266 D 1266 D 1552 UOP 425 UOP 375 D-1160 D 445 D 4007 UOP 155


OUTLINE OF SELECTED ANALYTICAL TEST METHODS NUMERICAL INDEX Test Number

Description

UOP 1 UOP 6 UOP 9

Distillation Range of Heavy Oils Induction Period of Gasoline H2S in Gases (Tutwiler)

UOP 11 UOP 41

Gum, Copper Dish Doctor Test (H2S and Mercaptans)

UOP 76 UOP 77 UOP 79 D 86 D 92 D 93 D 97 UOP 155 D 156 UOP 163

Vacuum Distillation Distillation of Petroleum Distillation of Petroleum Distillation of Petroleum Flash Point, Open Cup Flash Point, Closed Cup Pour Point of Petroleum Oil Weathering Test (Boilaway) Saybolt Color Mercaptans and H2S in Liquid Hydrocarbons

UOP 172 D 189 UOP 212

Flue Gas Analysis (Orsat) Conradson Carbon Residue H2S, Mercaptans and COS in Hydrocarbon Gases

UOP 233 UOP 251 UOP 254 UOP 262 D 270

Catalyst Loading Test Activity Test for Catalyst Apparent Bulk Density of Catalyst Phenols in Petroleum Products Sampling Petroleum Products


Test Number

Description

UOP 275 UOP 277 D 287 UOP 314 D 323 UOP 375 UOP 384 UOP 389 UOP 391 UOP 422 UOP 424 UOP 425 D 445 D 482 UOP 516

Loss of Ignition of Catalyst Existent Gum (Steam Jet) API Gravity pH, Iron, and Copper in Refinery Water Reid Vapor Pressure UOP Characterization Factor Nitrogen in Heavy Distillate Trace Metals in Oils Trace Metals in Oils Particle Size Distribution in Catalyst Sintering Index of Equilibrium Catalyst Surface Area, Pore Volume, and Pore Diameter of Catalyst Kinematic Viscosity of Oils Ash from Petroleum Products Sampling of Gasoline, Distillates and C3-C4 Fractions

UOP 539 UOP 546 UOP 682 UOP 683 UOP 703 UOP 709 UOP 725 UOP 740 D 1160 D 1266 D 1500 D 1552 D 2699 D 2700 D 3685 D 4007

Gas Analysis (GC) Metals in Cracking Catalyst Cyanide in Refinery Water Sulfide in Refinery Water Carbon on Catalyst Gas Analysis (GC) Pentenes and lighter in olefinic gasoline Ammonia in Water Vacuum Distillation Total Sulfur in Petroleum Products (Lamp) ASTM Color (formerly ASTM Union) Sulfur in Heavy Oils Octane Rating, Research Octane Rating, Motor Isokinetic Particle Determination in Flue Gases Water and Sediment in Oil


DISTILLATION RANGE OF HEAVY PETROLEUM OILS UOP METHOD 1

Scope This method is for determining the distillation range of heavy petroleum oils. It is applicable to petroleum products whose boiling range extends above that of kerosene; e.g., crude oils, gas oils and fuel oils. The method differs from ASTM Method D 86 in that a 200 ml flask is used and the distillation is continued past the thermal decomposition point to a dry or coke residue. The test is particularly useful in estimating gasoline, kerosene and/or distillate contents and the coking characteristics of these oils. Outline of Method A 100-ml sample is distilled under prescribed conditions. Systematic observations of thermometer readings and volumes of condensate are made, and the weight of coke or residue is determined. The results of the test are calculated and reported from these data. No corrections are applied to the data. Precautions Oils containing more than traces of water are very difficult to distill. However, if the heat is applied to the flask correctly, water can be distilled from the oil without “bumping.” When water is present, heat the flask evenly over its top and bottom surfaces; do not concentrate heat on the bottom of the flask. Keep the top of the flask hot enough to prevent water vapor from condensing there and allow time for the temperature to drop to ambient before continuing the distillation. Do not include water in the percentages reported for the temperatures named. The IBP obtained in this manner on samples containing water may, or may not, be a true IBP owing to the possibility of superheating the vapors when the flame is applied to the top surface of the flask. Note this on the distillation sheet.


Decomposition or cracking usually occurs when oils containing material boiling above 625°F (329°C) are distilled at atmospheric pressure. When this decomposition takes place it will be impossible to maintain a uniform distillation rate without causing a gradual drop in temperature. Therefore, disregard the 4-5 ml per minute rate from the temperature at which cracking begins and continue the distillation at such a rate that there is a steady rise in temperature. Precision See ASTM Method 86 for precision statement.


INDUCTION PERIOD OF GASOLINES BY THE UOP OXYGEN BOMB UOP METHOD 6

Scope This method is for determining the induction period of gasolines. It is useful in predicting the storage stability, in lieu of the more valuable storage tests and accelerated gum determination. It is a valuable control test and an excellent measure of inhibitor effectiveness. This method does not give the same numerical induction periods as ASTM Method D 525 because of differences in construction of the bombs and bath. However, the results are parallel and this method is preferred for its speed and convenience. Outline of Method The sample is placed in a bomb at 60-70°F (15-20°C) and subjected to oxygen at 100 psig. The bomb is heated rapidly to 211.6°F (99.7°C). The pressure is recorded continuously until the break point has been passed. The induction period is then determined from the chart record. Precautions The bombs and bottles must be scrupulously clean to obtain reproducible results. Precision Repeatability should be considered suspect it results differ from the mean by more than 5%.


HYDROGEN SULFIDE IN GASES BY THE TUTWILER METHOD UOP METHOD 9

Scope This method is for the determination of hydrogen sulfide in gas mixtures. Mercaptan sulfur, if present, is determined as hydrogen sulfide. The accuracy of this method is not sufficient to obtain reliable results below 5 grains of H2S per 100 cu. ft. Outline of Method The sample is admitted to a Tutwiler buret, displacing a starch solution. A known volume of starch solution is retained in the buret and a standard iodine solution is admitted and measured from the buret until the starch solution assumes a taint permanent blue color. The concentration of hydrogen sulfide is calculated from the volume of iodine used and its known normality. Precautions It is recommended that gases to be analyzed for hydrogen sulfide content be sampled directly from the plant stream into the buret. If the sample is to be transported, it should be done in a dry glass or stainless steel container. Do not confuse the blue color of the iodine-starch complex with the opalescent milky appearance resulting from the separation of free sulfur.


Precision Duplicate results by the same operator should be considered suspect it they differ by more than the following amounts, depending on the iodine solution used: Iodine Solution A:

10 grains*

Iodine Solution B:

20 grains*

Iodine Solution C:

5 grains*

*To convert to weight ppm: grain/100 SCF  542.1 = ppm (wt) MW gas


COPPER DISH GUM CONTENT OF GASOLINE UOP METHOD 11

Scope This is a method for determining the weight of the residue obtained when a gasoline or naphtha is evaporated in a copper dish. Considered in connection with the induction period, it is an indication of the stability of the gasoline in storage. Outline of Method The sample is evaporated in a clean, copper dish under controlled conditions and the weight of the residue is determined. Report Report the average weight of the residue as milligrams of copper dish gum per 100 ml of gasoline.


DOCTOR TEST FOR PETROLEUM DISTILLATES UOP METHOD 41

Scope This is a qualitative test for the presence of hydrogen sulfide and mercaptans in gasoline, jet fuel, kerosene and similar petroleum products. Outline of Method The sample is shaken with a sodium plumbite solution in a test tube. If hydrogen sulfide is present the following reaction occurs:

Na2PbO2 + H2S

PbS + 2NaOH

The lead sulfide is black and readily visible. If this reaction does not appear, sulfur is added to the test tube and the mixture shaken again. If mercaptans are present, on shaking they undergo a series of reactions, coloring the hydrocarbon layer first orange, then red and brown, and finally a black precipitate of lead sulfide appears. The overall reactions may be written:

Na2PbO2 + 2RSH (RS) 2Pb + S

(RS) 2Pb + 2NaOH RSSR + PbS


Report (a)

Hydrogen sulfide present. If hydrogen sulfide is detected, report it.

(b)

Sample sour. If a brown or black precipitate forms, the sample contains a relatively high concentration of mercaptans and is reported sour.

(c)

Sample borderline or sweet. If the mercaptan content of the sample is low, observe the sulfur layer and judge as follows: Discoloration of Floating Sulfur

Report

Definitely discolored Barely discolored Not discolored

“sour ” borderline “sweet”

Precaution Use only sufficient sulfur to form a thin film floating on the interface between the sample and the doctor solution.


HIGH VACUUM DISTILLATION OF HIGH BOILING RANGE PETROLEUM PRODUCTS UOP METHOD 76

Scope The distillation apparatus described in this method was devised to provide a means of determining the boiling range of heavy oils. It is intended for the determination, at reduced pressures, of the boiling temperature ranges of petroleum products which have an initial boiling point in excess of 460°F (238°C) and which decompose when distilled at atmospheric pressure. The method is applicable to petroleum products which can be partially or completely vaporized at a maximum liquid temperature of 750°F (399°C), at a pressure in the range of 0.2-0.3 mm of mercury absolute, and which may be condensed as liquids at the pressure of the test. Outline of Method The sample is distilled at a pressure of 0.2-0.3 mm of mercury absolute under conditions which provide approximately one theoretical plate fractionation. Data obtained, converted to 760 mm mercury, allow the preparation of a distillation curve relating volume distilled and boiling point.


CRUDE OIL EVALUATION BY HEMPEL DISTILLATION UOP METHOD 77

Scope This method is for determining the gasoline, naphtha and kerosene content of a crude oil as a guide in operating crude oil topping facilities. Outline of Method The method employs a Hempel column to secure the desired precision in a manner which simulates the results obtained from a commercial distillation. Directions are given to obtain end point gasoline, or naphtha, and kerosene of either specified end point or API gravity.


FRACTIONATION OF PETROLEUM DISTILLATES AND CRUDE OILS UOP METHOD 79

Scope This method describes laboratory fractionation equipment and procedures used in obtaining true boiling point data for petroleum distillates and crude oils. Procedures are also given for obtaining specific boiling range distillates for further analysis. Outline of Method A known volume of a petroleum distillate or a crude oil sample is fractionated in a high efficiency laboratory column. The distillation may consist of: (1) the precision fractionation of normally liquid hydrocarbons to collect fractions for further identification: (2) the quantitative separation of normally gaseous hydrocarbons, such as C3 and/or C4 hydrocarbons from C5 and heavier gasoline or crude oil fractions, and/or (3) the determination of true boiling point (TBP) distillation curves at atmospheric and reduced pressures.


DISTILLATION OF PETROLEUM PRODUCTS ASTM D 86 Scope This method covers the distillation of motor gasolines, aviation gasolines, aviation turbine fuels, special boiling point spirits, naphthas, white spirit, kerosenes, gas oils, distillate fuel oils, and similar petroleum products. A 100 ml sample is distilled under prescribed conditions and systematic observations of thermometer readings and volumes of condensate are made. Definitions 1.

Initial boiling point (lBP) – thermometer reading at instant first drop of condensate falls from the lower end of the condenser tube.

2.

End point (EP) – maximum thermometer reading obtained during the test.

3.

Dry point – thermometer reading observed at instant last drop of liquid evaporates from lowest point in flask. Any drops or film of liquid on side of flask or on thermometer are disregarded.

4

Decomposition point – thermometer reading that coincides with first indication of thermal decomposition of the liquid in the flask, as evidenced by fumes and erratic thermometer readings which usually show a decided decrease after any attempt to adjust the heat.

5.

Percent recovery – maximum percent recovered.

6.

Percent total recovery – combined percent recovery and residue in the flask.

7.

Percent loss – 100 minus percent total recovery.

8.

Percent residue – percent total recovery minus percent recovery.

9.

Percent evaporated – sum of percent recovered and percent loss.


FLASH AND FIRE POINTS BY CLEVELAND OPEN CUP ASTM D 92

Scope This method covers determination of the flash and fire points of all petroleum products except fuel oils and those having an open cup flash below 175°F (79°C). Summary of Method The test cup is filled to a specified level with the sample. The temperature of the sample is increased rapidly at first and then at a slow constant rate as the flash point is approached. At specified intervals a small test flame is passed across the cup. The lowest temperature at which application of the test flame causes the vapors above the surface of the liquid to ignite is taken as the flash point. To determine the fire point, the test is continued until the application of the test flame causes the oil to ignite and burn for at least 5 sec. Precision The following data should be used in judging the acceptability of results (95 percent confidence). Duplicate results by the same operator should be considered suspect if they differ by more than the following amounts:


Repeatability Flash point Fire point

15°F (8°C) 15°F (8°C)

The results submitted by each of two laboratories should be considered suspect if the results differ by more than the following amounts: Reproducibility Flash point Fire point

30°F (17°C) 25°F (14°C)


FLASH POINT BY PENSKY-MARTENS CLOSED TESTER ASTM D 93

Scope These methods cover the determination of the flash point by Pensky-Martens closed-cup tester of fuel oils, lube oils, suspensions of solids, liquids that tend to form a surface film under test conditions, and other liquids. Summary of Method The sample is heated at a slow, constant rate with continual stirring. A small flame is directed into the cup at regular intervals with simultaneous interruption of stirring. The flash point is the lowest temperature at which application of the test flame causes the vapor above the sample to ignite.


WATER AND SEDIMENT IN CRUDE OILS ASTM D 4007

Scope This standard defines a primary centrifuge method and two alternatives for determining the amount of water and sediment in crude oil. It further specifies a base method to be used when centrifuging is not suitable or when the accuracy of a centrifuge method is to be confirmed. Summary of Method The three centrifuge methods involve selection of number of factors such as type of solvent, type and amount of demulsifier, temperature of the sample during testing, and the duration of centrifuging. With many types of oil, the results are not dependent on the selected factors. Those factors which are not the most convenient can be used with good results.


POUR POINT OF PETROLEUM OILS ASTM D 97

Scope The test for pour point is intended for use on any petroleum oil. Summary of Method After preliminary heating, the sample is cooled at a specified rate and examined at intervals of 5°F (or 3°C) for flow characteristics. The lowest temperature at which movement of the oil is observed is recorded as the pour point. Definition Pour point – the lowest temperature, expressed as a multiple of 5°F (or 3°C) at which the oil is observed to flow when cooled and examined under prescribed conditions. Calculation and Report Add 5°F (or 3°C) to the temperature recorded and report the result as the Pour Point, ASTM D 97. Precision Repeatability – Duplicate results by the same operator should be considered suspect if they differ by more than 5°F (or 3°C). Reproducibility – The results submitted by each of two laboratories should be considered suspect only if the two results differ by more than 10°F (or 6°C).


WEATHERING TEST FOR GASES UOP METHOD 155

Scope This is a rapid procedure for the estimation of iso- and normal butane in liquefied butane samples. Propane does not interfere with the analysis and is determined if present in concentrations not exceeding 25%. Pentane and olefins will interfere in the analysis, if present. The test is sufficiently accurate for routine control of plant operations, the apparatus is inexpensive and the technique involved requires little experience. Outline of Method A 94 ml sample of liquefied butanes is drawn into a precooled centrifuge tube. It is then weathered in air to the 90 ml mark. The centrifuge tube and contents are then transferred to a water bath maintained at 60-70°F (15-20°C). Temperature readings are recorded when the liquid level has dropped to the 50- and 15-ml graduation marks of the centrifuge tube. From these temperatures and weathering test curve, the approximate concentrations of propane, isobutane and normal butane are determined.


Precision Duplicate results by the same operator should be considered suspect if they differ by more than the following amounts: Hydrocarbons Propane i-butane n-butane

Maximum Deviation, % + 2.5 + 1.5 + 1.0


SAYBOLT COLOR OF PETROLEUM PRODUCTS (SAYBOLT CHROMOMETER METHOD) ASTM D 156

Scope This method covers the determination of the color of refined oils such as undyed motor and aviation gasoline, jet fuels, naphthas and kerosene. A sample of the liquid is added to a tubular column through which a light source is seen. The color is compared with specified glass standards. The height of the liquid sample is decreased by levels until the color of the sample is lighter than that of the standard. The color number above this level is reported. The range of number is +30 (lightest) to -16 (darkest color). Color standards correspond to sample depth and color number. Precision Duplicate results by the same operator should be considered suspect if they differ by more than 1 color unit. Results submitted by one laboratory should be considered suspect if they differ from that of another laboratory by 2 color units.


HYDROGEN SULFIDE AND MERCAPTAN SULFUR IN LIQUID HYDROCARBONS BY POTENTIOMETRIC TITRATION UOP METHOD 163 Scope This method is for the determination of hydrogen sulfide and mercaptan sulfur in liquid hydrocarbons, such as gasoline, naphtha, light cycle oils and similar distillates. It is applicable to samples containing as little as 1.0 ppm mercaptan sulfur and 1.0 ppm hydrogen sulfide. Attention is called to the fact that an earlier version of this method (163-62) included determination of free sulfur. This has been deleted from the present method. Outline of Method The liquid hydrocarbon sample is titrated potentiometrically in ammoniacal isopropyl alcohol using alcoholic silver nitrate as titrant. A glass reference electrode and a silver-silver sulfide indicating electrode system are used. Estimation of the hydrogen sulfide and mercaptan sulfur content is made from the titration curves. Either an automatic recording titrator or a manually-operated instrument may be used. Free sulfur is a possible interference and instructions are given for analysis of samples containing it. Calculations

Hydrogen sulfide, as S, wt -ppm =

Mercap tan, as S, wt - ppm =

3 16  10 A N SV

32  103 (B - A) N SV


where: A = volume of silver nitrate solution used to reach the sulfide ion end point, ml B = volume of silver nitrate solution used to reach the mercaptide ion end point, ml N = normality of alcoholic silver nitrate solution S = specific gravity of sample at the temperature at which the sample is pipetted V = volume of sample, ml Precautions

Allow enough time for the titration cell to reach equilibrium before recording the volume of silver nitrate solution and the emf when the manual titration is made. When using a recording titrator, add the titrant at a rate of 3.0 ml per minute in the vicinity of the end point, otherwise the end point will be overshot.


VOLATILE NITROGEN BASES AND AMMONIA IN CATALYSTS, DEPOSITS, AND WATER SOLUTIONS UOP METHOD 169

Scope

This is a method for the determination of ammonia and steam-volatile nitrogen bases in catalysts, deposits, and water solutions. The procedure does not distinguish between ammonia and volatile nitrogen bases. Ammonia in the parts per million range can be determined with this method by using 0.005 N sulfuric acid and large samples. Outline of Method

A known quantity of sample is introduced into a Kjeldahl flask, diluted with distilled water, and made alkaline with 50% sodium hydroxide. The volatile nitrogen bases are distilled in a Kjeldahl apparatus and the condensate collected in a boric acid adsorbing solution. This solution is then titrated with standardized sulfur acid using methyl purple indicator.


FLUE GAS ANALYSIS (ORSAT) UOP METHOD 172

Scope

This method is for the quantitative determination of carbon dioxide, oxygen and carbon monoxide in flue gases. Outline of Method

Systematically, the gas sample is admitted into a series of pipets, each containing a reagent for the removal of an individual component. After contact with each reagent, the gas is returned to the buret. The difference in residual volume indicates the amount of component absorbed. In this manner, percentages of carbon dioxide, oxygen and carbon monoxide are determined. Precision

Duplicate results by the same operator should be considered suspect if they differ by more than the following amounts: Carbon dioxide

+ 0.2%

Oxygen

+ 0.3%

Carbon monoxide

+ 0.3%


CONRADSON CARBON RESIDUE OF PETROLEUM PRODUCTS ASTM D 189 Scope

This method covers the determination of the amount of carbon residue left after evaporation and pyrolysis of an oil, and is intended to provide some indication of relative coke-forming propensities. The method is generally applicable to relatively nonvolatile petroleum products which partially decompose on distillation at atmospheric pressure. Petroleum products containing ash-forming constituents as determined by ASTM Method D 482, Test for Ash from Petroleum Products, will have an erroneously high carbon residue, depending upon the amount of ash formed. Summary of Method

A weighed quantity of sample is placed in a crucible and subjected to destructive distillation. The residue undergoes cracking and coking reactions during a fixed period of severe heating. At the end of the specified heating period, the test crucible containing the carbonaceous residue is cooled in a desiccator and weighed. The residue remaining is calculated as a percentage of the original sample, and reported as Conradson carbon residue. Calculation

Calculate the carbon residue of the sample or of the 10 percent distillation residue as follows: Carbon residue = (A  100)/W where: A = weight of carbon residue, g W = weight of sample, g


Report Report the value obtained as "Conradson carbon residue, percent" or as "Conradson carbon residue on 10 percent distillation residue, percent," ASTM D 189.


HYDROGEN SULFIDE, MERCAPTAN SULFUR AND CARBONYL SULFIDE IN HYDROCARBON GASES BY POTENTIOMETRIC TITRATION UOP METHOD 212

Scope

This method is for determining hydrogen sulfide, mercaptan sulfur and carbonyl sulfide in gaseous hydrocarbons and in liquefied petroleum gas (LPG) of ordinary properties. Also covered is the determination of mercaptan in non-ordinary LPG which may contain a wide range of hydrocarbon types from ethane to such gasoline boiling range hydrocarbons as pentane and hexane. The hydrogen sulfide concentration range which can be determined is from 0.3 to several thousand wt-ppm. The method is also applicable to LPG samples containing as little as 1.0 ppm mercaptan sulfur. Outline of Method

The sample, taken either from a sample bomb or directly from a refinery stream, is scrubbed first through a potassium hydroxide solution and then through a monoethanolamine solution. A potentiometric titration of the absorbed hydrogen sulfide and mercaptan sulfur follows. The monoethanolamine solution, which contains the absorbed carbonyl sulfide, is titrated potentiometrically with alcoholic silver nitrate in an acidic titration solvent. The concentration of each item sought is estimated from the titration curve. Precision

Samples containing hydrogen sulfide mercaptan and carbonyl sulfide sulfur: An estimated standard deviation (esd) is not reported since insufficient data are available at present to permit this calculation with at least 4 degrees of freedom.


Samples containing mercaptan sulfur only and appreciable concentrations or pentane and higher boiling materials: The estimated standard deviation based on indicated replicates is shown below. Duplicate results by the same operator should not be considered suspect unless they differ by more than the amounts shown in the "allowable difference" column (95% probability).

Type of Sample

LPG containing appreciable concentrations of pentane and higher boiling materials

No. of Pairs

Mercaptan Level, wt-ppm, S

esd, wt-ppm, S

5

6.0

0.28

Allowable difference, wt-ppm, S

1.1


FLUID CRACKING CATALYST LOADING TEST UOP METHOD 233

Scope

This method is used to determine the proportion of silica-alumina type catalyst in recycle stock from a fluid catalytic cracking plant. Outline of Method

In this procedure the oil-catalyst mixture is washed free of oil with cold benzene, dried and weighed. Note: Toluene may be substituted for benzene. Precautions

Do not conduct the analysis near an open flame. It is preferably carried out under a hood. Be sure the crucible is dry before placing it in the oven. It is advisable to leave the oven door ajar for 5-10 minutes immediately after introducing the crucible. If the catalyst persists in adhering to the Erlenmeyer flask, dry and weigh the flask and add the increase in weight to the catalyst weight in the crucible. Precision

Duplicate determinations by the same operator should agree within 5%.


ACTIVITY TEST FOR FLUID CRACKING CATALYST UOP METHOD 251

Scope

This method is for evaluating the activity of fluid catalytic cracking catalysts relative to a reference catalyst by measuring the conversion of a Mid-continent gas oil to gas and gasoline. The procedure is applicable to both fresh and used catalyst. Outline of Method

A standard gas oil charge stock is cracked over the test catalyst in a fixed-bed operation under carefully controlled standard operating conditions. The conversion to gasoline and gas is measured. The activity, expressed as a percentage, is the ratio of the liquid hourly space velocity used with the test catalyst to the liquid hourly space velocity required with a primary reference catalyst to give the same conversion as that obtained with the test catalyst. The latter value is read from an experimentally-obtained reference catalyst calibration curve showing conversion as a function of space velocity. Notes

1.

Fresh and equilibrium synthetic catalysts are calcined prior to activity testing for 2 hours at 1112°F +25°F (600°C) in a muffle furnace having a vent in the door to permit slow circulation of air. The catalyst is loaded in 150-ml tall-form porcelain crucibles (without covers) filled about one-half full. Fresh catalyst is first dried at about 400°F (204°C) for 2 hours to remove moisture and avoid the spattering of catalyst which otherwise may occur if placed directly in the muffle furnace at 1112°F. Regenerated equilibrium catalysts have substantially all of the carbonaceous deposit (except embedded carbon) removed in calcining. For normal carbon deposits of 0.3 to 0.5 wt-% or less, the activity rating is not


affected by the carbon removal. Abnormal carbon deposits, up to 1 and 1.5 wt%, usually will reduce the weight activity by 1 to 2 numbers. In calcining fresh or used natural catalysts, the temperature employed is that used when regenerating the catalyst in commercial practice.


APPARENT BULK DENSITY OF FLUID CRACKING CATALYST UOP METHOD 254

Scope

This is a procedure for determining the apparent bulk density (ABD) of loosely packed fluid cracking catalyst. Outline of Method

The sample is poured into a weighed, 25 ml cylinder under carefully prescribed conditions. Excess catalyst is scraped off and the full cylinder is weighed. The weight of catalyst, in grams, divided by the volume of the container, in milliliters, is reported as the ABD of the sample. Definition

Apparent bulk density is defined as weight per unit volume. It is an empirical value for the particular type of solid particles to which the determination applies; in this case, fluid (powered) cracking catalyst in a size range of less than 200  m effective diameter. Calculations

Calculate the ABD as weight per unit volume.

where: W = weight of the catalyst, g


PHENOLS AND THIOPHENOLS IN PETROLEUM PRODUCTS BY SPECTROPHOTOMETRY UOP METHOD 262

Scope

This method is intended primarily for the quantitative determination of phenols and thiophenols in gasoline and in refinery caustics, but it also is applicable to crude cresylic acids derived from these refinery caustics. "Phenols" consist of a mixture of phenol, cresols and xylenols. "Thiophenols" denotes the analogous sulfur compounds. Outline of Method

Phenols and thiophenols are extracted from the petroleum fraction with 10% sodium hydroxide solution. The ultraviolet absorption spectrum of the caustic extract is then recorded. A base-line technique is used to compensate for background absorption of the sample. Calibration in a similar manner with known solutions and the application of Beer's Law permits calculation of weight-percent phenols and weightpercent thiophenols. In the presence of excessive concentrations of mercaptans, only the sum of the phenols and thiophenols can be determined directly. However, an accurate value for thiophenols can be obtained from the modified procedure described, involving adsorption of the sample on silica gel followed by selective elution of the thiophenols. The procedure for refinery caustics and crude cresylic acids is identical to that for petroleum fractions except that the extraction step is omitted.


Precautions

Use clean glassware and CP reagents. Shake vigorously in order to obtain quantitative extraction. Since this method of sample preparation is employed chiefly to avoid the oxidation of thiophenols during storage, the extraction must immediately follow sampling. An inert atmosphere and the proper sample container are mandatory. Do not use carbon dioxide as the inert gas.


SAMPLING PETROLEUM AND PETROLEUM PRODUCTS ASTM D 270

Scope

This method covers procedures for obtaining representative samples of stocks or shipments of crude petroleum and petroleum products, except electrical insulating oils, and butane, propane, and other petroleum products that are gases at atmospheric temperature and pressure. Summary of Method

Samples of petroleum and petroleum products are examined by various methods of test for the determination of physical and chemical characteristics. It is accordingly necessary that the samples be truly representative of the petroleum or petroleum products in question. The precautions required to ensure the representative character of the samples are numerous and depend upon the type of material being sampled, the tank, carrier, container or line from which the sample is being obtained, the type and cleanliness of the sample container, and the sampling procedure that is to be used. Each procedure is suitable for sampling a number of specific materials under definite storage, transportation, or container conditions. The basic principle of each procedure is to obtain a sample or a composite of several samples in such manner and from such locations in the tank or other container that the sample or composite will be truly representative of the petroleum or petroleum product.


LOSS ON IGNITION OF CATALYST AT 900°C UOP METHOD 275

Scope

This method is for determining the loss on ignition of fresh or used catalyst or catalyst bases of the various commercial shapes and sizes, when an ignition temperature of 900°C is specified. It is also applicable to other sample types, such as catalyst fines. A representative weighted sample is heated at 900°C to constant weight and the loss in weight calculated as percent loss on ignition. Precision

Duplicate determinations should not differ by more than 0.2% absolute for values below 3.0%, or more than 0.3% to 10.0%. Based on 10 pairs at the 2% loss on ignition level, the standard deviation calculated from the mean range was 0.072%. Time for Analysis

Elapsed time per test is about 21/4 hours (5.0 hours when fines are being analyzed).


EXISTENT GUM IN DIOLEFIN-CONTAINING GASOLINES AND NAPHTHAS BY THE STEAM JET METHOD UOP METHOD 277

Scope

Some gasolines and naphthas contain appreciable concentrations of diolefins or other materials which are sensitive to oxidation. Examples are: (1) pyrolysis naphtha (ethylene coproduct gasoline), (2) certain catalytically cracked gasolines, and (3) certain thermally cracked gasolines. This method is used to determine the existent gum in these types of materials. An inert gas, steam in this case, is used in the evaporation step in order to prevent gum from forming in the test beaker during the evaporation. It has been observed that the ASTM air-jet Method D 381, when used on oxidation-sensitive samples, tends to give falsely high gum values because of gum formation during the evaporation. Outline of Method

The steam jet method is similar to ASTM Method D 381 for Existent Gum in Fuels by Jet Evaporation. The sample in a tarred beaker is placed in a heated metal block apparatus maintained at 325°F (163°C) and evaporated to dryness by a jet of steam which has been superheated to 325°F (163°C). The residue is weighed and reported as milligrams of gum per 100 ml of sample. Precautions

The types of gasolines and naphthas for which this method is intended must be sampled and handled carefully in order to prevent oxidation. For best results it is recommended that: (1) the sample containers be purged with inert gas such as nitrogen, carbon dioxide, or sweet refinery gas prior to taking the samples, (2) any storage between time of sampling and analysis be in the dark (preferably


refrigerated storage), (3) the time lapse between sampling and analysis be as short as possible. Precision

Duplicate results by the same operator should not be considered suspect unless they differ by more than approximately 25% of the result.


API GRAVITY OF CRUDE PETROLEUM AND PETROLEUM PRODUCTS (HYDROMETER METHOD) ASTM D-287

Scope

Using a glass hydrometer the API gravity of crude petroleum and petroleum products which have Reid Vapor Pressures under 26 lbs. can be determined. Gravities are determined at 60°F (15°C), or converted to 60°F, by means of standard tables. Conversion tables are not applicable to nonhydrocarbons or essentially pure hydrocarbons such as the aromatics. Of interest: the ID of the sample cylinder must be at least 25 mm greater than the OD of the hydrometer. The height of the cylinder shall be such that the sample height is 25 mm more than the submerged portion of the hydrometer. Precision

Repeatability – Duplicates should not differ by more than 0.2 degrees API. Reproducibility – Results should not differ by more than 0.5 degrees API when done by different labs. The above judgments are valid providing API gravities were obtained at a temperature not differing from 60°F by more than 18°F (10°C).


ANALYSIS OF REFINERY WATERS FOR pH, IRON, AND COPPER UOP METHOD 314

Scope

This method is designed for measuring the extent of corrosion caused by refinery waters. Copper and iron are determined at a level of 0.1 ppm and higher. The metals may be present as simple dissolved ions, as complex ions with cyanide or other complexing agents, or as part of the suspended solids that often are present. Ammonia, hydrogen sulfide, hydrogen cyanide and organic matter do not interfere. The pH of the water is determined because it usually correlates with the extent of corrosion. Outline of Method

The water sample, including any suspended solids, is concentrated in the presence of sulfuric acid until sulfur trioxide fumes appear and then treated successively with nitric acid and aqua regia. Iron is determined colorimetrically. The lower limit of detection is 0.1 ppm iron. Copper is determined colorimetrically. The lower limit of detection is 0.1 ppm copper in the water sample. The pH of the original water sample is determined with a pH meter using a glasscalomel electrode system. Samples drawn for determination of pH at the refinery should be taken in glass bottles, leaving little or no air space above the liquid, and should be stoppered immediately with rubber or neoprene. The pH should be measured as soon thereafter as possible as it is often subject to fairly rapid change with time. Draw samples to be sent out of the refinery in polyethylene bottles to avoid breakage.


Precision

Insufficient data available to calculate an esd with at least 3.7 degrees of freedom. Time for Analysis

The elapsed time is about 8 hours per sample.


VAPOR PRESSURE OF PETROLEUM PRODUCTS (REID METHOD) ASTM D 323 Scope

This method determines the absolute vapor pressure of volatile crude oil and volatile nonviscous petroleum products, except LPG. The gasoline chamber of the testing apparatus is filled with a chilled sample and connected to the air chamber section which should be at 100°F (37.8°C). The container is then immersed in a constant-temperature bath and shaken periodically until equilibrium is reached. A manometer attached at the end of the cylinder like apparatus is read and corrected if the air chamber temperature is initially at something other than 100°F. Precision

Repeatability – Duplicate results by the same operators should be considered suspect it they differ by more than the following: Range

Repeatability

0-5 psi 5-16 psi 16-26 psi

0.1 0.2 0.3

Reproducibility – Results by two laboratories should be considered suspect it they differ by more than the following: Range

0-5 psi 5-16 psi 16-26 psi

Repeatability

0.35 0.3 0.4


CALCULATION OF UOP CHARACTERIZATION FACTOR AND ESTIMATION OF MOLECULAR WEIGHT OF PETROLEUM OILS UOP METHOD 375

Scope

This method is for determining the UOP Characterization Factor, which is indicative of the general origin and nature of a petroleum stock. Values of 12.5 or higher indicate a material predominantly paraffinic in nature. Highly aromatic materials have characterization factors of 10.0 or less. This method also may be used to estimate the molecular weight of typical petroleum fractions. It is not intended for estimating the molecular weight of a pure hydrocarbon compound. Outline of Method

This method gives directions for estimating: 1.

The UOP-Characterization Factor from API gravity and Engler distillation,

2.

The UOP Characterization Factor from API gravity and kinematic viscosity at a temperature of 100°, 122° or 210°F (37.8°, 50° or 98.9°C), and

3.

The molecular weight from API gravity and Engler distillation.

Definitions

The UOP Characterization Factor, K, of a hydrocarbon is defined as the cube root of its absolute boiling point, in degrees Rankine, divided by its specific gravity at 60°F.


Molecular weight, as employed herein, is the average molecular weight of a petroleum fraction and not that of a single, pure compound. Cubic average boiling point is the cube of the sum of the products of the volume fraction multiplied by the cube root of the boiling point of each component expressed in degrees Rankine. Mean average boiling point is the arithmetic average of the true molal boiling point and the cubic average boiling point, expressed in degrees Fahrenheit. True molal average boiling point is the sum of the products of the mol fraction multiplied by the boiling point of each component.


NITROGEN IN PETROLEUM DISTILLATES AND HEAVY OILS BY ACID EXTRACTION OR DIRECT KJELDAHL PROCEDURE UOP METHOD 384

Scope

This acid extraction method is specifically intended for the determination of combined nitrogen in petroleum distillates and heavy oils in concentrations ranging from 0.2 to 20 ppm. Nonextractable samples, and samples containing nitrogen concentrations ranging from 20 ppm to several percent, can be handled by a direct Kjeldahl analysis. The types of nitrogen compounds which can be determined are those which usually are determined by a macro-Kjeldahl procedure and include amines, amides, pyridines, pyroles and quinolines. The method does not apply to organic nitrocompounds nor to those containing a -N = N- linkage. Precision

Duplicate determinations run in the same laboratory by the same operator on the same equipment should not differ from the mean by more than the percent relative in the table below: Total Nitrogen, ppm

Percent, Relative

0.5 5 45 500

20 11 5 2

The estimated standard deviation was calculated to be 0.35 at the 16-ppm level, based on 5 replicate samples; 1.38 at the 54-ppm level, based on 6 replicates.


TRACE METALS IN OILS BY WET ASH-SPECTROGRAPHIC METHOD UOP METHOD 389

Scope

This method is applicable to the determination of iron, nickel, vanadium, lead, copper, sodium and molybdenum, specifically, in crude petroleum and such fractions as gas oils, fuel oils and let fuels. Additionally, manganese, chromium, magnesium, tin, calcium, aluminum and zinc may be determined by this method. Each of these elements can be determined over the concentration range of 0.02 to 1000 ppm if a 50-g sample of oil is ashed. Higher or lower concentrations can be determined by ashing appropriately-sized samples. Outline of Method

Cobalt and potassium are added to the oil sample as internal standards and spectroscopic buffers. The oil sample is then coked with fuming sulfuric acid, ignited and ashed at 1000°F (538°C), treated with aqua regia and the ash dissolved in dilute hydrochloric acid. A spark spectrum of the solution is obtained by means of a rotating-disk electrode technique. A microphotometer is used to measure the densities of selected metal lines. Concentrations are determined from standard calibration curves. Precision

The estimated standard deviations (esd) for various concentration levels of the metals determined by this method are shown below. Duplicate results by the same operator should not be considered suspect unless they differ by more than the amounts shown in the "allowable difference" column (95% probability).


Metal Level, ppm

No. of Pairs

esd, ppm

1 10 50 200

5 5 5 5

0.09 0.59 3.2 13

Allowable Difference, ppm

0.3 2.2 12 48


TRACE METALS IN PETROLEUM AND ORGANIC PRODUCTS BY WET-ASHING; FLAME PHOTOMETER AND SPECTROPHOTOMETRIC METHODS UOP METHOD 391

Scope

This method is for determining trace concentrations (parts per million) of vanadium, nickel, iron, copper and sodium in petroleum products such as crude oils and residues, and varied organic compounds (including nitroanilines, amines, chlorobenzenes, phenols and other related materials) produced or used in chemical manufacturing. Other elements commonly found in these materials do not interfere. Outline of Method

The sample is wet-ashed with fuming sulfuric acid and the coke is burned off in a muffle furnace. The inorganic residue remaining is dissolved in acid and diluted to a given volume. Vanadium, nickel, iron and copper are determined spectrophotometrically, and sodium is determined by flame photometry, using aliquots from the acid solution. Precaution

The sulfuric acid concentration in the standard sodium solutions must be the same as that in the sample solutions. If a different concentration of sulfuric acid is used for the sample solution than that specified in this method, the calibration curve must be prepared with sodium standard solutions containing this same sulfuric acid concentration.


Precision

Duplicate determinations by the same operator should not be considered suspect unless they differ by more than the following amounts:

Allowable Difference for Duplicates, ppm

Range, ppm

(Vanadium, Nickel, Iron, Copper) 0 to 2 >2

— —

0.1 ppm of metal 5% of the mean

(Sodium) 0 to 2 >2

— —

0.2 ppm 10% of the mean

Insufficient data are available to calculate a standard deviation for the metals listed.


PARTICLE SIZE DISTRIBUTION BY MICROMESH SIEVES UOP METHOD 422

Scope

This method is for the determination of particle sizes of fluid cracking catalysts by means of calibrated sieves having uniform precise square openings. It may also be used for determining particle sizes of other powdered materials. The method provides for the classification of particles in a range of sizes from about 20 to 149 microns. The method is a modification of Shell Development Company, Method EMS 5Z2/61. Outline of Method

A representative sample is appropriately humidified and then placed on the top sieve of a calibrated set of precision sieves. The set is mounted on a specified shaking apparatus and shaken for 20 minutes. The weight of catalyst retained on each sieve and in the pan is determined, and the percentage of the sample retained on each sieve and in the pan is calculated on the basis of the sum of weights of the recovered fractions.


SINTERING INDEX OF USED FLUID CRACKING CATALYST UOP METHOD 424

Scope

This method provides a means of separating and measuring a fraction of used fluid cracking catalyst having a reduced apparent density resulting from the pores having been sealed off by localized overheating. Precision

Duplicate results by the same operator should be considered suspect if they differ by more than 1% absolute. Duplicate results by different operators should be considered suspect if they differ by more than 2% absolute.


SURFACE AREA, PORE VOLUME AND PORE DIAMETER OF POROUS SUBSTANCES BY NITROGEN ADSORPTION UOP METHOD 425

Scope

This method is for the determination of surface areas of porous substances by a two-point system, using the basic Brunauer-Emmett-Teller (B.E.T.) theory of multilayer adsorption, and pore volumes of the substances, using the capillary condensation theory. Surface areas greater than 10 m2/g may be determined, as well as the volume of pores up to a diameter of 600 Angstroms. Pore diameter is calculated from surface area and pore volume. Outline of Method

The surface area of a substance is determined by measuring the volume of nitrogen gas adsorbed at liquid nitrogen temperature and relative pressures of 60/760 and 160/760, and applying the B.E.T. theory. The pore volume is determined by measuring the amount of gaseous nitrogen condensed in the pores at liquid nitrogen temperature and a relative pressure of 735/760. It can be shown by use of the Kelvin equation for capillary condensation and the proper correction for multilayer adsorption that this volume is that required to fill all pores of diameter less than 600 Angstroms.


Precision Surface Area

Duplicate determinations should not differ by more than 8% from the mean. Based on 2 sets of 5 replicate determinations, the standard deviation at the 500-600 m2/g level was 12. Based on 2 sets of 5 replicate determinations, the standard deviation at the 100-200 m2/g level was 2. Pore Volume

Duplicate determinations should not differ by more than 0.04 ml/g. Based on 4 sets of 5 replicate determinations, the standard deviation in the range of 0.30 to 0.90 m2/g was 0.013. Pore Diameter

Duplicate determinations should not differ by more than 5% from the mean. Based on 2 sets of 5 replicate determinations, the standard deviation at the 50-60 Angstrom level was 1. Based on 2 sets of 5 replicate determinations, the standard deviation at the 90-130 Angstrom level was 3.5.


Time for Test

The elapsed time for 1 sample is 9 hours. The elapsed time for 6 samples is 10.5 hours.


KINEMATIC VISCOSITY OF TRANSPARENT AND OPAQUE LIQUIDS (AND THE CALCULATION OF DYNAMIC VISCOSITY) ASTM D 445

Scope

This method covers the determination of the kinematic viscosity of liquid petroleum products, both transparent and opaque, by measuring the time for a volume of liquid to flow under gravity through a calibrated glass capillary viscometer. The dynamic viscosity can be obtained by multiplying the measured kinematic viscosity by the density of the liquid. Summary of Method

The time is measured in seconds for a fixed volume of liquid to flow under gravity through the capillary of a calibrated viscometer under a reproducible driving head and at a closely controlled temperature. The kinematic viscosity is the product of the measured flow time and the calibration constant of the viscometer. Precision

The following precision applies to clean, transparent oils tested between 60° and 212°F (15° and 100°C). Repeatability – Duplicate results by the same operator, using the same viscometer, should be considered suspect if their difference is greater than 0.35 percent of their mean. Reproducibility – The results submitted by each of two laboratories should not be considered suspect unless their difference is greater than 0.7 percent of their mean. ASH FROM PETROLEUM PRODUCTS


ASTM D 482

Scope

This method covers the determination of ash from distillate and residual fuels, gas turbine fuels, crude oils, lubricating oils, waxes, and other petroleum products, in which any ash-forming materials present are normally considered to be undesirable impurities or contaminants. The method is limited to petroleum products which are free from added ash-forming additives, including certain phosphorus compounds. Summary of Method

The sample contained in a suitable vessel is ignited and allowed to burn until only ash and carbon remain. The carbonaceous residue is reduced to an ash by heating in a muffle furnace at 1427°F (775°C), cooled and weighed. Report

Report the result to two significant figures as the ash, ASTM D 482, stating the weight of the sample taken.


SAMPLING OF GASOLINES, DISTILLATE FUELS AND C3-C4 FRACTIONS UOP METHOD 516

Scope

This is an outline of the sampling techniques necessary for obtaining, in a stainless steel cylinder, an air-free sample of a liquid hydrocarbon. Samples may consist of liquefied natural gases, high vapor pressure natural gasolines, various liquefied petroleum gases, air unstable gasolines or other liquid hydrocarbon products which require rigorous exclusion of air. Outline of Method

Proper sampling techniques are specified to obtain an uncontaminated, air-free sample in a suitable container; viz., a double-valved, stainless steel cylinder. The cylinder is mounted vertically near the sampling point and connected to the plant sample line by steel tubing or pipe. The connecting lines are flushed with the sample, which is then allowed to flow upward through the cylinder. Several volumes of sample are discarded before closing the sample cylinder valves of the cylinder. Identification and Shipment

Properly identify each sample by attaching a tag to the cylinder giving company name and location, time and date of sampling, identity of stream sampled, sample pressure and tests required. Crate the cylinder to protect the valves from being opened or damaged in shipment. Affix a "Red Label" or other proper shipping label to the crate.


Notes and Precautions

Samples taken in containers found leaking either during or after sampling should be discarded and new samples taken in leak-free cylinders. Consult the current Interstate Commerce Commission or other appropriate authorities for regulations for applicable specification for shipment. Consult ASTM Methods D 270, D 1145 and D 1265, if necessary, for procedures for measurement and sampling of petroleum and petroleum products. For liquid samples taken in the above manner it is mandatory to immediately provide a safe outage in the cylinder to prevent thermal expansion and cylinder rupture. Follow the recommendations in this method to safely provide that outage.


GAS ANALYSIS BY GAS CHROMATOGRAPHY UOP METHOD 539

Scope

This method is for determining the composition of a wide variety of gaseous hydrocarbon mixtures obtained from refining processes or from natural sources, including minor concentrations of the composite of C5 olefins and C6+ hydrocarbons. 1-Butene is not resolved from isobutylene and argon is determined as a composite with oxygen. The lower limit of detection for a single component is 0.1 mol-%. Outline of Method

All the constituents of a total gas sample cannot be resolved by a single chromatographic column. Therefore in this procedure, 3 columns are connected in series with appropriate coupling valves. Each column is used to separate a specific portion of the total sample. The first column is able to resolve gases in the C3-C5 boiling range. The second column separates the components in the intermediate boiling range – carbon dioxide, ethylene, and ethane – while the third resolves the light gases – hydrogen, oxygen + argon (composite), nitrogen, methane, and carbon monoxide. Precision

Based on 7 replicate determinations, the estimated standard deviations (esd) in Table 1 were calculated for the components listed.


TABLE 1 GAS ANALYSIS BY GAS CHROMATOGRAPHY

LEVEL, MOL-%

ESD, MOL-%

ALLOWABLE DIFFERENCE, MOL-%

HYDROGEN NITROGEN METHANE CARBON MONOXIDE

10 20 10 10

0.10 0.02 0.03 0.14

0.36 0.07 0.09 0.49

CARBON DIOXIDE ETHANE PROPYLENE n-BUTANE

10 10 10 10

0.03 0.04 0.02 0.03

0.10 0.14 0.05 0.09

3 4

0.03 0.04

0.11 0.14

COMPONENT

ISOPENTANE C5 OLEFINS/C6 PLUS


METALS IN CRACKING CATALYST BY EMISSION SPECTROSCOPY UOP METHOD 546 Scope

This method is designed for the determination of metal impurities in silica-alumina cracking catalyst containing 0 to 30% alumina and less than 1% Na. Impurities are determined in the following ranges: Cu and Mo, 0.001 to 1.0%; Ni, V, Mn, Cr, Pb, Sn, and Ti, 0.005 to 1.0%; Fe and Mg, 0.01 to 2.0%; Zn and Ca, 0.05 to 2.0%; Na, 0.05 to 1.0%. Outline of Method

Samples of the unknown diluted in graphite are burned to completion using a 6-amp dc arc. The photographed spectra are then examined to determine the elements sought. Iron, nickel, and vanadium concentrations are quantitatively determined by densitometry, using an internal standard, while the concentrations of the remaining metals are semi-quantitatively determined by visual comparison with spectra from known samples. Precision

Duplicate determinations for Fe, Ni, and V should not differ from the mean by more than 10% in the 0.01 to 1.0% range. Based on 10 replicates, the estimated standard deviation (esd) was calculated to be: Elemental

Level

esd

Fe Ni V

0.6 0.02 0.05

0.047 0.0017 0.0075


CYANIDE AND THIOCYANATE IN REFINERY WATERS AS CYANIDE UOP METHOD 682

Scope

This method is for determining cyanide and thiocyanate in refinery water, both of which are then calculated as weight-ppm cyanide. The lower limit of detection is about 0.02 wt-ppm in samples containing no sulfur and 0.8 wt-ppm in those containing 1% sulfide. Outline of Method

A sample of refinery water is placed in a flask and hydrogen sulfide added. The sample is then acidified with hydrochloric acid, bromine water is added and the mixture allowed to stand until clear. (1)

Reactions of bromine water with sample:

HCN + Br2

KCNS + 4 Br2 + 4 H2O

H2S + 4 Br2 + 4 H2O

CNBr + HBr

KBr + CNBr + H2SO4 + 6 HBr

8 HBr + H2SO4

The excess bromine is destroyed with arsenious acid and a pyridine-benzidine mixture is added. The sample is then allowed to stand for 10 minutes.


(2)

Reaction of the pyridine and benzidine with sample:

(a) + CNBr N

N

Br

CN

+

(b) + 2H 2 NC 6 H 4 -C 6 H 4 -NH N CN

Br

2

H

N

NHC 6 H 4 C 6 H 4 -NH

2

+ (Br)

-

+ H 2 N-CN

C 6 H 4 -C 6 H 4 NH 2

After standing the sample is diluted to volume. The absorbance is read on a spectrophotometer at 525 m and calculated as weight-ppm cyanide.


SULFIDE IN REFINERY WASTE WATER USING CADMIUM CHLORIDE UOP METHOD 683

Scope

This method is for determining sulfide in refinery waste water. It cannot be used on caustic solutions. Good precision can be obtained when a titrant volume of 10 ml or more is used. The estimated standard deviation of 0.013 was obtained on a sample containing 0.8 wt-% S. However, with titrant volumes of less than 10 ml, the results may be high by as much as 5% of the value found. Outline of Method

The sample is pipetted into 100 ml of water containing a small amount of ammonium hydroxide. It is then titrated potentiometrically with standard cadmium chloride and the result calculated as weight-percent sulfur. The titration curve may have 2 "breaks". The possible reactions are as follows: To the first "break":

Cd++ + (NH4)2S

CdS + 2NH4+

Between the first and second "break":

either (a)

Cd++ + (NH4)2S • S4

(b)

Cd++ + (NH4)2S • S4

CdS + 4S + 2NH4+ CdS3 + 2NH4+

However, in this method the sulfide is calculated to the second "break" with the equivalent weight of sulfur being 16.


Precision

The estimated standard deviation (esd) at the 0.8% sulfur level is shown below. Duplicate results by the same operator are not considered suspect unless they differ by more than the amount shown in the "allowable difference" column (95% probability).

Type of Sample

Sulfur Level, wt-%

No. of Replicates

esd, wt-% S

Allowable Difference, wt -%

Solution of (NH4)2S

0.8

11

0.013

0.026


CARBON ON CATALYST BY "LECO" WR-12 WIDE RANGE CARBON DETERMINATOR UOP METHOD 703

Scope

This method gives supplementary instructions for the determination of 0.01 to 80.0 wt-% carbon on catalysts. It is necessary that the analyst be provided with the history of the sample so that the proper sample preparation can be applied. Outline of Method

The sample is weighed in a ceramic crucible, mixed with accelerators and burned in an induction furnace using oxygen carrier gas. The products of combustion pass through a purifying train consisting of a dust trap, antimony metal for chloride removal, manganese dioxide for sulfur removal, and a heated catalyst lube for conversion of carbon monoxide to carbon dioxide and hydrogen to water. The carbon dioxide is adsorbed on molecular sieves at ambient temperature. After the adsorption period is ended, the molecular sieve column is rapidly heated to 600°F (316°C) and the carbon dioxide eluted and carried through the measuring thermistor by an auxiliary flow of oxygen. The output of the thermistor bridge is integrated and read on a digital voltmeter and converted to percent carbon by calculation.


Precision

The estimated standard deviation (esd) for carbon at different concentrations is shown below. For the 95% probability level, duplicate results by the same operator should not be considered suspect unless they differ by more than the amounts shown in the "allowable difference" column.

Carbon Level, %C

Number of Pairs Used

Number of Replicates

esd, %C

Allowable Difference, % C

0.01-0.6 0.6-2 2-5 20

5 7 7 5

8 – – –

0.005 0.015 0.022 0.334

0.02 0.05 0.07 1.25


GAS ANALYSIS BY GAS CHROMATOGRAPHY USING A TWO-INJECTION TECHNIQUE UOP METHOD 709

Scope

This method is for determining most of the components, including low concentrations of total C6+ hydrocarbons, normally found in a wide variety of gaseous hydrocarbon mixtures obtained from refining processes or from natural sources. Butene-1 is not resolved from isobutylene and argon is determined as a composite with oxygen. The lower limit of detection for a single component is 0.1 mol-%. Outline of Method

All the constituents of a total gas sample cannot be resolved by a single chromatographic column. Therefore, in this procedure, 2 columns are connected in series with appropriate coupling valves to a thermal conductivity detector. Each column is used to separate a specific portion of the total sample. The first column is able to resolve gases in the C2-C5 range. The second column separates the light gases: H2, O2 + A (composite), N2, CH4 and CO. Precision

Based on 6 replicate determinations, the estimated standard deviation (esd) in Table 2 was calculated for the gases listed. Duplicate results should not differ by more than the allowable difference indicated (95% probability).


TABLE 2 GAS ANALYSIS BY GAS CHROMATOGRAPHY USING A TWO-INJECTION TECHNIQUE

LEVEL, MOL-%

ESD, MOL-%

ALLOWABLE DIFFERENCE, MOL-%

HYDROGEN

21

0.186

0.68

NITROGEN

7

0.087

0.32

METHANE

12

0.180

0.66

CARBON MONOXIDE

8

0.266

0.97

PROPANE

8

0.077

0.28

PROPYLENE

6

0.203

0.74

ISOBUTANE

7

0.158

0.58

n-BUTANE

6

0.074

0.27

ISOBUTYLENE

6

0.264

0.96

COMPONENT


OILY MATERIAL IN REFINERY WATERS BY INFRARED SPECTROPHOTOMETRY UOP METHOD 726 Scope

This method is for determining oily material in refinery waters. It is especially suitable for routine monitoring of effluent streams known to be relatively constant as to the nature of the oily material present. The method has an advantage over physical test methods in that volatile hydrocarbons can be determined and are not lost. Outline of Method

Oily material is extracted from water with carbon tetrachloride. The infrared absorbances of the extract are determined at 2860 cm-1 (3.50 m) and 2930 cm-1 (3.42 m), and these are used to calculate the concentration of oily matter. Definition

Oily material means any substance containing -CH, -CH2-, or -CH3 groups which show infrared adsorption bands at 2860 cm-1 (3.50 m) and 2930 cm-1 (3.42 m) and which is extractable from acidified water with carbon tetrachloride. Sensitivity

The sensitivity of this method is about 1 ppm. The method can be extended to include oil concentrations of less than 1 ppm by using cells of longer path length (5and 10-cm cells are common), larger samples of the water and smaller volumes of carbon tetrachloride for the extraction. Precision and Accuracy

Results are reported to the nearest part per million. Relative esd as reported in the reference API method is 5%. Duplicate results should be considered suspect if they differ by more than 20% of the average value.


DISTILLATION OF PETROLEUM PRODUCTS AT REDUCED PRESSURES ASTM D 1160

Scope

This method covers the determination, at reduced pressures, of the boiling temperature ranges of petroleum products which can be partially or completely vaporized at a maximum liquid temperature of 750°F (400°C) at pressures down to 1 mm Hg, absolute. Summary of Method

The sample is distilled, at some predetermined and accurately controlled pressure, between 1 mm Hg, absolute, and atmospheric, under conditions which provide approximately one theoretical plate fractionation. Data are obtained from which a distillation curve relating volume distilled and boiling point at the controlled pressure can be prepared. Significance

Some petroleum products decompose when distilled at atmospheric pressure. This distillation method is for determination of distillation characteristics of such products. The apparatus and conditions of test provide approximately one theoretical plate fractionation. Results by this method are not comparable with those of other ASTM procedures for the determination of boiling point ranges of petroleum products, such as ASTM Method D 86 for Distillation of Petroleum Products.


SULFUR IN PETROLEUM PRODUCTS (LAMP METHOD) ASTM D 1266

Scope

This method determines the total sulfur in liquid petroleum products in concentrations above 0.002 weight percent. The procedure involves burning the sample in a closed system using a suitable lamp apparatus. An artificial atmosphere composed of 70% CO, and 30% O2 is used to burn the sample to prevent formation of nitrogen oxides. The oxides of sulfur are then oxidized to sulfuric acid. Sulfur as sulfate is determined acidimetrically by titration or gravimetrically by precipitation as barium sulfate. Precision

Samples should contain in the range of 0.01% to 0.4% sulfur. Repeatability – Duplicate results by the same operator should not differ by more than 0.005%. Reproducibility – Results by two laboratories should not differ by more than 0.010 + 0.025 S, where S = the total sulfur content, weight percent, of the sample.


ASTM COLOR OF PETROLEUM PRODUCTS (ASTM COLOR SCALE) ASTM D 1500

Introduction

This method has replaced the former ASTM Method D 155, Test for Color of Lubricating Oil and Petrolatum by Means of ASTM Union Colorimeter. Method D 155 was withdrawn as an ASTM Tentative on July 1,1960. Method D 1500 is better than the former Method D 155 in three respects: (1) the glass standards are specified in fundamental terms; (2) the differences in chromaticity between successive glass standards are uniform throughout the scale; and (3) the lighter colored standards more nearly match the color of petroleum products. Scope

This method covers the visual determination of the color of a wide variety of petroleum products such as lubricating oils, heating oils, diesel fuel oils, and petroleum waxes. Report

Report as the color of the sample, the designation of the glass producing a matching color, for example: 7.5 ASTM Color. If the color of the sample is intermediate between those of two standard glasses record the designation of the darker glass preceded by the letter "L", for example: "L7.5 ASTM Color". Never report the color as being darker than a given standard except those darker than 8, for example: ASTM Color. If the sample has been diluted with kerosene, report the color of the mixture followed by the abbreviation "Dil," for example: "L7.5 Dil ASTM Color".


Precision

The results obtained by different operators in the same laboratory should not vary by more than 0.5 number, and the same variation should apply for determinations between different laboratories at the 96 percent confidence level.


SULFUR IN PETROLEUM PRODUCTS (HIGH-TEMPERATURE METHOD) ASTM D 1552

Scope

This method covers two procedures for the determination of total sulfur in petroleum products, including lubricating oils containing additives, and additive concentrates. The method is applicable to samples boiling above 350°F (177°C) and containing not less than 0.06 percent sulfur. Chlorine in concentrations less than 1 percent does not interfere. Nitrogen when present in excess of 0.1 percent may interfere; the extent of such interference may be dependent on the type of nitrogen compound as well as the combustion conditions. The alkali and alkaline earth metals, as well as zinc, phosphorus, and lead, do not interfere. Summary of Method

The sample is burned in a stream of oxygen at a sufficiently high temperature to convert about 97 percent of the sulfur to sulfur dioxide. A standardization factor is employed to obtain accurate results. The combustion products are passed into an absorber containing an acid solution of potassium iodide and starch indicator. A slight blue color is developed in the absorber solution by the addition of standard potassium iodate solution. As combustion proceeds, bleaching the blue color, more iodate is added. The amount of standard iodate consumed during the combustion is a measure of the sulfur content of the sample. Report

Report the results of the test to the nearest 0.01 percent.


Precision

The following criteria should be used for judging the acceptability of results (95 percent confidence): Repeatability – Duplicate results by the same operator should be considered suspect if they differ by more than the following amounts: Sulfur, weight percent (Range)

Repeatability

0 to 0.5 0.5 to 1.0 1.0 to 2.0 2.0 to 3.0 3.0 to 4.0 4.0 to 5.0

0.05 0.07 0.10 0.16 0.22 0.24

Reproducibility – The result submitted by each of two laboratories should not be considered suspect unless the two results differ by more than the following amounts: Sulfur, weight percent (Range)

Reproducibility

0 to 0.5 0.5 to 1.0 1.0 to 2.0 2.0 to 3.0 3.0 to 4.0 4.0 to 5.0

0.05 0.11 0.17 0.26 0.40 0.54


KNOCK CHARACTERISTICS OF MOTOR FUELS BY THE RESEARCH METHOD ASTM D 2699 Scope

This method determines the knock characteristics of motor gasolines, intended for use in spark-ignition engines. A Research octane number (RON) of 100 or lower is the volume percent of iso-octane in a blend with n-heptane that matches the knock intensity of the unknown sample. For numbers above 100, a comparison is made to iso-octane and milliliters of tetraethyllead required to match knock intensity. Summary of Method

The RON of a gasoline is determined by comparing its knocking tendency with those for blends of reference fuels of known octane. Knock intensity is measured by an electronic detonation meter on a testing unit consisting of a single cylinder engine. Repeatability

Data for limits on duplicate results by the same operator have not been developed. Reproducibility

Results by different laboratories should be considered suspect if their difference is greater than the limits shown below: Ave. Research Octane No. Level

Limits

80 85 90 95 100 105

1.2 0.9 0.7 0.6 0.7 1.1


KNOCK CHARACTERISTICS OF MOTOR FUELS BY THE MOTOR METHOD ASTM D 2700

Scope

This method covers the determination of the knock characteristics of motor and aviation-type gasolines, intended for use in spark-ignition engines, in terms of ASTM motor octane numbers. Summary of Method The ASTM Motor Octane Number of a fuel is determined by comparing its knocking tendency with those for blends of ASTM reference fuels of known octane number under standard operating conditions. This is done by varying the compression ratio to obtain standard knock intensity, as measured by an electronic detonation meter. Definitions ASTM Motor Octane Number of motor and aviation-type gasolines of 100 and below – The volume percent, to the nearest tenth, of iso-octane (equals 100.0) in a blend with n-heptane (equals 0.0) that matches the knock intensity of the unknown sample, when compared by this method. ASTM Motor Octane Number of a motor gasoline above 100 – The value, to the nearest tenth, that corresponds to the equivalent engine rating in terms of milliliters of tetraethyl lead in iso-octane.


Precision

Repeatability – Data to determine acceptable limits for duplicate results obtained by the same operator have not been developed. Reproducibility – The difference between two, single, and independent results, obtained by different operators working in different laboratories on identical test material would, in the long run, and in the normal and correct operation of the test method, exceed the following values in only one case in 20 (see table). Average Motor Octane Number Level

Limits Octane Number

80 85 90 95 99 100.1 105

1.2 1.1 1.0 1.1 1.5 1.1 1.8


SAMPLING STACKS FOR PARTICULATE MATTER ASTM D 3685

Scope

This method covers the sampling and determination of particulate matter in stack gases. Significance

The following procedure describes a method of sampling in stacks and flues which has been standardized within the limits of the many conditions which are encountered in the normal course of sampling stacks. No one procedure or set of apparatus will apply to all problems. The recognition of one set of apparatus which will satisfy a number of commonly encountered conditions will often leave many other problems unanswered. The objective has been to select apparatus that will give a reliable answer when applied to a variety of problems. For compliance with regulations in the United States EPA method #5 is usually required. Check with local authorities before using any lab method required for compliance with environmental regulations.

157048 Procedures Page 1

PROCEDURES INTRODUCTION The procedures given in this section are general instructions which may serve as a guide for each unit. They cannot be specific because of the variations in the design and construction of each Fluid Catalytic Cracking Unit. These methods should be used by each refiner to develop a detailed set of operating instructions for his particular unit. Although basic procedures will be fairly consistent from unit to unit, the FCC is such a flexible process that several different strategies exist for operating an FCC unit. Individual refinery requirements and product markets should be carefully considered to determine the most economic strategy to employ for the FCC unit. These procedures will cover the basic operating steps for startup, shutdown and emergency situations. Specific operating strategies for maximum economic benefit will be left to the individual refiner. This section is divided according to topic. These subsections are: A. B. C. D. E. F. G.

Refractory Dryout -- Initial Startup Normal Startup Establish Normal Operating Conditions Normal Shutdown Emergency Procedures Catalyst Handling Special Operations

Throughout these procedures, references will be made to recirculation catalyst, combustor and upper regenerator, which implies a high efficiency style combustor regenerator, catalyst coolers, lift gas and alternate riser termination devices. Not all units will have these features. Where not applicable to your specific unit, certain steps in the procedures will not be followed. The basic principles for unit operation are the same, however, regardless of the specific features of the unit. Refer to Figures 1 through 3 for diagrams of the unit showing the key streams used in the following procedures.


A. REFRACTORY DRYOUT -- INITIAL STARTUP The startup operations on the FCC Unit require careful coordination and planning. The reactor-regenerator, main column, and Gas Concentration Unit must all be inspected to verify proper construction and that the equipment is ready for use. Utility systems should be commissioned and all catalyst and required chemicals on site. All instruments must be calibrated and control valves ready for service. Inspection and precommissioning work is covered in depth in the UOP FCC Operations Manual. The startup procedures that follow assume all equipment is ready for use. The protective refractory lining in the regenerator, riser, and parts of the reactor must be cured and dried before it is used. Failure to do this properly can result in premature failure of this important protection. The dryout procedure is time consuming, but once the refractory has been cured and dried, subsequent startups will use a shorter procedure. Minor repair work to the refractory will not necessarily require a full high temperature dryout, but sufficient time must be allowed for curing and air drying. In all cases the refractory manufacturer's guidelines should be consulted. The initial dryout isolates the reactor and regenerator from the main column at the vapor line blind. The vapor line vent opened upstream of the blind is opened to allow air flow from the reactor to atmosphere. Any work remaining on the column, gas compressor and Gas Concentration section can continue while the dryout is in progress. After application, the refractory lining contains three types of water: free, absorbed, and chemically bonded. The curing step at ambient conditions allows time for the chemical bonding to be completed. The free water must then be removed slowly to prevent rapid expansion to steam that could damage the refractory. The absorbed water is removed more slowly during a longer, high temperature step. AlI of the


refractory must be cured and dried; be sure that air is directed to every part of the reactor-regenerator system. The dryout may be summarized as: 1.

Controlled heating of regenerator and reactor with hold steps for water removal.

2.

Reactor-regenerator assembly field test for 24 hours.

3.

Controlled cooling of regenerator and reactor.

The source of heat for the dryout may be the air blower and direct fired air heater, as in the following procedure, or it may be supplied with internal local heaters. The choice may be dictated by the particular vendor preference and/or availability of suitable equipment. Procedure The following procedure is intended to supplement the refractory vendor procedure. During the dryout, several hold periods are maintained at a constant temperature. The time period for the hold may vary depending on the type and thickness of the refractory lining. Where the vendor procedure calls for more conservative steps, those steps should be followed. Should the following steps be more conservative than the vendor, further discussion is warranted to confirm that the dryout will be adequate. In all cases, the procedure of the vendor responsible for the application/performance of the refractory shall govern. 1.

Refractory Curing Step a.

The curing procedure shall begin immediately after installation and shall last a minimum of 24 hours. For the 4" low density gunned lining in the regenerator, water or membrane curing steps may also be included.


b.

2.

3.

During the curing procedure, the temperature of the shell and lining shall be kept above 60°F (16°C).

Refractory Air Drying a.

After curing, the refractory lining shall be tested by tapping with a one pound hammer at one foot intervals over the entire lining surface. Any voids or dry filled spaces will emit a dull sound and these areas shall be removed and replaced.

b.

After the 24 hour curing period, the refractory lining shall be air dried for at least another 24 hours by natural or forced ventilation.

Preparation Steps for Dryout a.

Isolate the structure with blinds in the feed, torch oil, fuel or gas to the CO boiler and air heater, steam to the riser and stripper, and reactor vapor line.

b.

Double block and bleed the riser blast points, flue gas quench nozzles and all steam purge points.

c.

Remove the blankoffs on the catalyst loading lines and the drain points in the flue gas system.

d.

Remove the blankoff on the vent nozzle on top of the reactor shell and install a gate valve. Attach a pipe stack approximately six feet (two meters) long to the valve for personnel protection. This valve will only be used if the reactor vapor line vent does not pass enough hot air to adequately heat up the reactor or if the riser termination device is highly contained so that there is little or no air flow through the reactor shell. Open the vapor line vent valve upstream of the vapor line blind.


4.

5.

e.

Connect the reactor skin thermocouples to a recorder for temperature monitoring.

f.

Open air purges to all instrument DA points and slide valve packing glands.

g.

Commission the steam systems of the flue gas cooler and catalyst cooler, if present. Steam drum boilout procedures can be combined with the dryout as long as the dryout procedure governs. The cooler tubes must have circulating water flow for protection during the dryout.

h.

Close the spent and regenerated catalyst slide valves. Check the blast and sample connections on the catalyst standpipes and drain any free water.

i.

Open the snort valve on the main air blower discharge.

j.

Prepare the direct fired air heater for firing. Prepare alternate internal heaters if they are to be used.

Start the Main Air Blower a.

Follow the manufacturer's instructions for the air blower startup.

b.

Begin closing the discharge snort valve, forcing air into the regenerator. As the pressure of the regenerator rises, the power required to drive the blower increases. Check the air blower's performance curve and keep the machine out of the surge condition.

Refractory Dryout a.

Open the blast and sample connections on the catalyst standpipes and drain any free water.


b.

Follow the vendor's instructions for dryout or the following instructions, if a suitable alternate is not available. Values given are approximate and to be used as guidelines. Good judgment should be used for any situations which deviate from the described steps.

c.

Raise the temperature in the regenerator at a rate of 50°F (30°C) per hour to 250°F (120°C). Adjust the regenerator pressure to control the air blower discharge temperature. Hold at this temperature for a minimum of 12 hours. Open the recirculation catalyst slide valve and cooled catalyst slide valve (if present), to ensure air circulates to all parts of the regenerator.

d.

As the regenerator is heating up, open the regenerated catalyst slide valve to circulate air up the riser into the reactor. Vent at the reactor vapor line vent and reactor shell if required. Raise the reactor temperature at 50°F (30°C) per hour to 250°F (120°C). Open the spent catalyst slide valve to circulate air through the standpipe into the stripper. Air flow should be primarily through the regenerated catalyst slide valve so that the insulating refractory in the riser is heated thoroughly. NOTE:

e.

It is important that the air blower is used only for initial dryout in the reactor. For later startups, only steam should be used to heat up the reactor. During operation, coke will form on the reactor internals and walls. There is a real danger of damaging oxidation or fire if this material is exposed to air at high temperature.

For the cold wall regenerated catalyst standpipe, wye section and riser with 5" high density vibracast refractory lining, the minimum hold period at 250°F (120°C) is 8 hours. It may be that some of this cold wall lining has already been installed and dried out in the shop. In that case, the remaining reactor lining will dictate the hold period required. For the 3/4" abrasion lining in the upper riser and reactor vessel, the minimum hold period at 250°F (120°C) is 4 hours.


f.

After the 250°F (120°C) hold period, light the direct fired air heater and raise the temperature in the regenerator and reactor at 50°F (30°C) per hour to 650°F (350°C) and hold for 12 hours for 4" gunned lining, 8 hours for 5" vibracast lining, or 4 hours for 3/4" abrasion lining. An operator should be stationed by the air heater any time it is in service.

g.

When the regenerator temperature reaches 500°F (260°C), start the purge steam flows to the torch oil guns and nozzles, and any quench nozzles, to keep them cool. The spent catalyst standpipe expansion joint steam purge can also be started at this time.

h.

While the reactor and regenerator are heating up, the equipment should be checked for expansion problems. Completely inspect the vessels, standpipes and structure every hour until the regenerator plenum has reached its maximum, generally ~1200°F (650°C), and every two hours thereafter. Check: (1)

That the equipment is free to expand and is not contacting any structural members.

(2)

That expansion joint tie rods are loose and not binding.

(3)

That catalyst lines and standpipes are free to move.

(4)

That small piping, especially instrument lines and electrical cables, is not under strain.

i.

During the 650°F (350°C) hold period is a convenient time to perform the first hot bolting step. Systematically hot bolt the entire reactor-regenerator section.

j.

All slide valves should be hot stroked at all hold temperatures. Position indicators should be checked with board readings and the valves checked


locally for free movement. Manual and handwheel operation should be checked and any problems corrected.

6.

k.

Raise the regenerator temperature at 100°F (55°C) per hour to 1300°F (700°C) or the highest temperature permitted by the air heater (may be 1200°F (650°C)). Raise the reactor temperature at 100°F (55°C) per hour to 950°F (510°C).

l.

Hold the regenerator temperature at 1250°F (680°C) or the highest achievable for 12 hours for 4" gunned lining or 8 hours for 5" vibracast lining.

m.

When the regenerator temperature reaches 1000°F (540°C), begin hot bolting the entire reactor-regenerator section again. Check for any expansion related problems.

High Temperature Field Test of Reactor-Regenerator At the end of the final refractory dryout hold period, a high temperature test is conducted on the reactor, regenerator and interconnecting piping under simulated low pressure operating conditions. The integrity of field welded and bolted joints under the strains developed by the expansion of lines and vessels is checked before putting catalyst and oil into the system. This test is only conducted during the initial start of the unit. a.

Follow the procedure as described in the UOP Schedule A, Project Specification 314, Reactor-Regenerator Assembly Field Testing Procedure. The test can only be conducted after: (1)

The refractory linings in the regenerator, reactor and associated piping have been fully cured and dried.

(2)

The direct fired air heater lining has been cured and dried.


(3)

7.

Any expansion related problems discovered in the dryout have been resolved.

b.

The reactor should be held at 950°F (510°C) and the regenerator at 1200-1250°F (650-680°C) for a 24 hour period before the system pressure is raised.

c.

Begin closing the reactor vapor line vent valve and flue gas slide valve and raise the reactor and regenerator pressure to approximately 28 psig (2 kg/cm2) or to the value specified in the 314 Specification.

d.

Hold the reactor and regenerator at these conditions for a minimum of 30 minutes. Check all flanges and joints in the system for leaks. Check the entire structure for any expansion related problems.

Cooldown and Inspection a.

After the high temperature field test is completed, reduce the reactor regenerator pressure back to the previous level.

b.

Begin reducing the air heater outlet temperature at 100°F (55°C) per hour. When the regenerator temperature reaches 500°F (260°C), stop the steam purges to the torch oil guns and nozzles.

c.

Shutdown the DFAH when the temperature control becomes erratic due to low fuel gas flow.

d.

Continue air flow through the vessels until the temperature is within 50°F (30°C) of the blower discharge temperature and then shut down the main air blower.

e.

Open all manways and vents. Use fans or other air moving equipment to cool the vessel internals to ambient conditions.


f.

Remove the distributor flange at the bottom of the wye section to allow inspection of the wye section.

g.

Inspect all refractory linings in the catalyst system, including the reactor, regenerator, riser, standpipes, cyclones, air heater, flue gas line and orifice chamber. Lock each slide valve in position to prevent accidental movement during inspection. Small hairline cracks will be present in the refractory and do not present a problem. Large cracks greater than 3/8" (10 mm) that extend all the way to the shell should be repaired.

h.

For the final flange assembly before startup, glue ¼” of ceramic blanket to the refractory retaining collars on all cold wall manways or blind flanges before they are closed (see UOP standard specification 3-24-2 Figure 4).

B. NORMAL STARTUP The normal startup of the unit can be divided into the following steps: 1.

Steam out the reactor and main column, and heat up the catalyst section.

2.

Heat up the fractionation section.

3.

Load catalyst to the regenerator and heat catalyst.

4.

Start the wet gas compressor.

5.

Circulate catalyst between the reactor and regenerator.

6.

Start oil to the riser.

7.

Establish normal operating conditions.


It is assumed that the unit is cold and empty, and that all precommissioning activities and refractory dryout have been completed. General guidelines are given in these procedures and each refiner should develop specific startup procedures for their particular unit. Variations from the steps given here are acceptable as long as the basic intent is followed and safety matters are not compromised. 1.

Steam Out Reactor and Main Column a.

The following procedure starts from the condition that all equipment and vessels are full of air. Steaming of the reactor and fractionation section is used to free the unit of oxygen. The following items should be completed before the steam out is begun: (1)

The unit must be completely flushed, all vessels and piping closed up, all orifice plates and instrumentation installed, and all equipment ready for startup.

(2)

The reactor vapor line blind should be removed at this time (refer to Figure 4). Some refiners prefer to leave this blind in place until later in the startup sequence. This is acceptable but it is generally more convenient to remove the blind now. The procedure following is designed to use the reactor containing steam as a protection buffer between the regenerator with air and the main column with hydrocarbon.

(3)

If the vapor line blind is out, the vapor line vent should be blocked and blinded.

(4)

Remove the blankoff from the main column high point vent and open the valve a couple turns.

(5)

Isolate the wet gas compressor at the suction and discharge lines. Steam should not be allowed to enter the compressor.


(6)

Line up backup nitrogen to the DG purge gas header and start purge gas to the reactor instrumentation DG points.

(7)

Make sure the regenerated and spent catalyst slide valves are fully shut. All slide valves should be operational with hydraulic systems commissioned.

(8)

Low points from various equipment and piping should be prepared to drain condensate as the steam out progresses.

b.

Start steam to the base of the riser, to the feed distributors, to the reactor stripper, and to the spent catalyst standpipe blast point. As the reactor is purged and heated, steam will flow through the vapor line into the main column. Continually drain condensate at the riser, reactor, stripper, and main column low point drains. The spent catalyst standpipe blast point will have to be used as a drain after the initial steaming is completed.

c.

Start steam into the bottom of the main column and into the sidecut strippers. Vent at the top of the main column and at the overhead receiver. Do not run the condenser fans or water to the trim cooler during this procedure.

d.

Steam through the overhead receiver to the wet gas compressor suction drum. Vent at the drum and drain condensate from low points. Steam through the spillback lines and the interstage receiver but make sure the compressor remains isolated. Connect steam hoses as needed if there is trouble achieving a good steam plume from vents.

e.

Continually drain condensate from all low points. As the vessels and piping heat up, less steam will condense and the low points can be throttled to match the condensate drain rate. Condensate collecting in the overhead receiver water boot can be pumped out with the sour water pumps.


f.

When there is a good steam plume out the top of the main column, main column receiver, compressor suction drum and interstage receiver for at least two hours, the system should be free of air and the steamout is complete.

g.

A quick leak check should be conducted on the system. Throttle all vent and drain valves to raise the pressure in the system to 10-12 psig (0.7-0.8 kg/cm2). Check all flanges, valve packings, etc. for leaks. It may be that too much steam is condensed in the main column condenser to permit building pressure for this leak test.

h.

Check all low points to ensure that all water is drained. Verify that the spent catalyst standpipe is being drained from the blast and sample connection.

i.

Start injecting fuel gas at the LCO stripper vapor return line. Raise the reactor and main column pressure to 10-12 psig (0.7-0.8kg/cm2) before reducing the steam injection. This will ensure that air is not drawn into the unit as the steam condenses.

j.

Close all vents and continue draining water from all low points. Never leave a drain point unattended with fuel gas in the system. NOTE: Fuel gas injection to the LCO stripper should be enough that the pressure transmitter on the main column receiver will keep the overpressure control valve to the flare open a small amount at all times. This will ensure that any air that might enter the system will be purged out to the flare. On older units which used air as the purge gas to the reactor instrument taps, air from the DA points in the reactor is carried by the steam flows to the main column and receiver. If this air is not purged out, the concentration of oxygen can build up in the receiver to a high level.


2.

k.

The main column overhead condenser can be put in service to condense the steam and build a concentration of fuel gas in the overhead receiver. Fuel gas pressure must be adequate to ensure that a vacuum cannot develop in any of the vessels. This is done in preparation for starting the wet gas compressor.

l.

The Gas Concentration Unit should be air-freed by steamout in parallel with the reactor-regenerator-main column sections. The columns and absorbers should be pressured with fuel gas at the conclusion of the steamout. Drain any free water from all low points.

Start the Main Air Blower a.

Be sure all DA and DG purges and all instrumentation in the reactorregenerator section are in service.

b.

Start the main air blower after the fractionation section is pressured up with fuel gas. Refer to the manufacturer's instruction manual for the blower startup procedure.

c.

The differential pressure between the reactor and regenerator should be maintained at a negative (reactor higher) 1.5 psig (0.1 kg/cm2). This will ensure that any leakage through the slide valves will put steam into the regenerator rather than air into the reactor. NOTE: The steam in the reactor acts as a buffer between the regenerator containing air and the main column which contains some fuel gas. As long as the reactor pressure is maintained as the highest level in the system, no contamination of air and fuel gas can occur.


3.

4.

d.

Open the recirculating catalyst slide valve and catalyst cooler slide valve (if present) to allow heated air to flow up the standpipes to the upper regenerator.

e.

If the unit is equipped with a power recovery unit, follow the manufacturer's instructions for starting flue gas to the expander turbine

f.

Begin water circulation through the catalyst cooler tubes (if present) and flue gas cooler tubes.

Light the Direct Fired Air Heater a.

When the flow from the main air blower has stabilized, light the direct fired air heater according to the manufacturer's instructions.

b.

Heat up the regenerator at a maximum rate of 200°F (110°C) per hour to a target temperature of 1000°F (540°C).

c.

Start fluidizing air to the upper regenerator fluffing rings and to the catalyst cooler fluidizing lances (if present) at minimum flow.

Inventory the Fractionation Section with Oil a.

As the regenerator is being heated, the fractionation section can be inventoried with oil to begin circulation and pumparound flows. During this process the main column overhead should be maintained above 230°F (110°C) to minimize steam condensation in the column. Fuel gas flow should be maintained to the LCO stripper vapor return line so that a small purge flow is maintained through the overhead receiver overpressure line to flare.

b.

Since LCO and HCO will be unavailable for flushing oil to the instruments and the main column bottoms pump flushes (gland seal, wear rings and throat bushing), the flush header should be commissioned with raw oil or


a light gas oil from outside the battery limits if the raw oil is very heavy. When the LCO and HCO products become available, the flush systems will be changed back to the proper stream. c.

Start raw oil feed through the startup filling line to the bottom of the main column. The addition of cold feed into the column will condense some steam since it passes countercurrent with steam over the disk and donut trays. Therefore, add raw oil to the column slowly until the column is hot. Pay special care to removing free water from the low points in all vessels, exchangers and piping because sudden water vaporization can damage equipment.

d.

Slowly warm up the main column bottoms steam generators by backing in steam through the bypass around the non-return valves. These steam generators will be used to heat the raw oil in the main column.

e.

Start the main column bottoms pump and send oil through one of the steam generators. This circulation must be done slowly at first to avoid cooling the bottom of the main column. Raise the raw oil outlet temperature as high as possible, then start flow through the second steam generator. Try to maintain the column overhead temperature high enough to drive water overhead.

f.

When the column bottom temperature has stabilized at its highest level from the steam heating, slowly start bottoms circulation through the other bottoms circulation loops. Circulate through the cold sections of the system slowly until they are heated to avoid cooling the column bottom excessively. Watch the bottoms level and bring in more raw oil as the total bottoms circuit is inventoried. Start flow from the net bottoms pumps through the startup line back to the raw oil charge pumps to complete the loop. Check for water at the low points throughout the bottoms system. This circulation and heatup will slowly dry out the bottoms system.


g.

After the bottoms circuit is hot and dry with good circulation flows, check for free water in the LCO and HCO system low points. Check that the sidecut strippers inlet level control valves are fully closed. Start a small flow of hot raw oil from the main column bottoms up to the LCO circuit return line via the startup filling line. This will put hot oil into the upper part of the column. When the LCO draw tray is full, it will overflow to inventory to HCO section and then overflow back to the bottom of the column. Add raw oil as necessary to maintain the bottoms level as the LCO and HCO sections are inventoried. Keep the bottoms temperature as hot as possible with the steam generators.

h.

Drain all free water in the LCO and HCO circuits and slowly start oil flow through the heat exchange circuits in the Gas Concentration Unit, bypassing the exchangers. Check that the main column overhead temperature is high enough to avoid steam condensation in the column. Do not place the sidecut strippers into operation at this time.

i.

Circulate oil flow through the LCO and HCO circuits until the lines are hot and all free water has been removed. Maintain a small flow from the hot bottoms circuit up to the LCO section to keep these sections hot and overflow the draw trays back to the bottoms.

j.

Inventory the main column receiver with startup naphtha. Don't start reflux at this time as it will cool off the top of the column and cause additional steam condensation.

k.

This method of starting oil circulation tries to minimize water accumulation in the fractionation section. It is very important to avoid circulating free water back to hot sections of the column as the rapid water vaporization could cause damage to the trays. All changes in operation and flows should be made slowly and only after first draining any free water from low points. Spare pumps should be drained and switched occasionally to prevent water accumulation.


5.

Prepare the Regenerator for Catalyst Inventory a.

The heating of the main column and the regenerator can be taking place at the same time. The regenerator is being heated with the direct fired air heater and steam is flowing through the reactor. Completely inspect the reactor-regenerator structure every hour until the maximum air heater outlet has been reached, around 1200°F (650°C), and once every two hours thereafter. Check: (1)

That the equipment is free to expand and is not contacting any structural members.

(2)

That expansion joint tie rods are loose and not binding.

(3)

That catalyst lines and standpipes are free to move.

(4)

That small piping, especially instrument lines and electrical cables, is not under strain.

b.

When the regenerator temperatures reach 500°F (260°C), start purge steam to the torch oil guns and nozzles, and any quench nozzles in the upper regenerator. Commission the flue gas steam generator before the temperature exceeds 500°F (260°C). Change air purges to packing glands and expansion joints over to steam at 500°F (260°C).

c.

As soon as the regenerator temperatures are above 450°F (230°C), the differential pressure transmitter across the spent catalyst slide valve can be placed in service. This transmitter should read the same as the reactor-regenerator differential pressure transmitter as it is measuring the same two pressures. If it does not, this is an indication that there may be condensate in the spent catalyst standpipe. The water can either be drained from the blast and sample point, or preferably, it can be slowly drained into the regenerator where it will vaporize and exit with the flue gas. Manually open the spent slide valve a small amount to continually


drain the standpipe so that water does not accumulate. The desired situation is that all of the condensate is kept drained, and if drained into the regenerator, it is done slowly so that when the water vaporizes, it will not cause a large pressure surge in the regenerator which could push air into the reactor. d.

When the air heater temperature reaches 600°F (315°C), start hot bolting flanges and manways and systematically hot bolt the entire regenerator and flue gas system. Repeat this procedure when the air heater temperature exceeds 1000°F (540°C).

e.

When the temperature in regenerator has reached 1000°F (540°C), catalyst can be loaded. A certain temperature is not required to load catalyst, but since the catalyst is cold and must be heated, it is best to have the regenerator already hot when loading is started. The main air blower should be set at the design rate or at the maximum firing limit of the air heater and the air heater should be adjusted to the maximum outlet temperature, typcially around 1200 – 1350 ºF (650 - 730°C). Since heating the catalyst inventory is very time consuming, a high rate of hot air flow is needed to help minimize this period. Close the recirculating catalyst slide valve and the catalyst cooler slide valve (if present) before loading catalyst.

f.

Check that all instrument purges have been started and contain the proper RO throughout the reactor-regenerator section. Check that all regenerator level and density transmitters are functioning. Ensure that air flow is going to the upper regenerator fluffing rings and catalyst cooler air lances (if present).

g.

Note that as catalyst is loaded into a high efficiency, combustor style regenerator and the combustor density increases, the pressure in the bottom of the combustor will increase. This will reduce the spent catalyst slide valve P and it may be possible for air to flow back into the reactor. If this P approaches zero reduce the regenerator pressure (with the


reactor-regenerator PDIC) to ensure the spent catalyst slide valve P remains positive. 6.

Loading Catalyst to the Regenerator The initial load of catalyst into the regenerator should ideally be enough to provide the total unit inventory. However, due to the size of the reactor and stripper, this is usually not the case. The regenerator should be loaded with as much catalyst as possible to high levels, then after catalyst circulation is started and the reactor inventoried, additional catalyst will need to be loaded into the regenerator.

The cyclone inlet velocities in the regenerator should be maintained greater than 35 ft/sec (11 m/sec) as much as possible to ensure good catalyst separation efficiency. Lower pressure during heatup and catalyst loading can help increase the velocity when the regenerator is cool. For bubbling bed and RFCC regenerators the superficial bed velocity should not exceed 3 ft/sec (0.9 m/sec) during startup to minimize catalyst loading to the cyclones. a.

Ensure that the following items have been accomplished before loading catalyst into the regenerator: (1)

All slide valve P transmitters should be in service and the low P override controller operable.

(2)

Check that all instrument purges, slide valve packing purges, and expansion joint purges are commissioned.

(3)

Check that steam is not being used in the regenerated catalyst standpipe, as any condensate will make mud when the catalyst is loaded. Check the blast and sample point to make sure no condensate is present.


(4)

Make sure the catalyst cooler, regenerated catalyst and spent catalyst slide valves are fully closed.

(5)

Ensure that fluffing air is on to the upper regenerator rings and to the catalyst cooler (if present). Water must be circulating through the cooler tubes to keep them cool.

(6)

Raise the feed atomizing steam to 150% of design rates to ensure that any catalyst passing through to the riser can not plug the distributors.

b.

Prepare the catalyst hoppers for transferring catalyst. Refer to the specific procedures outlined in the Catalyst Handling section. Gauge the hoppers before starting to establish the initial catalyst inventory.

c.

Start catalyst loading to the upper regenerator. Observe the level instruments for signs that catalyst is accumulating. For a high efficiency regenerator, open the recirculation slide valve a small amount to begin circulation of catalyst to the combustor when a level is established and a differential pressure appears across the recirculation slide valve. The regenerator will begin to cool as cold catalyst is added. Keep the air heater firing at its maximum temperature to help heat the catalyst.

d.

For a high efficiency style regenerator, open the recirculation slide valve further to increase the density in the combustor as the level increases in the upper regenerator. The combustor density will remain low even with the recirculation valve full open since there is no catalyst circulating through the spent standpipe yet. If possible, try to establish a combustor density of 4-7 lb/ft3 (65-110 kg/m3).

e.

Near the end of the catalyst loading step, when a high level exists in the regenerator and the combustor density is as high as possible with the recirculation slide valve open, it is advisable to reduce the main air blower rate to around 50% of design. This will cause the combustor density to


increase, thereby increasing the combustor catalyst inventory. The upper regenerator level will drop, which allows additional catalyst to be loaded, increasing the total catalyst inventory. Adjust the air heater firing when the air rate is reduced so the maximum temperature is not exceeded. f.

7.

At the completion of the catalyst loading step, gauge the hopper again to determine the quantity of catalyst loaded.

Heat Up the Catalyst Inventory a.

As the cold catalyst is being loaded, it will cool the regenerator. Keep the air heater firing at its maximum temperature to heat the catalyst. The rate of heating the catalyst is not critical; the size of the catalyst inventory, the speed of catalyst loading and the duty of the air heater will affect how fast the temperature can be raised. A rate of 200-300°F (110-170°C) per hour is a good target.

b.

For a high efficiency regenerator, adjust the catalyst recirculation rate to obtain a density of at least 4-7 lb/ft3 (65-110 kg/m3) in the combustor during the heatup. Torch oil should not be fired if the combustor density is less than 4 lb/ft3 (65 kg/m3). Excessive particle temperatures or afterburning can result if sufficient catalyst is not available to absorb the heat from torch oil firing. For a bubbling bed regenerator or 2 stage, RFCC regenerator the catalyst level should be a minimum of 1 ft (0.3 m) over the torch oil guns before starting torch oil.

c.

If torch oil is to be used to heat up the catalyst, the minimum temperature at which it should be fired is 800-850°F (425-450°C). If an increase in the combustor temperature is observed, the torch oil has ignited. If no temperature change is observed, the torch oil has not ignited, and its use should be discontinued until the catalyst temperature has been raised a further 50°F (30°C). When using torch oil, adjust the torch oil atomizing


steam pressure to a minimum of 50 psig (3.5 kg/cm2). Control the amount of torch oil to give a smooth even rise in temperature.

8.

d.

If a catalyst cooler is present on the regenerator, heating the catalyst will take additional time, as the cooler will be taking some heat out as the entire system is being heated. This is because water is circulating through the tubes and the fluffing air is on. Water circulation is required to protect the tubes and some air flow is recommended to keep the air lances clear of catalyst. The minimum amount of fluffing air should be used throughout the startup to minimize this heat removal effect.

e.

When the regenerator circulating catalyst inventory has been heated to 900°F (480°C), catalyst circulation to the reactor can be started. Continue heating the regenerator catalyst to a target value around 1250°F (675°C) in preparation for cutting in feed.

Start the Gas Concentration Unit Wet Gas Compressor In order to control catalyst circulation between the reactor and regenerator, it is necessary to have a constant pressure on the main column. Starting the wet gas compressor at this time allows a better control over the pressure and removes this task from the very busy time when feed is cut in. However, it may not be possible to start the compressor now if fuel gas supply is insufficient or molecular weight is too low. In that case, pressure control is maintained as before, with a fuel gas purge to the LCO stripper and venting to flare at the overpressure control valve. The wet gas compressor can be started after feed is started to the riser. a.

It is advisable to start the wet gas compressor early and have it operating smoothly before circulating catalyst. Set the process controls in preparation for compressor startup as follows: (1)

Set the main column overpressure control to hold the system pressure at 10 psig (0.7 kg/cm2).


9.

(2)

Manually open both compressor spillback control valves and their bypasses.

(3)

Set the compressor speed control on manual at minimum speed or open the suction valve fully.

b.

Start the compressor interstage cooler fans and open the cooling water to the trim cooler.

c.

Increase the fuel gas makeup to the LCO stripper to provide an operating cushion before starting the compressor.

d.

Start the compressor auxiliaries and the compressor according to the manufacturer's instructions.

e.

Keep the compressor operating on total spillback until feed is charged to the reactor. The discharge valve can be cracked open slowly to help pressure up the gas concentration unit at this time but be careful to do this very slowly so that the main column pressure is not sucked down quickly.

Start Catalyst Circulation The following procedure is general in nature and the specific arrangement of the reactor and regenerator system needs to be carefully considered. The velocity in the lift zone, upper riser and cyclones needs to be considered at all times to ensure smooth catalyst circulation and to minimize catalyst losses. For all types of riser terminations the velocity throughout the riser should always be greater than 10 ft/sec (3 m/sec) when circulating catalyst. 15 ft/sec (4.5 m/sec) is preferred. This will ensure smooth catalyst flow. When heating up or circulating with steam only the heat demand and therefore the catalyst circulation and cyclone loading is very low so that the cyclone efficiency is not critical.


Once startup naphtha (if used) or feed is introduced to the unit the additional heat required to vaporize the hydrocarbon increases the catalyst circulation and therefore the cyclone loading quickly. Before starting these streams the catalyst separation efficiency of the riser termination device and cyclones must be increased to minimize losses to the main column. For a direct connect, SCSS, VSS or VDS riser termination systems the cyclone inlet velocity should be increased to 35 ft/sec (11 m/sec) or greater with the startup steam to the wye before starting raw oil or startup naphtha to the riser. Note that this velocity includes the stripping steam vapor flow. This is very important for direct connect or SCSS systems. The vortex chamber on VSS and VDS riser termination systems is less sensitive to changes in velocity than other types of termination devices so that this is not as critical but it is still recommended. For vented risers the velocity out of the riser is critical for catalyst separation. Catalyst should not be circulated with startup naphtha or feed with less than 35 ft/sec (11 m/sec) riser exit velocity. This velocity does not include the stripping steam vapor flow. a.

When the regenerator has reached 900°F (480°C), the unit is ready to start catalyst circulation. Temporarily stop the flow of oil from the MCB circulation up to the LCO and HCO sections of the main column. Shutdown the HCO and LCO circulation pumps if the inventory in these sections is lost. When catalyst circulation is first started, it is possible that catalyst can be carried into the main column. It is best to contain these fines in the bottoms rather than having them spread through the LCO and HCO circuits.

b.

Check the spent catalyst standpipe for any condensate. Crack open the spent slide valve occasionally to drain condensate into the regenerator. The P controller for the reactor and regenerator should be set at a negative (reactor higher) 1.5 psig (0.1 kg/cm2) or more if required to keep the spent catalyst slide valve P positive to provide a steam buffer


(reactor) between the air in the regenerator and hydrocarbon in the main column. c.

Set the lift and/or startup steam to the wye to maintain a velocity of 15 ft/sec (4.5 m/sec) throughout the riser. If sufficient fuel or natural gas is available, lift gas flow can be started to help reduce the steam requirement. This flow can come from the normal lift gas source (sponge absorber) as recycle if the wet gas compressor is operating, or can be piped in externally and vented at the main column overhead receiver if the compressor is not operating.

d.

Feed steam to the Optimix feed distributors should be set at 150% of design to ensure that catalyst can not plug the nozzles.

e.

Adjust the stripping steam and fluffing steam flows to design rates.

f.

Check the P across the regenerated catalyst slide valve and blast the standpipe with air if there is low or no P. Slide valve differential pressures will be erratic at low catalyst circulation rates.

g.

Start opening the regenerated catalyst slide valve with the reactor temperature controller in manual. Closely watch the reactor temperature which will rise as soon as catalyst begins to circulate. If no response is observed after several minutes, blast the regenerated catalyst standpipe again.

h.

The catalyst circulation from the reactor back to the regenerator should be started as soon as possible to minimize any potential mud formation (catalyst + condensate) in the spent catalyst standpipe. Do not wait until the reactor level is fully established to start catalyst flow to the regenerator. As soon as catalyst creates an increasing P across the spent slide valve, open the valve on manual to start returning catalyst to the regenerator. If the P does not increase across the valve, blast the standpipe with steam.


i.

The reactor level should be increased without delay, particularly if the cyclone diplegs are to be submerged in the catalyst bed. It is possible that some catalyst can be lost to the main column before the diplegs are submerged, but if the stripper level is raised smoothly and quickly, catalyst losses will be minimized. It is important that a good flow of catalyst is leaving the stripper back to the regenerator during this period. On units without submerged diplegs the level may be increased more slowly. j.

As the reactor stripper level is increased, it may be necessary to bring in more catalyst from storage to maintain levels in the regenerator. On units with submerged diplegs this should be done before the stripper level approaches the bottom of the diplegs. The catalyst level should never be held just below the diplegs for any reason as it is possible to create a vacuuming action through the cyclones if the seal is lost and draw catalyst up the diplegs and out to the main column. Once the catalyst level approaches the diplegs, they should be submerged as quickly as possible.

k.

Place the reactor level controller on automatic as soon as possible. The reactor temperature controller should be maintained in manual.

10. Raise Reactor Temperature a.

As catalyst circulation is started, raise the reactor temperature smoothly at a rate of 200-300°F (110-170°C) per hour. For certain reactor configurations, it may be important how fast the skin temperatures are increased. In some cases, special guidelines will be specified. It is useful to record the reactor skin temperatures during the heatup for future reference and analysis of developed stresses during the thermal expansion.


b.

When the reactor temperature has reached 600°F (315°C), startup naphtha flow can be started to the riser. Startup naphtha may be used to increase the catalyst flow and open the slide valves further for better control. It smoothes the transition when feed is cut into the riser. It also aids in wetting the main column trays with hydrocarbon and helps displace water in the main column. Startup naphtha is an optional step in the procedure. Any type of light naphtha, straight run or cracked, can be used. Increase the startup steam to achieve a cyclone inlet velocity of 35 ft/sec (riser outlet velocity for vented riser terminations) before starting naphtha flow to ensure that the separation efficiency is good as the catalyst circulation rate and cyclone loading will increase significantly with the heat required to vaporize the naphtha. As the naphtha is vaporized and the cyclone (or riser outlet) velocity increases the startup steam may be reduced. Add additional naphtha to the main column overhead receiver as needed. This naphtha will be recycled through the main column, to the overhead receiver, back to the riser. Keep raising the reactor temperature at the specified rate when naphtha is added.

c.

It is important that the bottom of the main column be maintained as hot as possible during the reactor heatup to prepare for eventual cracked product flows. The steam generators should be used to heat the bottoms and maintain at least 350°F (175°C) in the bottom of the column. A high temperature will ensure that all water and much of the naphtha is driven overhead to minimize the accumulation in the column.

d.

Continue to heat up the regenerator catalyst inventory to 1250°F (675°C) as the reactor temperature is increased. Torch oil can be used to maintain this temperature while the reactor is being prepared for starting feed. Make sure the combustor density is maintained above 4 lb/ft3 (65 kg/m3)


when torch oil is fired (or the level is maintained above the torch oil guns in a bubbling bed or RFCC regenerator). e.

At some point after stable catalyst circulation has been established, and both spent and regenerated catalyst slide valves have good P across them, the reactor-regenerator P controller can be adjusted to a positive value (regenerator pressure higher) to help balance the two slide valve P's. Make sure the slide valve low P override controllers are commissioned.

f.

Increase the reactor temperature to a target point around 980°F (525°C) and the regenerator temperature to ~1250 – 1300 ºF (675 - 705°C). Maintain stable catalyst flow and levels before starting feed to the riser.

11. Charge Oil to the Reactor Riser Feed can be started to the riser as soon as the preceding operations have been stabilized. a.

Stop backing steam into the main column steam generators and fill them with boiler feed water to prepare for eventual heat removal/steam production. This should be done slowly to avoid excessive thermal shock when the tubes are changed from hot steam to cooler BFW.

b.

Set the feed flow through the bypass valve to the main column at approximately 10% of design charge rate. The flow may be lined up through the main column bottoms recycle flow meter. The bottoms recycle meter is convenient to use for starting feed to the riser because the normal feed control valve is too large to accurately control such small flows.

c.

Feed steam should be set at 150% of design. Stripping steam and fluffing steam flows should be at design.


If startup naphtha is in service the cyclone inlet velocity (riser outlet velocity for vented risers) should already be greater than 35 ft/sec (11 m/sec). If startup naphtha is not used then the startup steam rate should be increased to achieve a cyclone inlet velocity (riser outlet velocity for vented risers) of 35 ft/sec (11 m/sec) or greater. d.

Prior to starting feed to the riser, be sure to drain all free water from the feed line between the diverter valve and the feed nozzle block valves. Start feed to the riser by switching the feed bypass switch to the normal position. This will close the valve on the line to the main column and open the line to the riser. Begin opening the regenerated catalyst slide valve further at the same time to provide the additional heat required to maintain riser temperature and velocity. NOTE: Start feed very slowly at first to avoid thermal shock to the feed distributor tips. The feed distributor tips can be cracked if subjected to excessive thermal shock. When cutting in oil, catalyst circulation must be increased to maintain the reactor temperature. Do not allow the temperature to drop below 930°F (500°C). Should the reactor temperature drop too low, feed can be reduced until the temperature is increased again. Initially, the regenerator temperature may drop until catalyst containing coke enters the regenerator. Try to maintain the combustor temperature around 1250°F (675°C).

e.

Begin increasing the feed rate smoothly in increments of 10-20% through the normal feed control valve. As the feed rate is increased, startup naphtha can be smoothly backed out. The lift steam rate can also be reduced. It is best to wait until the feed rate is up to 40-50% before beginning to reduce the startup naphtha and startup and feed steam


rates. The reduction should be done gradually so that riser velocity does not suddenly drop. f.

As the circulation of relatively cool catalyst from the reactor to the regenerator increases the regenerator temperatures will decrease temporarily until spent catalyst with coke has displaced the clean catalyst in the stripper. Additional torch oil will be required to keep the regenerator temperature at 1250 –1300 ºF (675-705 ºC) When spent catalyst starts entering the regenerator and the coke starts burning the regenerator temperatures will increase. Torch oil flow can be reduced and eventually stopped. The air heater firing can also be reduced and eventually stopped. The combustor temperature should be maintained around 1275-1300°F (690-705°C). The regenerator upper dense bed temperature should be slightly higher than the combustor.

g.

Continue increasing the feed rate smoothly in 5-10% increments to the design value. During this time, gradually increase the air rate to the regenerator as necessary. Adjust the reactor-regenerator P controller as needed to balance the spent and regenerated catalyst slide valve P's.


C. ESTABLISH NORMAL OPERATING CONDITIONS At the completion of the unit startup, stabilize the unit operation using the following process control guidelines: Reactor Variables

Control

Raw Oil Charge Rate

As desired.

Raw Oil Preheat Temperature

Set to balance coke yield, conversion, and gasoline RON requirements.

Lift Steam

Total flow set to achieve optimum lift zone velocity, typically 10-20 ft/sec (3-6 m/sec). Flows may be used in any ratio depending on wet gas compressor, main column overhead or sour water stripping constraints. Lift gas is beneficial for metals passivation in units with high nickel on Ecat (>3000 wppm)

And Lift Gas

Feed Steam

Typically 1-2 wt% of design feed rate. Should be adjusted to optimize yields. May be used at high flow rates during startup and emergencies.

Reactor Temperature

Adjust to obtain desired conversion, yield pattern, coke yield and gasoline RON.

Reactor Pressure

Indirectly set pressure.

Reactor Catalyst Level

Set to cover the top stripping grid or to seal the diplegs in units with submerged primary cyclone diplegs.

by

main

column

receiver


Reactor Variables

Control

Stripping Steam

Use just enough to strip the catalyst of residual hydrocarbons. Typical rate is 1.7-2.5 lb (kg) per 1000 lb (kg) of catalyst circulation. Adjust by observing effect of changes on regenerator temperature.

Main Column Bottoms Recycle

Normally zero. During turndown or when light feeds are processed, some recycle may be necessary to increase coke and help the unit heat balance.

Naphtha to Riser

Used to assist catalyst circulation during startup and to help control the regenerator temperature when the unit is behind in burning (old style unit – partial combustion operation).

HCO Recycle

For units operating in maximum distillate production, used to increase coke yield or improve LCO yield.


Regenerator Variables

Control

Combustion Air Rate

Adjust for sufficient air to burn all coke off spent catalyst. Maintain 1-2% excess oxygen in the flue gas for full combustion units. Typical value for full combustion is 14 lb (kg) air per lb (kg) coke. The air/coke ratio on partial combustion units is lower and is adjusted to control the heat of combustion.

Combustor Temperature

Adjust for proper coke and CO combustion and minimize afterburning. Typical value 1275°F (690°C) minimum.

Combustor Density

Adjust to optimize coke and CO combustion. Normal value between 5-10 lb/ft3 (80-160 kg/m3).

Regenerated Catalyst Temperature

Function of coke operation. May be influenced by catalyst cooler if present. Dense bed normally 20-50°F (10-30°C) above combustor temperature. Dilute phase normally 10-20°F (5-10°C) above dense.

Reactor-Regenerator Differential Pressure

Adjust to obtain stable and balanced spent and regenerated catalyst slide valve P's.

Slide Valve P's

Dependent on reactor-regenerator pressure and catalyst levels. Normal values between 512 psi (Low pressure over-ride typically set about 2 psi (0.14 kg/cm2).


Regenerator Variables

Control

Upper Regenerator Catalyst Level

Adjusted by additions and withdrawals to maintain a suitable catalyst surge capacity for the unit.

Torch Oil Rate

Used during startup to aid in catalyst inventory heatup. The use of torch oil should be minimized for the protection of the catalyst.

Air to Catalyst Cooler

Used to control the catalyst cooler duty and therefore the regenerated catalyst temperature. A minimum air rate of 10-20% of design should be maintained at all times. The maximum air rate specified for the cooler should never be exceeded. On flow through catalyst coolers the air rate can be adjusted to keep the cooled catalyst slide valve in a good operating range.

Catalyst Cooler Slide Valve

A secondary control for adjusting heat removal. Can be used to limit T across cooler to about 200°F (100°C).


Fractionation Section Variables

Control

Main Column Receiver Press

Adjust as required by the reactor-regenerator P, and the relative main air blower discharge pressure and wet gas compressor suction pressure.

Overhead Receiver Temp

Generally maintained around 110-120°F (4050°C).

Main Column Top Temperature

Set to control the endpoint of the unstabilized gasoline.

Main Column Reflux Rate

Controlled on cascade by the overhead temperature controller. Reflux rate is set by the overhead condenser duty required to heat balance the column after the duty of the lower pumparound streams are set. Primary adjustment is with the MCB steam generators. Set to control the draw temperature and endpoint of the heavy gasoline product.

Heavy Naphtha Product Draw Rate LCO Product Draw Rate

Set to control the LCO draw and LCO product endpoint or to control the MCB temperature.

Cycle Oil Circulation Rates

Set by the process requirements of the associated heat exchangers.


Fractionation Section Variables

Control

Main Column Bottoms Circulation Adjust flow to steam generators to balance Rate the main column heat removal and set overhead reflux flow. Minimum total flow back to column must satisfy disc and donut liquid rate of 6 gpm per ft2 (15m3 per m2) of column area.. Main Column Bottoms Temp

Controlled primarily by the LCO product draw rate. Quench from the steam generators may be used to subcool the liquid in the bottom of the column. Maximum bottoms temperature is generally ~680ºF (360ºC) but is dependant on feed type and reactor severity.

Unstabilized Gasoline Yield

Depends on charge rate and conversion. Controlled by the level in the overhead receiver.

Main Column Bottoms Product Rate

Adjust to control the main column bottoms level.

Cycle Oil Stripping Steam Rate

Adjust for product flash point specification.

Net Overhead Gas Flow

Depends on charge rate, reactor severity and catalyst condition. Controlled by the wet gas compressor speed or spillbacks to control the main column overhead receiver pressure.

Flush Oil

Adjust as required to keep catalyst out of instruments, and flush the main column bottoms pump packing gland and wear rings.


D. NORMAL SHUTDOWN The shutdown of the FCC Unit is essentially the reverse of the startup steps. It should be carried out in an orderly and planned sequence. Some main points to remember are: 1.

Maintain good catalyst fluidization and circulation through the reactor and regenerator throughout the shutdown. Increase riser steam to ensure smooth catalyst circulation.

2.

Always decrease the charge rate before decreasing the air rate. Maintain excess oxygen in the flue gas at all times and keep the regenerator hot to ensure the catalyst is fully regenerated.

3.

Make sure all pumparound circuits are flushed out to eliminate problems with heavy oils or catalyst fines.

During scheduled shutdowns, the catalyst section and the main column exchanger circuits will usually be inspected and cleaned. Depending on the work to be done, the main column might have to be water washed for entry. The Gas Concentration Unit will be pumped out and purged with steam. If columns are to be entered, they will need to be water washed. Precautions must be taken to cool the reactor sufficiently before allowing air to enter the vessel. This is done to guard against the possibility of auto-ignition of hot coke deposits in the reactor or vapor line. The reactor should be cooled below 350°F (175°C) before any manways or nozzles are opened. Riser and stripping steam should be used to assist cooling the vessel as required. The following procedure describes a full shutdown for maintenance entry to vessels. Depending on the reason for the shutdown, the full procedure may not be followed, and catalyst may or may not be left in the regenerator. In most cases, catalyst will be transferred from the reactor to the regenerator during a shutdown. These


procedures should be considered only guidelines. Detailed shutdown instructions should be prepared by the refiner for his specific unit. If the shutdown is temporary, the unit may be maintained in an operating mode and catalyst circulation continued. However, circulating hot catalyst on steam for long periods of time will damage the catalyst. Therefore, if the shutdown will be for several days, it is usually best to stop catalyst circulation. Procedure 1.

Notify offsites and utility systems well in advance that the FCC Unit will be shutting down. Prepare the regenerator by withdrawing catalyst to drop the upper regenerator level to a low value. This will make room for the eventual transfer of the catalyst in the reactor and reduce the time needed to unload the unit catalyst inventory.

2.

Slowly begin reducing the reactor temperature to 900°F (480°C). At the same time, begin reducing the naphtha and LCO product flows. This will make the main column bottoms material lighter, aiding in flushing out the bottoms circuits.

3.

The regenerator temperatures will begin to drop when the reactor changes are made. Adjust the recirculation catalyst and catalyst cooler to keep the combustor hot, around 1250°F (675°C), to ensure all coke is burned off the catalyst.

4.

Maintain the main column bottoms level low around 30%. Control the level by drawing more or less bottoms product to storage.

5.

Begin decreasing the raw oil charge rate in increments of 5 to 10% to 50% of design. For full combustion units decrease the combustion air rate as the charge is reduced, but always maintain excess oxygen in the flue gas (2-5% provides a good safety margin) and good cyclone velocities. On partial combustion units reduce the air rate to control the heat of


combustion and move towards full combustion as the regenerator temperature drops. Control the regenerator temperature at 1250-1300ºF (675-705ºC). Maintain balanced main column heat removal and product draw temperatures by decreasing pumparound circulations and product flows as required. 6.

As the raw oil charge rate is reduced, increase the lift steam and the feed steam to assist catalyst circulation. Begin reducing the reactor pressure to increase vapor velocity in the riser and the cyclones. Lift gas flow to the riser may need to be reduced as the gas production is decreased. When control of the lift gas becomes difficult or unstable, shutdown the flow and block in the control valve.

7.

As the coke make decreases, the regenerator will cool off. Reduce the catalyst cooler air rate and begin closing the cooled catalyst slide valve. The slide valve may be closed completely but do not stop the fluidizing air until catalyst is removed from the regenerator. Fire the air heater when needed to hold the combustor temperature at 1225-1250°F (665-675°C). Torch oil may be used but should be avoided if possible due to its harmful effect on the catalyst.

8.

The main column overhead gas production will decrease as the reactor temperature and charge rate are decreased. Check that the spillback valves for the wet gas compressor remain in a controlling range. Start fuel gas flow into the LCO stripper vapor return line if needed to maintain main column pressure control as the unit is shut down.

9.

Slowly decrease the reactor-regenerator P controller to a negative value (reactor pressure higher than the regenerator) in preparation for cutting feed. This is intended to create a higher pressure buffer of steam in the reactor between the air in the regenerator and hydrocarbon in the main column.


10. Prepare to bypass feed from the riser. Make sure the main column has a low level in the bottom. Keep the regenerator at a lower pressure than the reactor and put the regenerated catalyst slide valve on manual control. Increase riser steam to maintain catalyst circulation, bypass raw oil to the main column, and reduce the flow of oil. Keep raw oil flowing in and out of the main column until it is verified that any catalyst carried over to the column during shutdown has been flushed out. 11. With the catalyst circulating on steam, begin dropping the reactor level to transfer catalyst into the regenerator. Begin withdrawing catalyst from the regenerator to the hopper to make room for the reactor inventory. Start decreasing the air heater outlet temperature at 100-200°F (50-100°C) per hour. When the regenerator temperatures have dropped below 1000°F (540°C) close the regenerated catalyst slide valve and stop circulating catalyst to the reactor. Maintain steam flow to the riser. 12. When oil is bypassed from the reactor, the gas make will decrease very rapidly. Shutdown the wet gas compressor according to the manufacturer's instructions and block it in. Nitrogen purge the compressor casing. 13. After the wet gas compressor is shut down, block in the pressure controller on the sponge absorber overhead line. Pressure as much liquid as possible back to the main column or to the debutanizer from the other gas concentration columns. After all the liquid is pumped out, depressure the columns to the fuel gas system. When the fuel gas header pressure is reached, depressure the remaining gas to flare. 14. As the main column starts to cool, stop the withdrawal of naphtha and LCO to the sidecut strippers. The naphtha and cycle oils will then overflow their accumulators, wash down the column, and dilute the bottoms material. Open the bypass on the net bottoms product to the raw oil line and start bottoms circulation as during startup. Switch the flushing header source from LCO/HCO to the raw oil line. When the column and


bottoms circuit is adequately flushed, pump out all circuits, the sidecut strippers, and the column bottoms. Shut down the flushing headers. When the column is empty, depressure it to flare. Pump out the overhead receiver to the primary absorber column. Steam out the raw oil charge line, the heat exchange train, and the reactor bypass line to the main column from the raw oil pump discharge. 15. When all the catalyst has been transferred from the reactor into the regenerator, close the spent catalyst slide valve. Lock both slide valves closed. Unload catalyst from the upper regenerator to the hopper. Open the recirculation and cooled catalyst slide valves to drop catalyst from the standpipe and cooler into the combustor, where it will be carried back to the upper regenerator. When all the catalyst is removed from the regenerator, close the unloading valves, shut off the air heater and blind the fuel gas line. Continue cooling the regenerator with the main air blower. When the regenerator temperatures are slightly above the blower discharge temperature, the air blower can be shutdown. 16. Connect vacuum hoses to the catalyst unloading connections throughout the catalyst section and systematically clean out any remaining catalyst. Vacuum catalyst from the unloading nozzles provided at the air heater, the bottom of the riser, the bottom of the reactor stripper cone, and the upper regenerator cone. When cleaning the reactor riser, the regenerated catalyst slide valve should be temporarily opened to unload any remaining catalyst from the regenerator standpipe. 17. When the main column is empty, start steam to the bottom and stripping steam to the sidecut strippers. Be sure fuel gas to the LCO stripper is shutoff and blinded. Maintain steam flow to the riser, feed distributors and reactor stripper. Open the vents on top of the main column and overhead receiver, and drain condensate from the low points. 18. When the reactor and main column are hydrocarbon free, decrease the riser and column bottoms steam flows until only a trace is showing at the


drains located at the vapor line blind. Remove the blind in the vapor line vent, open the vent and install the vapor line blind. Increase the steam flows and continue to steam out for several more hours. Drain condensate from all low points. 19. Connect steam hoses and steam out all pumparound circuits. Drain condensate from low points When the steamout is completed, be sure vents and drains are fully open before stopping steam to avoid pulling a vacuum. 20. When the main column has cooled to 100°F (40°C), start plant water to the overhead receiver. Start the reflux pump and send water to the top of the column. Drain at the bottoms and low points. When water flushing is complete, blind where required for entry. 21. When the gas concentration columns are empty, steam out the unit as required. Any column which will be opened for entry should be water washed after steaming. 22. When the reactor and regenerator have cooled to 300°F (150°C), the manways can be opened to ventilate and cool the vessels. Install air movers as required. Note that when the reactor and regenerator are entered, hot catalyst can still be present, particularly in any diplegs which are closed by their flapper valves. 23. Install blinds to isolate all vessels to be entered. A specific blind list should be prepared for each particular unit.


E. EMERGENCY PROCEDURES Emergencies on an FCC Unit may come in many forms. The operator must first recognize the problem and then take quick action. Response will depend to a large extent on plant design and individual features, the potential danger from the event, and specific circumstances present at the time of the event. As it is impossible to foresee every potential situation, operator judgment, anticipation and training are key components for handling emergencies successfully. The best protection is a thorough understanding of the process, the equipment and the potential dangers involved. The most important aspect of any emergency situation is how to make it safe; personnel safety, environmental safety, and equipment safety are the major objectives for any emergency handling program. Once the safety issues are under control, then the less important concerns, such as maintaining or restoring unit operation can be addressed. Most FCC Unit emergencies will eventually involve a fundamental decision: should raw oil charge to the riser be stopped for a period of time to correct the problem? If this action is necessary, the basic steps to keep in mind are: 1.

Increase lift steam, feed steam and stripping steam.

2.

Bypass raw oil to the main column and stop lift gas.

3.

Establish a negative reactor-regenerator pressure differential (reactor at higher pressure).

4.

Reduce the reactor pressure if possible.

These steps remove the hydrocarbon from the reactor, establish a steam barrier between the regenerator and the main column, and maintain good vapor velocity in the riser to assist catalyst circulation. Once these conditions are established, the problem area can be investigated and corrected.


It is emphasized that every emergency situation must be handled individually depending upon the conditions existing at that particular time. The following procedures must be considered as only guidelines which are not unit specific. Each refiner is responsible for preparing a detailed set of procedures, specific to his unit, for handling any type of emergency event. 1.

Oil Reversal This is one of the most severe emergency events which can develop on the FCC unit, but thanks to built-in safety features in the design, is a very rare event in today's modern unit. A reversal can develop due to a sudden increase in reactor or main column pressure, or a sudden decrease in regenerator pressure. The higher pressure in the base of the riser can cause oil to backflow up the regenerated catalyst standpipe into the regenerator. Once in the regenerator, the oil will burn rapidly, resulting in extremely high temperatures. A severe reversal can cause temperatures to exceed 2000°F (1100°C), well over the design temperature of the regenerator internals. Fortunately, today's slide valves are designed to close within 5 seconds and if the valve low P override is in service, the amount of oil which could potentially reach the regenerator is considerably reduced. If a reversal occurs, the following actions should be taken: a.

Bypass the raw oil charge, increase steam to the riser, increase feed steam and close the regenerated and spent catalyst slide valves. Adjust the reactor-regenerator P to a negative value (reactor pressure higher) and reduce the reactor pressure. Stop lift gas and block in the control valves.

b.

Check the regenerator temperatures. If they pass 1400°F (760°C), decrease the combustor air rate. Increase fluidizing air to the catalyst cooler and open the cooler slide valve to help remove heat.


c.

If the regenerator temperatures continue to rise, start a small circulation of catalyst to the reactor. This will help remove some of the heat from the regenerator. However, do not allow the reactor temperature to exceed 1000°F (540°C).

d.

If it is not possible to circulate catalyst with riser steam, decrease the air rate as much as possible and wait for the oil to burn off and temperatures to drop. Do not shutdown the air blower if it can be avoided, as a lengthy unit shutdown may develop to clean coke out of the regenerator. Monitor and record regenerator temperatures every five minutes.

e.

Add fuel gas to the LCO stripper to maintain main column and reactor pressure. Shutdown the wet gas compressor if necessary and control the pressure from the overhead receiver overpressure control valve.

f.

If it is expected that raw oil feed will be stopped for a long period, begin pre-startup main column bottoms circulations. When temperatures drop below 1300°F (700°C) in the regenerator, start increasing the air rate and internal catalyst circulation. Use the air heater or torch oil if necessary to hold the catalyst at 1200°F (650°C), then continue with the unit restart per the normal procedures.

2.

Behind in Burning / Afterburning While not necessarily an emergency, getting behind in burning or afterburning can develop into one if not recognized and addressed. The high efficiency regenerator is designed for complete CO combustion and a small percentage of excess oxygen should always be maintained to avoid getting behind in burning. However, although extremely difficult, it could be possible that some sudden and unusual feedstock changes might cause this problem to develop. A conventional unit operating in partial combustion is the typical situation where behind in burning is a concern.


It is essential that the coke be burned off the catalyst at the same rate it is produced. This is easily achieved in full CO combustion by maintaining a small amount of excess oxygen in the flue gas. In partial CO combustion, however, excess oxygen is not present and the burn characteristic is represented by the CO2/CO ratio. It is much easier to change the coke production fast enough to exceed the oxygen availability in partial combustion. When all the coke is not burned, the unit gets "behind in burning" with the result that coke begins to accumulate on the catalyst. The catalyst may turn dark gray or black in color and will start to lose activity, causing a drop in conversion. Coke formation in the reactor will continue and the overall coke accumulation on the catalyst can snowball, eventually forcing feed to be bypassed. Afterburn can be defined as CO combustion in the regenerator dilute phase. There will always be some minor degree of afterburn and it is not a problem until it becomes excessive. It is undesirable because there is little catalyst present in this area to absorb the heat and very high temperatures can develop. Signs that the unit is behind in burning are: a.

The temperature difference between the regenerator dilute phase and the dense bed is less than usual, and

b.

The catalyst is noticeably darker in color.

The following actions should be taken to catch up in burning: (1)

Increase the air rate to the regenerator. Take care, however, because when the extra coke is burned off, there will be an increasing amount of excess oxygen in the regenerator and afterburning may occur. To avoid this, air rate increases should be made gradually.


(2)

Increase internal catalyst circulation through the recirculation slide valve.

(3)

Monitor the regenerator temperatures to make sure the dilute phase and flue gas temperatures are increasing.

(4)

Draw frequent regenerated catalyst samples. Compare these to check that the catalyst is becoming whiter, indicating that the accumulated coke is gradually being burned off. This is the only way to know for sure if the problem is being resolved.

(5)

In the event that increasing the air rate does not correct the problem, reduce the reactor temperature, drop the raw oil charge rate, and/or start naphtha quench to the riser.

An afterburn problem is indicated by the following: a.

Increasing T between the regenerator dilute phase and the dense bed.

b.

Increasing regenerator dilute phase temperature.

The following actions should be taken to reduce the afterburn: (1)

Increase the combustor temperature by increasing the catalyst recirculation rate.

(2)

Increase the combustor recirculation rate.

(3)

Increase the regenerator pressure if possible.

(4)

Add CO combustion promoter or increase the rate of promoter addition.

density by increasing the catalyst


(5)

3.

For a conventional bubbling bed regenerator, increase the regenerator catalyst level.

Circulating Water Pump Failure This procedure is in reference to the loss of the flue gas cooler circulating water flow and the catalyst cooler circulating water flow. This could also be caused by a loss of the boiler feedwater supply. a.

Switch to the spare pump if this has not automatically occurred from the low flow switch auto-start control.

Flue Gas Cooler: b.

If the flue gas cooler circulating water cannot be restored, bypass oil from the riser immediately. This is required to prevent high temperatures in the flue gas cooler tubes and downstream piping. The maximum allowable temperature for the cooler tubes is usually 700°F (370°C).

c.

Increase lift steam and feed steam to the riser to maintain reactor pressure so that air cannot enter from the regenerator. Stop the lift gas and block in the control valve.

d.

Close the regenerated and spent catalyst slide valves to stop catalyst circulation.

e.

Shutdown the main air blower as quickly and safely as possible to eliminate hot flue gas flow through the flue gas cooler.

f.

Start fuel gas to the LCO stripper. Continue running the wet gas compressor on total spillback.

g.

If feed to the riser is stopped for more than four hours and the reactor instruments are purged with air (DA), block in the purges. This will


minimize the possibility of auto-ignition of coke due to an accumulation of oxygen. h.

When the water circulation is restored, start the main air blower and bring the unit back on stream following the normal startup procedure. Restart the DA purges just prior to beginning catalyst circulation if these had been stopped.

Catalyst Cooler:

4.

i.

Stop the fluidizing air and close the cooler slide valve. The catalyst in the cooler should be allowed to become stagnant and cool off to minimize the overheating of the tubes.

j.

It is not necessary to bypass feed from the riser. Unit operation will have to be adjusted due to the loss of heat removal from the cooler. Reactor severity will have to be reduced or the feedstock made lighter.

Main Air Blower Failure Loss of the main air blower requires that the unit be shutdown. Usually this is due to failure of the blower lube oil system, though faulty instrumentation may also be responsible. a.

Immediately bypass oil from the riser to the main column. This must be done quickly as the regenerator pressure will drop, causing a loss of regenerated catalyst slide valve P which could set up the potential for an oil reversal. Increase riser steam and feed steam.

b.

Check that the air blower discharge check valve closes completely following the blower shutdown to ensure that no catalyst backs into the blower.


c.

Fully close the catalyst slide valves. The valves would eventually close due to loss of differential pressure across them after the blower fails, but they should be manually closed as fast as possible. Reduce fluidizing air to the catalyst cooler to a minimum.

d.

Start fuel gas to the LCO stripper as quickly as possible. This may allow the wet gas compressor to continue operating with the spillbacks fully open. Reduce the main column receiver pressure as low as possible.

e.

Decrease the main column pumparound flows as possible to hold heat in the column. Stop all net product flows except the bottoms to storage. Establish the main column bottoms recirculation using the bottoms product bypass to the raw oil charge line. If the main column top temperature drops below 230°F (110°C) before oil can be restarted to the riser, back steam into the bottoms steam generators and provide heat to the circulating bottoms stream.

f.

If feed to the riser is stopped for more than four hours and the reactor instrumentation is purged with air (DA), block in the purges. This will minimize the possibility of auto-ignition of coke due to an accumulation of oxygen.

g.

Restart the main air blower when it is ready for service. Begin internal catalyst circulation in the regenerator through the recirculation catalyst slide valve. If the regenerator temperature is above 800°F (430°C), torch oil can be used to reheat the catalyst back to 1200°F (650°C). If it is below 750°F (400°C), light the air heater first. Be sure to raise the reactor pressure above the regenerator pressure after the blower is restarted.

h.

When the catalyst temperature is at 1200°F (650°C), follow the normal startup procedure. Restart the DA purges in the reactor just prior to restarting catalyst circulation if these had been stopped.


5.

Wet Gas Compressor Failure Usually the gas compressor would be lost due to failure of the lube oil system or instrumentation. If this occurs, the main column pressure will be controlled by the overhead receiver overpressure controller vent to flare. The unit can be kept on stream for a short period at reduced throughput if the flaring can be tolerated. If the compressor cannot be restarted soon, the unit will have to be shutdown.

6.

a.

Bypass oil from the riser to the main column and increase steam to the riser and feed distributors. Keep catalyst circulating to the reactor if the shutdown will be for a short duration. Establish a reactor pressure higher than the regenerator.

b.

Continue internal catalyst circulation in the regenerator and use torch oil to keep the catalyst at 1200°F (650°C).

c.

Start just enough fuel gas to the LCO stripper to maintain pressure control on the main column.

d.

Reduce the main column pumparound flows to keep the main column hot. Back steam into the bottoms steam generators as needed to heat the circulating bottoms stream.

e.

Restart the wet gas compressor as soon as possible. Bring the unit back on stream following the normal startup procedure.

Raw Oil Charge Pump Failure If the charge pump fails and cannot be restarted immediately, the unit will need to be shutdown. This could also result from a loss of charge from storage.


7.

a.

Increase riser steam and feed steam. Keep catalyst circulating to the reactor if the shutdown will be for a short duration.

b.

Start fuel gas to the LCO stripper to maintain pressure in the main column. Establish a reactor pressure higher than the regenerator.

c.

Keep the wet gas compressor running on total spillback in preparation for restarting charge to the riser.

d.

Continue internal catalyst circulation in the regenerator and use torch oil as needed to maintain the catalyst at 1200°F (650°C).

e.

When the charge pump is available, bring the unit back on stream following the normal startup procedure.

Main Column Bottoms Circulating Pump Failure The main column bottoms circulation circuits remove about 25% of the heat from the reactor vapors. If the flow is lost for more than ten minutes, the unit will need to be shutdown since the bottoms residence time and temperature would increase to the point where coke formation would begin. The loss of bottoms circulation could also result in heat damage to the disc and donut trays and fines carried up the column. a.

The first priority is to get the material out of the column and try to keep the bottoms temperature down. Try to start the spare pump immediately. Reduce the charge rate to 75% of design and drop the reactor temperature by 50°F (30°C). Decrease column product draws slightly to aid in quenching the bottoms.

b.

If it is not possible to start the spare pump immediately, continue decreasing the charge rate. Increase steam to riser when the charge rate is below 60% of design to assist catalyst circulation. If it is not possible to start either pump within ten minutes, increase the riser steam and feed


steam, bypass the raw oil charge to the main column, and stop the raw oil feed. Establish a reactor pressure higher than the regenerator. Keep catalyst circulating to the reactor if the shutdown will be for a short duration.

8.

c.

Continue internal catalyst circulation in the regenerator and use torch oil to maintain the catalyst at 1200°F (650°C).

d.

Add fuel gas to the LCO stripper as needed to maintain main column and reactor pressure. Keep the wet gas compressor running on total spillback.

e.

When a bottoms pump can be restarted, pump the column bottoms down to a normal level. Restart the normal bottoms circulation and follow the normal startup procedure.

Slide Valve Failure It may be necessary to manually operate a slide valve during an emergency condition resulting from the following faults: a.

Hydraulic Oil Supply Failure

b.

Controller Malfunctions

c.

Physical Damage to the Valve

The slide valve has many redundant safety features to minimize the potential for loss of control. Four methods are provided to move the slide valve: 1.

Electronic actuator (hydraulic power)

2.

Local manual positioner (hydraulic power)


a.

3.

Local handpump (hydraulic power and manpower)

4.

Local handwheel (manpower)

Hydraulic Oil Supply Failure A loss of hydraulic oil pressure could occur due to a loss of the hydraulic oil pump or a ruptured hydraulic oil line. The loss of pressure will result in the loss of control of the slide valve and should leave the valve in the position it occupied when the failure occurred (fail in place). The slide valve must immediately be put on manual control. If the unit operation is steady, a major upset need not result. However, the unit will drift off control over time and immediate action should be taken. (1)

For those cases where the hydraulic lines remain intact, the slide valve main hydraulic oil accumulator will provide enough fluid to the actuator to move the valve two full strokes. However, if the valve has not moved within approximately four minutes, the accumulator will have depressured and be unable to move the valve.

(2)

When the main accumulator pressure is depleted, switch to the reserve accumulator which will again provide two full strokes of the valve or about four minutes of operating time. When the reserve accumulator is switched on, it will activate the control board alarm "Reserve Accumulator in Service". To extend the use of the reserve accumulator, the board operator can switch the reserve accumulator in and out of service to make valve position changes.

(3)

When the reserve accumulator pressure is low, the "Reserve Accumulator Low Pressure" alarm will be activated in the control room. At that time it will be necessary for a field operator to manually move the valve with the local handpump or handwheel. Refer to the manufacturer's instructions for this operation.


(4)

When operating the regenerated catalyst slide valve by handwheel, watch the valve P very closely. If P is lost, a flow reversal could develop.

(5)

If the hydraulic oil pressure can be reestablished quickly, put the valve back into regular operation. For a prolonged failure, the unit will need to be shutdown. Do not attempt to operate any slide valve by the handwheel for normal operation. This manual action is much too slow to react to sudden process changes.

b.

Controller Malfunctions A slide valve will close upon the loss of its actuator electronic power or instrument signal. There is no alarm for these failures and they will only be apparent to the board operator when he sees the slide valve drifting to the closed position. In this case, the slide valve should be operated using the hydraulic oil system local manual positioner, the local handpump or the handwheel. When operating the slide valve locally, care must be taken to maintain a positive slide valve P. If the electronic input signal can be reestablished quickly, put the valve back into regular operation. For a prolonged failure, the unit will need to be shutdown.

c.

Physical Damage to the Valve The loss of slide valve control could be the result of excessive disc erosion or a broken stem. Corrective actions to be taken depend on which valve is affected: (1)

Spent Catalyst Slide Valve Failure (a)

The worst situation occurs if control of the valve is lost in an open position and the valve cannot be closed. For this case, increase riser steam and feed steam. Bypass raw oil charge to


the main column and shut down the charge pump. Establish a reactor pressure higher than the regenerator and reduce the main column pressure.

(2)

(b)

Since the valve is stuck open, it is possible that the reactor could empty of catalyst. This may allow air to enter from the regenerator which could result in a fire or explosion. Continue catalyst circulation using steam to the riser to hold a reactor catalyst level. Allow the regenerator temperatures to fall to 1000°F (540°C) and maintain this level with torch oil or the air heater. Having a lower catalyst temperature will allow more catalyst to be circulated without generating a high reactor temperature.

(c)

Shutdown the wet gas compressor and allow the main column to cool down. Pump out as much oil as possible. Start steam to the bottom of the column and vent from the overhead receiver to flare. Maintain enough pressure to keep the reactor pressure above the regenerator.

(d)

Start unloading as much catalyst as possible from the regenerator without losing circulation to the reactor. Continue with the shutdown procedure to eventually install the reactor vapor blind, and prepare the regenerator for entry and repair of the valve. If the level of catalyst is lost in the reactor and the spent catalyst slide valve loses its P, the air blower should be shutdown immediately and steam to the riser increased as much as possible to keep air out of the reactor.

Regenerated Catalyst Slide Valve Failure (a)

If control of the valve is lost with the valve in an open position, increase the riser steam and feed steam while decreasing the raw oil charge to 60% of design. Increase the riser steam as


necessary to keep catalyst circulating through the riser. Reduce the reactor pressure as possible to assist catalyst flow up the riser. (b)

Allow the reactor temperature to drop to reduce coke make but do not go below 900°F (480°C). As the regenerator temperatures drop, transfer as much catalyst as possible to the storage hopper without losing the recirculation and regenerated catalyst flows.

(c)

When the raw oil charge rate is decreased to 60% of design, start a small flow of dry steam to the main column bottoms.

(d)

When prepared, bypass the raw oil to the main column and shut down the charge pump. Immediately shutdown the main air blower, riser steam and feed steam. Maintain reactor stripping steam. With no steam to transport the catalyst, it will slump and plug the riser wye section. This is done to prevent any air from entering the reactor. While this is less than desirable for easily restarting operation, it is a necessary safety protection for the unit.

(e)

Fully close the spent catalyst slide valve to maintain a reactor level. Increase steam into the main column bottoms. Increase the reactor stripping steam to maximum to purge the reactor.

(f)

After bypassing charge, shutdown and block in the wet gas compressor.

(g)

Continue steaming the main column and vent pressure to the flare. Pump out the oil from all circuits in preparation for installing the reactor vapor blind. Continue with the shutdown procedures to isolate and prepare the reactor and regenerator for entry to repair the valve.


(3)

(4)

Recirculation Slide Valve Failure (a)

Usually, failure of this valve does not require a unit shutdown unless it fails near its fully closed position. Small adjustments to other operating variables should make it possible to keep the unit on stream.

(b)

If the valve is stuck in one position, there will be no direct control over the combustor temperature. Check the regenerator temperatures, combustor catalyst density and main air blower discharge pressure.

(c)

If the valve is stuck too far open, the catalyst density in the combustor will increase which will increase the blower discharge pressure. If the air flow becomes limited as a result, adjustments can be made to the regenerator pressure to compensate.

(d)

If the valve is stuck too far closed, the catalyst density and temperature will decrease, and there may not be enough heat to complete the coke combustion or enough catalyst to absorb the heat from the combustion. This could lead to afterburning in the upper regenerator. In this case, the air rate can be decreased slightly to compensate. In severe cases, the charge rate and reactor temperature should be reduced to decrease the coke make.

Flue Gas Slide Valve Failure If the control of one of the discs is lost on the double disc valve, it may be possible to control the reactor-regenerator differential pressure using only the second operable disc. However, in the event that adequate P control cannot be maintained, or that control of both discs is lost, then the unit would need to be shut down.


If the valves fail toward the closed position resulting in a rising regenerator pressure, the main air blower would have to be shutdown immediately. The rising regenerator pressure would stop spent catalyst transfer and create a dangerous situation in which air could be forced into the reactor. On the other hand, if the valves were to fail toward the open position, the regenerator would depressure as if the main air blower had failed. Therefore, the unit should be handled as in a main air blower failure. (5)

Catalyst Cooler Slide Valve Failure The loss of control on this valve will not require a unit shutdown. The cooler will either take more or less heat out of the regenerator than desired, depending on whether the valve is too far open or closed. This situation can usually be handled by adjusting the cooler fluidizing air to compensate. In the extreme case, the reactor and feedstock conditions may have to be adjusted to make more or less coke.

9.

Catalyst Cooler a.

Loss of Circulating Boiler Feed Water The following steps must be taken if circulating water flow cannot be restarted to the cooler. With the loss of circulating water, the catalyst cooler tube temperature will quickly increase to the surrounding catalyst temperature. (1)

To minimize heat transfer, immediately stop fluidizing air flow to the catalyst cooler.

(2)

Confirm that the cooled catalyst slide valve has closed.


b.

(3)

Depressure the catalyst cooler steam drum to 50 psig (3.5 kg/cm2). This will reduce the metallurgical stress on the cooler tubes.

(4)

Determine the cause of the water circulation failure and make necessary repairs.

(5)

To avoid thermally shocking the tubes, let them cool to below 450°F (230°C) before restarting the water circulation. Use the thermocouples located in the cooler as an indication of the tube temperature.

(6)

Restart water circulation and watch for a high BFW makeup rate indicating a tube is leaking.

(7)

When water circulation is stable, begin fluidizing air to the cooler. Adjust the air rate for the desired heat removal. If air will not flow through the lances, it may be necessary to attach a higher pressure gas source, air or nitrogen, to the header to help blow the lances free.

Catalyst Cooler Tube Rupture If a tube in the catalyst cooler ruptures, as indicated by a slight pressure surge in the regenerator and a sudden increase in boiler feed water demand, the following actions must be taken: (1)

The circulating water pumps must be shutdown and the makeup water stopped.

(2)

Stop fluidizing air to the cooler.

(3)

Confirm that the cooled catalyst slide valve has closed.

(4)

Depressure the steam drum and isolate the system.


(5)

It is possible to leave the cooler in this state and continue the FCC operation indefinitely. The stagnant catalyst in the cooler will eventually cool to protect the tubes against high temperature. Shutdown of the FCC unit can be planned for a later date to repair the ruptured tube.

In both of the above situations, loss of the catalyst cooler will require an adjustment in FCC operating conditions to compensate for the lost heat removal and limit regenerator temperature. This may require a reduction in charge rate or a lighter feedstock. The FCC unit does not need to be shutdown in either event. 10. Lift Gas Failure Lift gas is not required to keep the FCC unit operating. Usually, lift steam can be used to compensate for any loss of lift gas. The system is designed to shut off lift gas to the riser when oil charge is bypassed. a.

Loss of Lift Gas Increase lift steam to the riser. If lift gas cannot be restored, block in and isolate the line.

b.

Upset in Gas Concentration Unit If an upset occurs which increases the heavy material in the lift gas (greater than 10% C3+), reduce or discontinue the lift gas, increasing the lift steam to compensate. The C3+ material can crack to large quantities of light gas which could upset the compressor and absorbers.


11. Instrument Air Failure Loss of instrument air will result in all control valves moving to their fail position. If instrument air cannot be restored immediately, the unit must be shutdown. a.

The raw oil charge is bypassed to the main column on air failure. The charge pump should be shut down until air is restored.

b.

Steam to the riser and feed distributors opens fully on air failure. Lift gas to the riser will fail closed.

c.

The main air blower governor will fail to the minimum speed and the antisurge snort valve will fail open. The catalyst slide valves will not fail as they are electrohydraulic, so they must be closed by operator action. If instrument air cannot be restored quickly, the blower may be left running and the snort valve, recirculation catalyst slide valve and flue gas slide valves and be adjusted to reestablish catalyst circulation in the regenerator. The catalyst temperature can then be maintained at 1200°F (650°C) using the air heater or torch oil in preparation for the unit restart.

d.

The wet gas compressor spillback valves will fail open. Unless sufficient fuel gas can be put into the LCO stripper to maintain column pressure, the compressor should be shut down.

e.

The circulating main column bottoms control valves to the steam generators will fail open. If instrument air is not restored quickly, close the block valves and use the bypass valves to reduce these flows and minimize cooling in the main column.

f.

Usually an instrument air failure will be of short duration. When air is restored, control valve action will return and the unit can be restarted according to the normal startup procedure.


12. Electrical Power Failure In a general power failure, all process equipment will shut down except those pumps and compressors driven by steam turbines. The power pack for instrumentation will fail momentarily until the emergency power system cuts in, restoring instrument circuits. In the event of a power failure, the following actions should be taken: a.

Increase lift steam and feed steam to the riser, bypass the raw oil charge to the main column and close the raw oil charge control valve. If it is expected that the power outage will be temporary, continue catalyst circulation using steam. Stop lift gas flow. Establish a reactor pressure higher than the regenerator and reduce the pressure in the main column.

b.

Close the catalyst cooler slide valve and reduce fluidizing air to minimum flow to minimize heat removal in the regenerator. If catalyst circulation is continued, use the air heater to maintain the catalyst at 1200°F (650°C). The turbine driven circulating water pumps for the catalyst cooler and flue gas cooler should continue operating.

c.

If the wet gas compressor is motor driven, it will shut down. If turbine driven, keep the compressor running on total spillback. In either case, start fuel gas to the LCO stripper to maintain column pressure.

d.

The main column bottoms pumps are turbine driven and will continue operating. Maintain circulation but flow through the steam generators may have to be stopped if BFW makeup has stopped. With all column pumparounds and overhead reflux stopped, the column will hold heat for some time.

e.

When power is restored, restart all pumps and follow the normal startup procedure.


13. Cooling Water Failure If cooling water failure occurs for a period greater than about ten minutes, the unit will probably need to be shut down. The largest potential danger is to the wet gas compressor. Loss of cooling water to the interstage cooler will result in a high suction temperature to the second stage which can cause significant damage to the compressor. If cooling water cannot be restored, take the following action: a.

Shutdown the wet gas compressor and bypass raw oil charge to the main column. Increase lift steam and feed steam to the riser. Stop lift gas flow and block in the control valve.

b.

Maintain reactor pressure by starting fuel gas to the LCO stripper. Establish a reactor pressure higher than the regenerator.

c.

Continue catalyst circulation using riser steam if it is expected that cooling water will be restored within a reasonable period. If not, catalyst circulation to the riser can be stopped.

d.

Continue internal catalyst circulation in the regenerator and use torch oil to maintain the catalyst at 1200°F (650°C).

e.

When cooling water is restored, bring the unit back on stream following the normal startup procedure.

14. Steam Failure Since the FCC unit is a major steam producer, it is less susceptible to steam failures. When a boiler failure does occur, steam pressure will begin dropping. Non-critical users will be shed first, so the FCC unit has time for an orderly shutdown. A main objective is to purge hydrocarbon from the reactor before steam is lost completely. The following actions should be taken:


F.

a.

Bypass raw oil charge to the main column and shutdown the charge pump. Increase lift steam, feed steam and stripping steam to maximum flows to purge the reactor quickly.

b.

Continue catalyst circulation on riser steam for a short period, then close the regenerated and spent catalyst slide valves.

c.

Since the main column bottoms pumps are turbine driven, they will not be operating for long. Therefore, cool the main column quickly to lighten the material in the bottom to avoid coking when flow has been lost. Pump out bottoms to hold a low level. Shutdown the wet gas compressor and begin to depressure the main column to flare. Start steam to the main column bottom to help purge the column.

d.

The main air blower is usually turbine driven and will eventually be stopped. Internal catalyst circulation in the regenerator can be maintained until that time. If it is expected that steam will be restored within a short period, keep the catalyst hot with the air heater until the blower shuts down.

e.

When steam is restored, start turbines carefully due to the potential for condensate in the lines. Restart the unit according to the normal startup procedures.

CATALYST HANDLING Handling FCC catalyst is a relatively simple job. FCC catalyst is a relatively strong material and will function quite well if not seriously abused. Because of its characteristic of being easily fluidized, it is easily moved to and from the unit using the principles of a pneumatic conveyor.


Storage is provided usually in two large hoppers, one for fresh catalyst and the other for equilibrium catalyst (removed from the unit). Capacity of these hoppers may vary from 100-500 tons, depending on the size of the unit. The fresh catalyst hopper will have an automatic catalyst loader, and is usually slightly smaller than the equilibrium hopper. The equilibrium hopper must be sized to hold the entire catalyst inventory of the unit, plus some contingency. Both hoppers have relief valves to prevent overpressuring, and a variety of loading and unloading lines. Instrumentation is limited to several pressure gauges and a level gauging device. Both hoppers are usually built to withstand a full vacuum. A steam ejector is used to provide this vacuum to unload catalyst into the hopper. Refer to Figure 5. 1.

Loading Catalyst to the Hopper Catalyst is delivered from the manufacturer in trucks, railcars, or large plastic-lined boxes. Equilibrium catalyst usually leaves the refinery in the same manner. Before the catalyst is loaded to the hopper, all lines and vessels should be inspected. The important points to check are: a.

All lines and vessels are built according to specification.

b.

Pressure taps are open and the level gauging devices are working.

c.

The lines and vessels are free of foreign material, especially water or oil. The air lines should be checked by blowing all lines until they are clean and dry. The large diameter loading lines can be blown using the hot air from the regenerator during dryout. Any water in the lines or hoppers will form a sticky mud which makes normal handling impossible as this wet catalyst is extremely difficult to remove. Prolonged blowing with air or cleanout by hand will work, but the best method is prevention by keeping the system dry.


The next step is to load catalyst to the appropriate hopper. This is important because of the tremendous activity difference between fresh and equilibrium catalyst. Mistakes here can be highly embarrassing. The catalyst may be loaded to the hopper by blowing it in from the truck or railcar, evacuating the hopper and pulling it in, or a combination of the two. A vacuum can be pulled on the hoppers by commissioning the ejector after lining up the valves to the appropriate hopper (Figure 5). Open the loading valves first at the top of the hopper, then at the catalyst loading area. The catalyst should flow freely. If the catalyst arrives on site in boxes, a cone shaped open hopper can be used to transfer it to the catalyst hopper. The appropriate hopper would be evacuated, and a small amount of carrying air supplied at the base of the cone to help move the catalyst. The amount of catalyst in the hopper should be measured regularly. A chart showing tons of catalyst as a function of outage will give a quick inventory reference. This can be calculated knowing the hopper volumes as a function of height (be careful when calculating the cone section at the base) and the catalyst densities of approximately 50 Ib/ft3 (800 kg/m3) for fresh catalyst and 55 Ib/ft3 (880 kg/m3) for equilibrium catalyst. More exact values of these multipliers can be obtained from the catalyst data sheets or calculated after a known weight of either catalyst has been loaded into its appropriate hopper. Knowing the amount of catalyst remaining in the hoppers is the only way to control inventory and to determine how much catalyst is being used in the unit. Using the hopper level gauging devices may not always provide accurate catalyst level measurements. These have a tendency to sink in the catalyst and give false readings. Also, the catalyst may form a cone inside the hopper which can give a false reading. A proper measurement can be obtained by depressuring the hopper and letting it settle for at least 1


hour. Then the gauging hatch is opened and the outage measured with a hand tape. 2.

Loading Catalyst to the Regenerator The procedure for loading the catalyst to the regenerator begins with pressuring up the hopper. Blower air is good for this purpose because it is hot and dry, although dry plant air can also be used. If possible, the air should be injected at the base of the hopper so that it will fluff the catalyst as it flows upward. The pressure in the hopper must be higher than the pressure in the regenerator, or the catalyst will not move. Refer to Figure 5. a.

Make sure the regenerator is ready to receive catalyst. Refer to the normal startup procedures.

b.

Gauge the hopper and pressure it to 40 psig (2.8 kg/cm2) by adding air through the bottom.

c.

Line up the loading line from the hopper to the regenerator start the carrying air to the line while leaving the block valve under the hopper closed. Set the air rate to achieve a velocity of ~50 ft/sec (15 m/sec) in the line or so that the pressure at the end of the line increases by ~1-2 psi (0.1-0.2 kg/cm2) if no FI is provided.

f.

Open the catalyst loading valve at the bottom of the hopper slowly until the loading line pressure rises 10-15 psi (0.7-1.0 kg/cm2), signifying that catalyst is flowing through the line. The catalyst block valve should be 1/4 to 3/4 open. The operator can avoid plugging the line by carefully watching the pressure gauge and/or carrying air FI on the loading line at the base of the hopper. With normal loading, the gauge will oscillate slightly around a pressure about 5-10 psi (0.35 –0.7 kg/cm2) above the regenerator pressure. If this gauge on the loading line shows an increase in pressure, the catalyst hopper


block valve should be closed. The carrying air can then clear the line before excess catalyst can plug it. The optimum catalyst transfer technique is best determined through trial and error operation for each specific installation. g.

Maintain the hopper pressure by adding air to the bottom of the hopper as this generally gives smoother catalyst flow. However, if this operation reduces the loading rate, the hopper can be pressure from the top.

h.

When the correct regenerator inventory is achieved, close the loading valve at the bottom of the hopper. Blow out the loading line and close the loading valves at the regenerator, leaving the loading line under plant air pressure.

i.

Gauge the hopper and determine the quantity of catalyst transferred during the loading.

Plugging Problems If there is a problem with the catalyst bridging above the block valve at the bottom of the hopper, quickly opening and closing the valve 1-2 turns will usually break the bridge. If this fails, blast the bottom of the hopper with air. The catalyst loading line may plug occasionally. When this happens, close the catalyst block valve at the base of the hopper. Starting at the regenerator and working back to the hopper, fully open the blast points one by one. If this fails, pounding on the line with a non-sparking hammer will work as a last resort. When the pressure gauge on the loading line at the base of the hopper falls to just above the regenerator pressure, the line is clear. High pressure air or nitrogen may be used to clear a plug, but do not exceed safe working pressure of the line.


3.

Fresh Catalyst Makeup For loading fresh catalyst, an automatic loader usually works quite well. These feed catalyst to the regenerator in small amounts on a steady basis. This is much better than dumping a day's supply into the unit in 20 minutes, leading to a marked activity change. The fresh catalyst loading line is usually much smaller than the large line used for initial catalyst loading. The smaller line has a larger volume of air per unit area passing through it, which decreases the potential of plugging problems. The rate of fresh catalyst makeup will vary depending on unit performance objectives, feedstock quality and how well the unit "holds" catalyst. Some refiners add only enough fresh catalyst to balance losses. Others add makeup at a higher rate to maintain a certain activity level and then periodically withdraw equilibrium catalyst to balance unit inventory. The level of metals in the feedstock will have a strong impact on catalyst makeup rates.

4.

Unloading Catalyst from the Regenerator The pressure in the regenerator is the driving force to unload catalyst to the equilibrium hopper. Commission the steam ejector and line up the valves to pull a vacuum on the equilibrium hopper. Then open the unloading valves, first on the hopper, then on the regenerator. It is usually not necessary to use the blast connections, although they are available if needed. Withdraw the appropriate amount of catalyst by watching the regenerator level. When the unloading is finished, close the regenerator block valve. Clean out the unloading line by opening the nearest blast point. Then shut off the ejector and bleed enough air into the hopper to raise it to atmospheric pressure.


5.

Catalyst Valve Purges An air purge is normally provided to the seat and bonnet of all valves in the catalyst loading/unloading system. This is to prevent catalyst from collecting in these areas and interfering with valve operation. Forcing a valve shut when the seat is full of catalyst will only damage the valve. For proper operation, open the purges and clear the seat and bonnet before the valve is operated. Do not leave the purges on during the loading operation as excessive valve erosion can result. Rather, purge the valves each time they will be adjusted and particularly just before the valves are closed.

G. SPECIAL OPERATIONS 1.

Catalyst Sampling When drawing hot catalyst samples, gloves, long sleeve clothing and face shields should always be worn. Always use a metal container to collect the sample. Refer to Figure 6 for the valve locations in the following procedure for drawing catalyst samples. a.

Spent Catalyst Sample NOTE: Air should never be used when purging into the reactor or the spent standpipe as oxygen can initiate coke burning. (1)

The reactor level controller should be switched to manual as purging the sample connection could upset the slide valve differential pressure controller.


b.

(2)

Check that all blast and sample connection valves are closed, then open sample valve E and steam valve B. Vent steam to the atmosphere until it is dry.

(3)

Close sample valve E and open the nozzle valve F. Steam purge the sample line into the standpipe.

(4)

Open sample valve E and reduce the steam flow through valve B until a small flow of catalyst is obtained. If catalyst does not flow out of sample valve E, close valve E and blast the standpipe through steam valve C. Repeat the step until catalyst can be obtained.

(5)

Take the sample and close valve E.

(6)

Open steam valve B and purge into the standpipe.

(7)

Close nozzle valve F and then steam valve B.

(8)

Open sample valve E.

(9)

Put the reactor level control back in automatic.

Regenerated Catalyst Sample (1)

The reactor temperature controller should be switched to manual as purging the sample connection could upset the slide valve differential pressure controller.

(2)

Check that all blast and sample connection valves are closed, then open sample valve E and air valve A. Vent air to the atmosphere until it is dry.


2.

(3)

Close sample valve E and open the nozzle valve F. Air purge the sample line into the standpipe.

(4)

Open sample valve E and reduce the air flow through valve A until a small flow of catalyst is obtained. If catalyst does not flow out of sample valve E, close valve E and A, and blast the standpipe through steam valve C. Repeat the step until catalyst can be obtained.

(5)

Take the sample and close valve E.

(6)

Open air valve A and purge into the standpipe.

(7)

Close nozzle valve F and then air valve A.

(8)


(9)

Put the reactor temperature control back in automatic.

Blasting Catalyst Standpipes Sample and blast connections are located just above the spent, regenerated, and recirculation catalyst slide valves. There may also be blast connections located higher on the regenerated standpipe. These connections are used not only to sample catalyst but also to clear the standpipes of any plugs which might occur. The following is a procedure for blasting these standpipes using steam (refer to Figure 6). NOTE: Air should never be used to when purging into the reactor or spent standpipe. a.

Place the appropriate slide valve controller in manual when blasting a standpipe.


3.

b.

Check that all blast and sample valves are closed, then open sample valve E and steam valve C. Vent steam to the atmosphere until it is dry.

c.

Close sample valve E and open nozzle valve F. Purge steam into the standpipe until the plug is cleared. Do not blast for extended periods (more than several minutes) with the steam valve full open. Potential erosion to the standpipe refractory is a concern. It is better to use short intermittent blasts than one long continuous blast.

d.

Close nozzle valve F and steam valve C.

e.


f.

Place the slide valve controller back in automatic control.

Direct Fired Air Heater The direct fired air heater is used to heat up the circulating catalyst inventory during startup or maintain heat during temporary shutdowns. The heater is only fired when the main air blower is operating. A potentiometer should be connected to the temperature indicator located at the outlet of the air heater so the field operator can monitor the temperature locally when the heater is fired. Refer to the manufacturer's operating manual for detailed operating information. The general heater steps are as follows: a.

Pull the blind in the air heater fuel gas line after the main air blower is put in operation.

b.

Switch on the ignition power.

c.

Adjust the inlet damper to direct less air behind the burner block.


d.

Open the fuel gas line supplying gas to the pilot burner. Press the button on the local panel supplying power to the pilot ignitor. Use the observation ports to confirm that the pilot has been lit.

e.

When the pilot has been lit, slowly open the main burner gas valve until a flame is obtained. Use the observation ports to confirm that the main burner is firing.

f.

Adjust the inlet damper to get a proper flame color.

g.

While the air heater is relatively simple to operate, the following precautions should be observed: (1)

An operator should be stationed at the air heater to monitor the flame any time the heater is in operation.

(2)

If the air flow to the regenerator stops for any reason, the fuel gas supply valves must be shut off immediately.

(3)

The fuel gas to the heater must be free of any liquid. Slugs of liquid hydrocarbon temperatures.

4.

can

cause

damage

due

to

high

Flushing System The flushing oil system is required for two reasons: a.

The main column bottoms contains catalyst fines which can settle out and plug instruments and small piping, and which can erode pumps and control valves.

b.

Maintenance on the packing of the hot main column bottoms pumps is reduced if the packing glands are cooled by a flushing stream.


The throat bushings and wear rings of the main column bottoms pumps are flushed with hot heavy cycle oil taken from the circulating HCO pump discharge. Hot material is used to avoid the thermal stresses which could result if a cooled oil was injected into the hot pump casing. HCO is used since LCO would flash and cause pump cavitation. The flush rate is adjusted by closing the inlet flush valve, observing the static pressure, and opening the flush valve until the pressure increases about 10 psi (0.7 kg/cm2). Light cycle oil is used for all other flushing services with the exception that raw oil is used during startup and can be used during an emergency. On units with resid feed stocks light gas oil from storage should be used as a backup instead of raw oil. The flushing oil is taken from the LCO product cooler outlet and passes through a 30 mesh strainer, which should be cleaned whenever the P exceeds 15 psi (1 kg/cm2). The main column bottom level instrument and sight glass similarly flushed to prevent the accumulation of catalyst. A 1/8 inch (3 mm) restriction orifice is used to regulate the flow to each location. The main column bottoms pump packing glands are flushed with LCO through a lantern ring to cool the shaft and packing. The flush is returned to the main column with the circulating HCO return. This flush is adjusted by closing both the supply and return valves, and observing the lantern ring pressure. The flushing supply valve should then be adjusted so that the flush oil pressure is 15 psi (1 kg/cm2) higher, while the return valve is adjusted to maintain the return temperature above 150°F (65°C). It is advisable to mark the desired flush oil pressures on the gauges so that adjustments can be easily made.


5.

Main Air Blower Discharge Check Valve a.

Application The main air blower discharge check valve is a swinging disc type valve installed on the discharge of the air blower for safety reasons. The weight of its disc is balanced to cause the valve to close when air flow stops, preventing the reverse flow of air and catalyst through the blower. The valve is equipped with a spring loaded air cylinder and an oil dashpot. The air cylinder provides an automatic device which assists the valve to close quickly, while the dashpot dampens the excessive swinging of the disc under a low or pulsating flow condition so that the disc does not slam on the seat.

b.

Counterweights The steel disc of the valve is of heavy construction to resist distortion from pressure loads or high temperatures. Counterweights are provided so that the full weight of this disc will not have to be carried by the gas stream going through the valve. By counterweighting approximately 75% of the disc weight, the valve opens wide under the design flow condition and results in minimum pressure drop through the valve. Counterweighting to hold the valve open when there is no gas flow through the valve must not be done as this negates the safety protection of the valve. If the disc is moved off the seat by manually pulling on the counterweight lever when there is no gas flow through the valve, the disc should return to the seat freely and rapidly upon release of the counterweight lever.

c.

Air Cylinder The external spring loaded air cylinder is designed to assist in the rapid closure of the valve under emergency conditions. It can not be used to open the valve and under normal operating conditions with


air pressure supplied to the cylinder, there is no physical connection nor restriction to the movement of the valve. Air pressure in the range of 60 to 125 psig (4 to 9 kg/cm2) is required to offset the air cylinder spring force. The air pressure supplied to the cylinder is through a 3-way solenoid valve which is activated by the main air blower discharge flow transmitter low flow switch. This valve operates such that when it is actuated, it vents the air pressure in the cylinder to atmosphere. Once the cylinder is depressured, its spring force acts to jolt the valve disc loose so that it may close freely. The release of air from the cylinder, however, does not assist in the closing action. There are two advantages associated with using this type of externally actuated check valve. If the valve tends to stick open after several months of operation, the cylinder's spring provides a breakaway force to start the valve closing. Also, the air cylinder is tripped before the blower stops rotating, so that the spring forces the disc nearer to the seat before air flow stops. This ensures that the valve will seat before reverse flow can develop. While the spring in the air cylinder exerts a substantial force, the closing force creates only a relatively small back pressure. This is important from the standpoint of operation, because even in the event of accidental tripping, flow through the valve will continue. Thus, if the shutdown system should malfunction while the blower is operating normally, the unit can continue in operation until the instrumentation is repaired. d.

Oil Dashpot The oil dashpot contains a loose fitting piston which moves through an oil filled cylinder. As the check valve disc moves, it moves the piston through the oil, forcing oil from one side to the other through a


bypass line containing a regulating check valve. This valve restricts the oil flow as the disc opens, preventing rapid opening, while allowing unobstructed oil flow and rapid movement of the disc when the check valve closes. This system will dampen the action of the disc in a low flow or pulsating condition to protect the valve seat. The regulating valve should be set initially in a 3/4 open position to provide some restriction to the oil flow on opening of the disc. Before the check valve is put into operation, the dashpot should be filled through the plug on the bypass connection with a light lubricating oil such as SAE 10W. 6.

Slumped Riser During Shutdown With raw oil charge bypassed, riser flow is maintained usually with only steam. If this steam is stopped during catalyst circulation, or the regenerated catalyst slide valve is maintained open manually with insufficient steam to the riser, the catalyst can slump and fill the wye section and bottom of the riser. If such a situation develops with hot catalyst and oil present, the wye section stagnant catalyst can coke up and require hammer and chisel to remove. If such a situation develops with cooler catalyst and condensate, catalyst mud can form which is also difficult to remove. Thus, it is imperative that steam always be maintained to the riser during shutdown situations, at least until the regenerated catalyst slide valve has been closed and the riser is free of catalyst. Should a slump occur, blast points at the bottom of the wye section should be used to clear the plug with steam if possible. Steam can also be blasted through the normal lift steam line. If mud has formed and cannot be blasted clear, the 4 inch (150 mm) blankoff connection at the base of the wye can be opened to attempt to drain the wet catalyst. A vacuum hose can be connected to assist pulling the catalyst out.


If all attempts fail, the lift or feed distributor in the wye should be pulled to allow entry for manual removal of the catalyst. NOTE: Always assume the catalyst will be hot and take all necessary safety precautions.


Figure 1: Reactor/Regenerator Section Products to Main Column Flue Gas

FIC

Prestripping Steam

FIC

Stripping Steam

FIC

Fluffing Steam

FIC

Atomizing Steam

Steam HS

Split Range

Reactor Bypass to Main Column

FO

FIC

Raw Oil from Preheat

FC

Torch Oil from HCO or Raw Oil

Main Column Bottoms Recycle

Air Fuel Gas Pilot Gas

FIC

Direct Fired Air Heater

FIC

Lift Steam

Startup Naphtha from Main Column Overhead Steam

FIC

Startup Steam

FIC

Lift Gas from Gas Concentration Unit FCC-P001


Figure 2: Main Column Lower Section Circulation Bypass

FIC

LIC

Startup Filling Line To LCO Pumparound Return Line

Debutanizer Reboiler Torch Oil

FIC

Pump Flushing Oil

Disc and Donut Minimum Flow FIC

MCB Quench Reactor Product Vapor

Steam to Superheater

FIC

Startup Steam

FIC HS FO

Raw Oil to Reactor

FC

T

Startup Heating Steam Trap

Startup Circulation/ Recycle

FIC FIC

MCB Product

Torch Oil Raw Oil from Surge Drum

Flushing Oil

CW

FCC-P002


Figure 3: Main Column Upper Section Wet Gas Compressor First Stage Spillback Signal to Wet Gas Compressor Controls

PRC

PIC

To Flare Header

FC

PSV

To Wet Gas Compressor Suction Drum

CW LIC

TIC FIC

FIC

Net Overhead Liquid to High Pressure Receiver Sour Water

Startup Circulation Bypass FI

Gas Concentration Unit Heat Exchangers

Startup Filling from Raw Oil/MCB

HCN Stripper FI

Startup Fuel Gas

FIC

HCO

HCN Product

LCO Stripper Steam

CW

FIC LCO Product BFW CW Preheater

Gas Concentration Unit Heat Exchangers

LCO to Flushing Oil Startup Circulation Bypass FCC-P003


Figure 4: Reactor Vapor Line Blind Reactor Overhead Vapor

Vent to Safe Location Blind or Spacer Steam DG PI

Blind or Spacer

Drains

FCC-P004


Figure 5: Catalyst Handling Facilities To Atmosphere or ESP Inlet

Ejector Manual Gauging Hatch

Steam LI Air Purge

Catalyst Hopper

GRATING

Plant Air

Plant Air Plant Air RO

Catalyst Volume Pot

Everlasting Valves

Plant Air

Catalyst Makeup Logic Controller

Sight Flow Glass

FI

PG

FIC

Catalyst Loading and Unloading from Shipping Containers Catalyst Filling and Withdrawal To/From Regenerator

"DA"

Plant Air

Removable Spoolpiece (blankoff when not unloading Ecat)

PI "DA"

Catalyst Makeup to Regenerator

FCC-P005


Figure 6: Catalyst Standpipe Blast and Sample Point

Plant Air Steam A B Strainer D

C RO

E Sample Point

F

Catalyst Standpipe

Min

FCC-P006

157048 Safety Page 1

SAFETY A.

GENERAL

Safety is a broad field that covers a variety of topics from fire fighting to toxic chemicals. Awareness of safety in the refinery is extremely important. As a basis for a safety program, regulations and guidelines from government and professional groups should be studied for their effect on this unit and then implemented by the refinery staff. The staff must develop its own coordinated plant-wide method of implementing safe and proper procedures, and of handling emergencies, which should include: 1.

Training to develop the operator's knowledge of equipment operation, and the nature of the FCC unit.

2.

Good communication among all involved groups concerning activities in progress, and of prime importance, as well, is the use of common sense.

3.

Emergency planning and practice drills.

The following discussion is not intended to supersede or otherwise replace refinery safety practices, but rather, should be used as a supplement or reminder of some of the important safety features.

B.

PRECAUTIONS FOR ENTERING VESSELS

The reactor must be adequately cooled before air is admitted into the vessel. If the reactor is not sufficiently cooled, auto-ignition of hot coke deposits in the reactor or its vapor line may result. The reactor must be cooled to below 300°F (150°C) before any manways or nozzles are opened. Riser or stripping steam can be used to help cool the vessel.


The following safety precautions should be followed to protect personnel entering a vessel against toxic materials, such as H2S, which are present in the RCC Unit: 1.

The vessels should be isolated by positive action, such as blinding, to exclude all sources of hydrocarbon, steam, etc.

2.

An air mover should be installed at the vessel's manway to sweep away any vapors and provide a continuous supply of fresh air.

3.

Responsible personnel must test the atmosphere in the vessel for explosivity, toxic fumes, oxygen content, dust, etc. Permission for vessel entry should be given only after this testing has been done.

4.

Personnel entering the vessel must be equipped with a pressure demand respirator that is in proper working condition, and is connected to a suitable fresh air supply.

5.

Separate air supplies which are independent of electrical power should be available for immediate use and transfer to personnel in the vessel.

6.

Personnel entering the vessel should wear a safety harness with a properly attached safety line.

7.

If the work is to be performed at a high level above the bottom of the vessel, such as cyclone inspection, scaffolding and support flooring must be built to prevent falls.

8.

There should be a minimum of two backup men at the vessel manway in continuous surveillance of the personnel in the vessel.


9.

There should be spare pressure demand respirators, complete with their own separate air supplies, to allow backup personnel to enter the vessel quickly in case of an emergency. This spare equipment must be compact enough to allow the users to enter through the manway while wearing the equipment.

10.

It is recommended that any personnel working in a vessel which has an inert or contaminated atmosphere not be permitted to move too far away from the entryway, or into any tight areas, such as through a fractionator tray manway. If the person should develop some difficulty in an inaccessible area to a point where he could not function properly or lost consciousness, it would be extremely difficult for the surveillance team to assist or move the person by use of his safety line.

C.

HIGH TEMPERATURE PRECAUTIONS

A hazard that is commonly encountered on an RCC Unit is the rapid heating of unloading lines and the catalyst storage hopper when catalyst is unloaded from the regenerator. The unloading lines and hopper will heat up very quickly from their normal ambient temperature. Anyone working around the uninsulated lines or vessels should be warned that these will get hot. Combustible materials and trash should be kept away from the unloading lines and equilibrium catalyst hoppers.

D.

WATER VAPORIZATION HAZARD

The danger of vaporizing water in a liquid-full vessel exists anywhere in the refinery where a heating source exists and a water pocket can form. In the FCC unit, this is most likely to occur in the main column bottoms circuit. Anytime a vessel such as a slurry settler, or a bottoms coke strainer or filter is opened for maintenance, particular care must be paid to removing all free water and slowly heating the vessel back up to operating temperature. This is true for the bottoms heat exchangers as well. If water is heated in a hydrostatically full vessel, it will follow its vapor liquid equilibrium curve and begin increasing in pressure. If the vessel design pressure is


exceeded and the vessel cracks and relieves the internal pressure, the superheated water will expand approximately 1600 times in volume, resulting in a catastrophic failure of the vessel. This is similar to a classic boiler explosion mechanism. Operating and maintenance procedures must be written and followed to prevent equipment water pockets and rapid heating.

E.

CHEMICAL HAZARDS

The hydrocarbons, catalyst, and chemicals encountered in an RCC Unit are not hazardous as long as they are handled according to safe refinery practices. If they are mishandled, or allowed to escape into the atmosphere, some of them may be a hazard to the health of anyone in the area. It is recommended that the refiner determine the allowable safe limits according to laws, rules, and regulations issued by various agencies which apply to the material in question. It is suggested that methods ASTM D-170, ASTM E-300, and UOP 516 be followed regarding sampling techniques. Hazardous materials present in the RCC Unit, and their emergency treatments, include: 1.

Catalyst

Fluid cracking catalyst is a fine dust capable of causing eye and lung irritation. When unloading catalyst cars, drawing samples, etc., goggles or a face shield should be worn. If the circumstances are such that the working atmosphere becomes dusty, a dust mask should also be worn. The Material Safety Data Sheet provided by the catalyst manufacturer needs to be consulted for additional detailed information. When drawing hot regenerated catalyst samples, gloves and long sleeved clothing should always be worn. In case of hot catalyst leaks, care must be exercised to avoid the contact with hot catalyst on unprotected skin.


Loose catalyst is a relatively poor conductor of heat. While the surface of a pile of catalyst may be cool, the interior may be very hot. Personnel should not walk over or work around piles of spilled catalyst. 2.

Iron Sulfide

Iron sulfide is a black or gray powder deposit found in vessels where sulfur corrosion has occurred. It is easily mistaken for coke. The danger of iron sulfide lies in its "pyrophoric" properties. It will ignite spontaneously when exposed to air, which is most likely to occur in recently opened refinery vessels. A vessel suspected of containing iron sulfide must be thoroughly steamed out and washed with water before air is permitted to enter. Any suspected deposits must be kept wet until they can be removed and disposed of properly.

3.

Heavy Cracked Hydrocarbons

Heavy cracked oils are skin irritants. When drawing samples, draining equipment, and cleaning vessels or exchangers, suitable care should be exercised to avoid skin contact. In case of actual contact, the skin should be thoroughly washed with hot soapy water, and any oil-saturated clothing should be removed. Should any hydrocarbon enter the eye, the recommended First Aid is to wash with a copious amount of clean water and obtain trained medical assistance as quickly as possible. 4.

Light Cracked Hydrocarbon Liquids

Hydrocarbons in the gasoline boiling range will remove the natural oil from the skin, leaving it unprotected and subject to irritation or infection. Those gasoline streams which contain aromatics will be particularly dangerous. Benzene is a poison, and heavier aromatics have a narcotic effect.


In case of exposure, no time should be lost in removing any gasoline-soaked clothing and washing the skin with hot soapy water. First Aid for eye injuries is the same as that discussed above under "Heavy Cracked Hydrocarbons." 5.

Aromatic Hydrocarbons

a.

Benzene2

The most toxic hydrocarbon present in the unit is benzene. The danger in exposure to benzene lies in its carcinogenic effect on the body's blood-forming organs, an effect which is cumulative with each exposure. This aromatic component is contained in the gasoline and heavy naphtha streams in the unit. If clothing, including gloves, becomes wet from benzene, immediately remove the clothing. Wash the skin areas exposed to benzene with soap and water. Take a complete bath if the body area is wetted with benzene. Do not wear clothing that has been wet with benzene until the garment has been washed and dried. Wearing clothing that has been wet with benzene almost assures that the person will inhale benzene vapors over a long period of time with serious hazard to health. Avoid draining benzene on the ground or into the sewers where it can vaporize and create a health hazard. If benzene is accidentally spilled, flush it from the area and the sewer catch basin with large quantities of cold water. Do not use hot water or steam which further vaporizes the benzene. If you must enter an area of high benzene vapor concentration resulting from a spill, wear a pressure demand respirator.

2

For detailed information to take on exposure and potential hazards, consult Chapter XVII of OHSA Publication 1910.1028, Appendix A.


Though not specifically a health hazard, a problem resulting from benzene entering the sewer is that benzene is much more soluble in water than any other hydrocarbons. This places an extra load on the effluent treating system. b.

Toluene, Xylenes, and Heavier Aromatics3

These aromatic compounds are present in the gasoline and heavy naphtha streams in the unit. These compounds are only mildly toxic and do not have the destructive effect on the blood-forming organs as does benzene. Their principal effect is skin, eye, and respiratory irritation. If clothing becomes wet with such aromatics, remove the clothing, bathe, and put on fresh clothing. Avoid breathing aromatic vapors. All employees should be alerted as to the early signs and symptoms of excessive absorption of aromatics, and all workers should report such symptoms to the Medical Department. In addition, all employees should, of course, be advised of the hazards involved and precautions to be taken when working with aromatics.

6.

Light Hydrocarbon Vapors

The inhalation of any light hydrocarbon vapor should be avoided. Such vapors can be toxic, since they may contain aromatics, H2S, or other lethal compounds. A person who has breathed quantities of hydrocarbon vapors should be removed from the area and kept warm and quiet. If necessary, artificial respiration with or without the use of oxygen should be administered and medical aid summoned. Professional medical attention should be obtained at once.

3

See attached sheets for further hazard information.


7.

Hydrogen Sulfide (H2S)3

Hydrogen sulfide is present in the gases produced by the cracking of hydrocarbons containing sulfur. It will occur in the overhead receiver gas and will be dissolved in the unstabilized gasoline. Hydrogen sulfide is one of the most poisonous gases known. Exposure to an atmosphere containing less than 0.1% H2S may be fatal in 30 minutes or less. At very low concentrations, hydrogen sulfide has the characteristic odor of rotten eggs, but at higher concentrations, or extended exposure at low concentrations, the sense of smell is paralyzed so that personnel may be unaware of its presence. Extreme care must be exercised when opening lines and equipment which have contained even low concentrations of H2S, and an H2S detector should be used. When drawing samples, venting instruments, bleeding pumps, etc., precautions should be taken to avoid breathing the vapors. A person exposed to H2S may become excited or dizzy, may stagger, and can ultimately lose consciousness. First Aid consists of removal from the area and the administration of artificial respiration with or without oxygen if breathing has stopped. The patient should be kept warm and medical aid summoned.


8.

Flue Gas

Flue gas from the regenerator contains little or no oxygen. Asphyxiation can result if a person enters an improperly ventilated duct or a low area where the high density of flue gas will cause it to accumulate. Asphyxiation may be preceded by symptoms of dizziness, headache, or shortness of breath. RCC flue gas is very dangerous since it contains carbon monoxide, which is toxic. A concentration of 0.4% can be fatal in about one hour. One visible symptom of carbon monoxide poisoning is a bluish-red color of the skin. First Aid in cases of flue gas asphyxiation or poisoning consists of keeping the victim warm and administering artificial respiration and oxygen, if necessary, obtain professional medical attention immediately.


HYDROGEN SULFIDE EXPOSURE CALL FOR MEDICAL AID. VAPOR POISONOUS IF INHALED. Irritating to eyes. Move to fresh air. If breathing has stopped, give artificial respiration. If breathing is difficult, give oxygen. IF IN EYES, hold eyelids open and flush with plenty of water. HEALTH HAZARDS PERSONAL PROTECTIVE EQUIPMENT: Rubber-framed goggles; approved respiratory protection. SYMPTOMS FOLLOWING EXPOSURE: Irritation of eyes, nose and throat. If high concentrations are inhaled, hyperpnea and respiratory paralysis may occur. Very high concentrations may produce pulmonary edema. TREATMENT FOR EXPOSURE (THRESHOLD LIMIT VALUE): 10 ppm SHORT-TERM INHALATION LIMITS: 200 ppm for 10 min., and 50 ppm for 60 min. TOXICITY BY INGESTION: Hydrogen sulfide is present as a gas at room temperature, so ingestion not likely. LATE TOXICITY: Data not available. VAPOR (GAS) IRRITANT CHARACTERISTICS: Vapor is moderately irritating such that personnel will not usually tolerate moderate or high vapor concentration.


LIQUID OR SOLID CHARACTERISTICS: Minimum hazard. If spilled on clothing and allowed to remain, may cause smarting and reddening of the skin. ODOR THRESHOLD: 0.0047 ppm.


TOLUENE EXPOSURE CALL FOR MEDICAL AID. VAPOR Irritating to eyes, nose and throat. If inhaled, will cause nausea, vomiting, headache, dizziness, difficult breathing, or loss of consciousness. Move to fresh air. If breathing has stopped, give artificial respiration. If breathing is difficult, give oxygen. LIQUID Irritating to skin and eyes. If swallowed, will cause nausea, vomiting or loss of consciousness. Remove contaminated clothing and shoes. Flush affected areas with plenty of water. IF IN EYES, hold eyelids open and flush with plenty of water. IF SWALLOWED and victim is CONSCIOUS, have victim drink water or milk. DO NOT INDUCE VOMITING. HEALTH HAZARDS PERSONAL PROTECTIVE EQUIPMENT: Air-supplied mask; goggles or face shield; plastic gloves.


SYMPTOMS FOLLOWING EXPOSURE: Vapors irritate eyes and upper respiratory tract; cause dizziness, headache, anesthesia, respiratory arrest. Liquid irritates eyes and causes drying of skin. If aspirated, causes coughing, gagging, distress, and rapidly developing pulmonary edema. If ingested causes vomiting, griping, diarrhea, depressed respiration. TREATMENT FOR EXPOSURE: INHALATION: remove to fresh air, give artificial respiration and oxygen if needed; call a doctor. INGESTION: do NOT induce vomiting. Call a doctor. EYES: flush with water for at least 15 min. SKIN: wipe off, wash with soap and water. TOXICITY BY INHALATION (THRESHOLD LIMIT VALUE): 100 ppm. SHORT-TERM INHALATION LIMITS: 600 ppm for 30 min. TOXICITY BY INGESTION: Grade 2; LD50 0.5 to 5 g/kg LATE TOXICITY: Kidney and liver damage may follow ingestion. VAPOR (GAS) IRRITANT CHARACTERISTICS: Vapors cause a slight smarting of the eyes or respiratory system if present in high concentrations. The effect is temporary. LIQUID OR SOLID CHARACTERISTICS: Minimum hazard. If spilled on clothing and allowed to remain, may cause smarting and reddening of the skin. ODOR THRESHOLD: 0.17 ppm.


XYLENES EXPOSURE CALL FOR MEDICAL AID. VAPOR Irritating to eyes, nose and throat. If inhaled, will cause headache, difficult breathing, or loss of consciousness. Move to fresh air. If breathing has stopped, give artificial respiration. If breathing is difficult, give oxygen. LIQUID Irritating to skin and eyes. If swallowed, will cause nausea, vomiting or loss of consciousness. Remove contaminated clothing and shoes. Flush affected areas with plenty of water. IF IN EYES, hold eyelids open and flush with plenty of water. IF SWALLOWED and victim is CONSCIOUS, have victim drink water or milk. DO NOT INDUCE VOMITING. HEALTH HAZARDS PERSONAL PROTECTIVE EQUIPMENT: Approved canister or air-supplied mask; goggles or face shield; plastic gloves and boots. SYMPTOMS FOLLOWING EXPOSURE: Vapors cause headache and dizziness. Liquid irritates eyes and skin. If taken into lungs, causes severe coughing, distress, and rapidly developing pulmonary edema. If ingested, causes nausea, vomiting, cramps, headache, and coma; can be fatal. Kidney and liver damage can occur.


TREATMENT FOR EXPOSURE: INHALATION: remove to fresh air, administer artificial respiration and oxygen if required; call a doctor. INGESTION: do NOT induce vomiting; call a doctor. EYES: flush with water for at least 15 min. SKIN: wipe off, wash with soap and water. TOXICITY BY INHALATION (THRESHOLD LIMIT VALUE): 100 ppm. SHORT-TERM INHALATION LIMITS: 300 ppm for 30 min. TOXICITY BY INGESTION: Grade 3; LD50 50 to 500 g/kg LATE TOXICITY: Kidney and liver damage. VAPOR (GAS) IRRITANT CHARACTERISTICS: Vapors cause a slight smarting of the eyes or respiratory system if present in high concentrations. The effect is temporary. LIQUID OR SOLID CHARACTERISTICS: Minimum hazard. If spilled on clothing and allowed to remain, may cause smarting and reddening of the skin. ODOR THRESHOLD: 0.05 ppm.

157048 Environmental Page 1

ENVIRONMENTAL Introduction The refiner today is facing ever tighter environmental regulations. This section will discuss some of the normal FCCU pollutants and potential methods for their reduction. It is assumed that each refiner will be familiar with the restrictions placed on him by the appropriate authorities, as these widely varying and ever-changing rules are beyond the scope of this book. There are four primary sources of emissions from the FCCU. These are: •

Regenerator Flue Gas

•

Sour Water

•

Main Column Bottoms Catalyst Fines

•

Fired Heater Stack Gas

The major source is the regenerator flue gas, it will be discussed first. Regenerator Flue Gas The flue gas flow rate is about 105 - 110% (wt) of the inlet air rate on a dry basis. The composition of the gas will vary with the mode of regeneration, i.e., partial or complete combustion, and with other factors such as feedstock composition. Typical uncontrolled emission concentrations are shown in Table 1. Normally, the breakdown of the flue gas will be 75-80 vol% N2 + Argon, 15-22 vol% CO + CO2 and 8 -12% water vapor.


TABLE 1 FCCU REGENERATOR UNCONTROLLED FLUE GAS EMISSIONS FLUE GAS CONCENTRATIONS

Species CO Particulates (note 2)

Bubbling Bed Partial Combustion

Bubbling Bed Complete Combustion

High Efficiency Regenerator Complete Combustion

90,000 ppm

<500 ppm

<100 ppm

20-30 lb/mm SCF 350-450 mg/Nm3

15-30 lb/mm SCF 250-450 mg/Nm3

25-50%

20-50%

Opacity (note 2)

SOx

200-2000 ppm

(note 3)

Hydrocarbons

<200 ppm

<10 ppm

<10 ppm

NH3

<200 ppm

<10 ppm

<10 ppm

NOx

50-100 ppm

150-350 ppm

<60 ppm

Notes:

1. 2. 3.

All ppm concentrations are by volume No external particle removal such as ESP or WGS No SOx reduction such as catalyst additives or WGS


CO, Hydrocarbons, and NH3 The amount of hydrocarbons and NH3 present will depend primarily on the feedstock characteristics and operating severity. Hydrocarbon and NH3 are normally present in only trace quantities, while CO and CO2 from coke combustion are major constituents of the flue gas. The disposal of these pollutants can best be handled by burning them, either in a complete combustion regenerator or with a CO boiler. In most cases complete combustion is the favored method for a variety of economic and process reasons. Complete combustion will convert most CO and hydrocarbons to CO2 and water vapor, while the higher oxygen atmosphere of the complete combustion unit decreases the amount of ammonia formed. This observation is based on commercial plant experience; the mechanism of ammonia formation has not been definitely proven. Particulates The particulates from the FCCU regenerator cyclones are primarily catalyst fines of less than 40 microns and the particulate loading is in the range of 15-30 lb/MMSCF (250-450 mg/Nm3). In addition to the catalyst fines, some condensables such as sulfates will be present if the sampling temperature is low enough. In the USA, the temperature specified by EPA Method #5 is 248°F + 25°F (120°C). At this temperature some sulfur compounds and hydrocarbons will condense, which may cause a higher particulate measurement. In the United States the particulate level in the flue gas is typically limited to 1 lb of solids per 1000 lb of coke burned or approximately 4.7-6.5 lb/MMSCF of flue gas 3 3 (80-110 mg solids/Nm ); the European regulation varies from 80-500 mg/Nm . Future regulations may reduce this value to 50 mg/Nm3. These environmental regulations typically require an electrostatic precipitator (ESP) or wet gas scrubber (WGS) to bring particulate emissions down to an acceptable level. The precipitator is the less expensive of the two options in many cases, depending on flow rates, pressures, temperatures, and particulate loadings, but


neither one is inexpensive to construct or operate. Disposal of the collected wastes can be difficult. The precipitator yields dry catalyst fines which can be used for landfill or as a raw material for cement. The scrubber generates a waste liquid or slurry stream high in fines and solids, which must be further treated. The removal of SOx is an added advantage for the scrubber system, with efficiencies of up to 95% claimed by some manufacturers. Improvements in third stage separator (TSS) technology which uses cyclonic separators external to the regenerator have improved efficiency to the point where tit may be considered a less expensive alternative capable of meeting environmental regulations. Historically though, TSS use has been limited primarily to protection of power recovery equipment and did not eliminate the need for the ESP or WGS. Opacity Opacity is the quality or state of a substance which renders that substance impervious to rays of light. Opaque stack emissions that block all light would have an opacity of 100%, while clear emissions that do not attenuate light have an opacity of zero. Another scale which is sometimes used for gray or black emissions is the Ringelmann Number, going from 0 (clear) to 5 (opaque). A Ringelmann Number of 1.5 would correspond to an opacity of 30%. A high opacity for an FCCU regenerator stack would be caused by: 1.

Catalyst fines (either greater content or shift to smaller PSD)

2.

Unburned hydrocarbons

3.

Condensibles such as SOx and NH3

4.

Water vapor


Each of these would have a separate solution, and would normally be accompanied by other problems. Large amounts of catalyst fines from the stack could indicate excessive attrition or poor cyclone performance. If the plume were caused by unburned hydrocarbons, this would indicate poor regenerator operation. Control of condensables such as SOx are discussed elsewhere in this section. Water vapor is sometimes mistaken for catalyst fines. This vapor usually does not cause problems meeting environmental regulations and it should not be excessive; large increases in water output should be investigated if the water vapor rate is excessive. Sulfur Oxides The flue gas sulfur oxides are formed when sulfur in the coke is oxidized to SOx in the regenerator. These oxides are primarily SO2 (~90%), with lesser amounts of SO3 (~10%); the total SOx increases as feed sulfur increases and depends on the type of compounds containing the sulfur. There are basically two methods of reducing SOx besides feed pretreatment: using a catalyst additive, i.e. a “SOx control catalyst" and wet gas scrubbing (dry scrubbing with limestone has also been used, but it is impractical for most refiners). The SOx control catalyst additive is injected independent of the FCC catalyst (a small injection device is required) and is typically 3-10% of the catalyst inventory. The additive adsorbs SOx in an oxidizing environment (regenerator) and liberates sulfur as H2S in a reducing environment (reactor). SOx control additives can typically remove 10-20 lb SOx/lb additive down to a limit of ~300 vppm in the flue gas. This process relies on oxidizing the SO2 to SO3 and is therefore typically more effective in a complete combustion unit with excess oxygen present. UOP has acquired the right to license the Exxon Wet Gas Scrubbing (WGS) process (see Figure 1) to UOP-licensed FCC or RFCC units. The WGS process removes both SOx and particulates from the FCC or RFCC flue gas. The WGS uses a venturi device which provides intimate contacting of the flue gas with a mildly alkaline solution. The system can also tolerate large solids carryover from the regenerator if an upset should occur. The catalyst fines are removed as moist filter


cakes and the water effluent contains dissolved salts (sodium sulfate). Wet gas scrubbing can remove up to 90% of the SOx from the flue gas down to <50 vppm. Precise economic comparisons between using a SOx control catalyst and the WGS process are difficult because the value assigned to particulates removal has a major effect on the selection. In general, the SOx control catalyst will be more attractive at low-to-moderate amounts of SOx removal (below 1000 vol ppm) and the WGS is more attractive for removing large quantities of SOx (above 1500 vol ppm) or for meeting very low SOx emissions (<200 vppm). Between 1000 and 1500 ppm of SOx removed, the choice depends on a number of factors including variability of installed cost, presence of existing particulate removal equipment, local environmental laws, utility and chemical costs and relative amount of SOx removal. Nitrogen Oxides The amount of NOx emitted from the regenerator is highly dependent upon several variables such as mode of operation, regenerator style, excess O2, promoter concentration, coke distribution, feed nitrogen, and dense bed temperature. Studies have shown that about half of the feed nitrogen goes to coke on catalyst, but only about 10% of the nitrogen in coke goes to NOx. It is believed that the NOx emission is limited by the reaction of CO + NO to form N2 and CO2. Thus, when Pt CO promoter is added to a unit, CO is converted to CO2 so quickly that there is less CO available to react with the NOx and NOx emissions increase. In a partial burn operation, there is always CO available to react with the NOx formed so emissions are lower. Good coke and air distribution is important so that concentrations of CO, O2 and NOx are evenly distributed throughout the regenerator. There are fundamental differences in the operation of a bubbling bed operating in full combustion with CO promoter and a high efficiency regenerator operating in full combustion without the need for CO promoter. Bubbling Bed Regenerator The bubbling bed type regenerator burns coke from spent FCC catalyst in a backmixed dense phase fluidized bed. The regenerator consists of a closed cylindrical pressure vessel sized to contain a dense phase fluidized bed of catalyst in the bottom of the vessel and multiple sets of cyclone separators within the dilute phase existing in the upper portion of the vessel.


Spent catalyst enters the regenerator at the side of the vessel at the surface of the fluidized dense phase of catalyst. Air for coke combustion is distributed uniformly across the cross-section of the vessel near the bottom of the fluidized bed through an air distributor. The carbon burning reactions proceed largely in series, with coke burning to CO first and then CO burning to CO2. Since the fluidized bed is substantially back-mixed, there is a rather wide distribution of residence times for both the combustion reactants and products. Further, since fluidized dense beds mix better vertically than laterally, it is difficult to mix reactants (air and coke) uniformly across the cross-section of the vessel. This leads to non-uniform burning profiles and requires longer average residence times, higher quantities of excess oxygen and higher levels of Pt combustion promoter to assure that the catalyst is burned clean (regenerated fully) and that CO is fully converted to CO2 to avoid exceeding CO emission limits. This combination of factors, i.e. long residence times, non-uniform burning profiles, higher required levels of excess oxygen (~2 mol%), and higher required levels of Pt combustion promoter, result in this style of regenerator typically producing on the order of 3 to 4 times the NOx emissions relative to the equivalent High Efficiency Combustor style regenerator. Combustor Style Regenerator The High Efficiency Combustor style regenerator burns coke from the spent catalyst in a quasi-plug flow fast fluidization burning zone. The High Efficiency style regenerator is divided into two separate zones. The lower section is the combustor zone where the coke burning occurs. The upper section of the regenerator (second zone) serves to hold a reservoir of regenerated catalyst and also contains multiple sets of cyclones in the dilute phase of the upper regenerator. Very little, if any, coke burning occurs in this upper portion of the regenerator vessel. Spent catalyst (carrying the coke), air and a substantial quantity of hot clean regenerated catalyst are mixed together in the bottom of the combustor. The combustor vessel is sized so that it operates in the velocity and density regime characterized as fast fluidization. This permits quasi-plug flow transport of material from the bottom of the combustor vessel upward and out to the upper regenerator through a vapor/catalyst disengaging device. The moderate operating density of the combustor (burn zone) permits rapid and uniform mixing of material entering the bottom of the combustor. The recycle of hot regenerated catalyst permits a degree


of control over residence time and initial rates of coke combustion. The quasi-plug flow behavior of the fast fluidized bed assures relatively narrow and controlled residence time distributions. Overall, the High Efficiency Combustor style of regenerator provides better control of the burning zone. The results of the better (or more efficient) control of the burn zone in the Combustor style regenerator are that:  



Oxygen is used more efficiently so that lower levels of excess oxygen are required to completely burn the coke to CO2 while minimizing CO emissions Generally Pt combustion promoter is not needed (at design rates) to accelerate the burn to completion because hot recycled catalyst is used to increase the burning rate by preheating the combustion reactants (air and coke on spent catalyst) The preheating effect of recycling hot regenerated catalyst, combined with the efficient mixing and uniform residence time distribution in the combustor permits the time spent in the combustor to be minimized

The net result of these combustion characteristics of the Combustor style regenerator is that substantially lower levels of NOx are produced.

Plant Upsets The values given for pollutant concentration in Table 1 are for normal plant operation. During upsets these numbers may be grossly exceeded. An example of this would be an oil reversal into the regenerator. Massive amounts of hydrocarbons may be emitted. While the FCCU has been designed for safe operation, something as simple as a sticky slide valve may thwart initial corrective action. The best way to minimize these upsets is careful attention to the unit, with well trained operators that understand what action should be taken, and why it is done.


Main Column Overhead Receiver Sour Water Wash water is used in the FCCU to remove ammonia, cyanides, and some sulfur compounds which can cause corrosion and fouling in the unit. The water is injected into the interstage cooler of the wet gas compressor at a rate of ~2 GPM/1000 BPD (5-7 vol% of fresh feed). The water goes through the coolers and is pumped from the interstage receiver to the high pressure separator. From here it circulates back to wash the main column overhead condensers, and leaves the plant from the overhead receiver water boot. In many units stripping, lift and feed atomizing steam in the reactor will double the amount of water drained over the amount injected in the gas concentration unit. The concentrations of the various contaminants in the water will depend on the feed concentrations and the total sour water rates. For a 2 GPM/1000 BPD water injection rate, the concentrations shown in Table 2 would be expected for a unit with a feed sulfur of 1-2% and less than 1000 ppm of feed nitrogen. TABLE 2 FCC SOUR WATER CONCENTRATION IN MC OVHD WATER

Pollutant

Mode of Regenerator Operation Partial Complete Combustion Combustion

Sulfide, ppm Ammonia, ppm Cyanide, ppm Phenols, ppm Hydrocarbons, ppm

3000 – 4000 1000 – 2000 40 – 150 100 – 300 100 – 2000

3000 – 4000 1000 – 2000 30 – 50 200 – 600 100 – 2000

All values are ppm by weight. The pH of the water from the overhead receiver should be in the basic range, pH = 8-9.


Some refiners do not show pollutant levels as high as the values given in Table 2, especially if the feed sulfur or nitrogen is low. The difference between the partial and complete combustion mode of regenerator operation would involve oxygen carryover with the regenerated catalyst, different oxidation levels in the regenerator and other factors. The normal disposal method for the sour water is a sour water stripper. The overhead vapor stream from the stripper is normally sent to a Claustype unit, although in some cases it may be simply burned. The stripper bottoms is sent to waste water treatment; it has also been used for the crude unit desalter. MCB Catalyst Fines Catalyst fines leaving the reactor with the hydrocarbon product are concentrated in the bottom of the main column and leave the unit with the bottoms product. Some of these fines settle out in the product tank so that the tank needs to be cleaned occasionally. The oil soaked fines removed from the tanks are considered hazardous waste and must be disposed of properly. Catalyst fines in the bottoms product can also cause problems with heaters and boilers firing heavy fuel oil. Because of the high cost of tank cleaning and problems with downstream heaters catalyst fines removal technology is being added or considered by many refiners. UOP, through an alliance with Pall Corporation, offers filtration technology capable of reducing the fines in the bottoms product to less than 50 wppm. Cyclonic separation devices and slurry settlers are also in use but are less effective.

Fired Heater Stack Gas Any environmental problems with the FCC fired heaters would be the same as those encountered with the other refinery heaters. Efficient firing normally reduces CO content in the stack gas to a minimum. SOx and NOx can be minimized by hydrotreating the fuel. Low Nox burners are also an effective means of NOx reduction in fired heaters and boilers. Flue gas treating could also be used to reduce emissions. Proper furnace operation will minimize emissions problems with the stack gas.


Fugitive Hydrocarbon Emissions This category would cover hydrocarbon vapors from leaks, sampling or storage. Refineries have always tried to minimize these losses, because uncontrolled hydrocarbons are a fire and safety hazard, in addition to an economic loss. Closed sampling systems and controlled venting on storage tanks will probably become more common in the future.

RFCC Process Technology Manual

Recommend Documents