process-simulation-of-solvent-deasphalting-plants-with-proii.pdf

Process Simulation of Solvent Deasphalting Plants with PRO/II based on Thermodynamic Equilibrium Data "VLE K -Values -Values - Fill Options"

Moscow, 28-30 March 2012

Introduction

SDA, the Solvent Deasphalting technology is actually one of the most interesting „Bottom to Barrel “ technology for heavy residues in modern refineries

? • Large price differences between light sweet crudes and heavy sour crudes created strong incentives for bottom processing tecchnologies • Old fashion technology SDA in the lube oil refinery will be more and more also applied in fuel oil refineries, too • The SDA technology technology is the lowest investment cost variant compared to Gasification, Visbreaker, Visbreaker, Hydrocracker H ydrocracker,, Coker et al • In the best case both products DAO as well as Pitch can be converted in further process steps to valuble products, no further f urther residues will be available

2EDL_SDA_PRO/II_Fill-Options

Integration of SDA in modern Refinery Bright Stock

Lube Oil Refinery

FCC -Plant Crude

Fuel Oil Refinery

Hydrocracker

ADU DAO

SDA

VDU AR

Asphalt (Pitch) Bitumen plant ADU…Atmospheric Distillation Unit VDU…Vacuum Distillation Unit SDA…Solvent Deasphalting AR…..Atmospheric Residue VR…..Vacuum Residue


VR

Typical SDA

a typical SDA plant is a PDA, e.g. - Propane Deasphalting Plant The plant normally consists of the following parts:

• Propane Extraction • DAO Train • Asphalt Train • Propane Condensation and Recovery • Pressure Relieve -, Slop - and Flushing system


SDA simulation and design

The simulation and design of a SDA , especially the liquid –liqud extraction process is a challenging task !

? • the feed, Vacuum Residue can only be simulated with pseudo components e.g. it is impossible to use components from the data base with a defined structure • the pseudo components have no structure to produce BIPs from the UNIFAC –Model to build activity coefficients between the components • the liquid –liquid extraction process for the SDA is based on the different solubilities between the solvent and the hydrocarbons • BIPs are necessary to calculate the equilibria in the two liquid phases, the DAO and the Asphalt phase BIP… Binary Interaction Parameters 5EDL_SDA_PRO/II_Fill-Options

Model Basis: „Chueh and Prausnitz“ The BIPs can be calculated based on a publication from Chueh and Prausnitz (1967) for an EOS (Equation of State) via the partial molar volumes of the multicomponent liquid mixture: Equation of State: P 

RT   b



a T 0.5 (  b)

Partial differentiation of EOS: RT

(

 P 

)T c  0 and (

2

 P 

2

)T c  0

 k



  b

bk

(1 

  b

2( )

RT (  b) 2

with the mixing rules for a mixture and used for

a, b, aii , bi , aij ........ f ( xi ,  i , ci , P ci , T ci , ai , bi ) 6EDL_SDA_PRO/II_Fill-Options



 x a i

ki

)  abk /(  b)

i

 (  b)T

1/ 2

a  T 1/ 2

  2 (  b) 2    2  b

a  0.4278 and b  0.0867

Model Basis: „Chueh and Prausnitz“ applying the pseudocritical rules to the critical region:

T R  0,93

T 'CM  T CM  (T CT  T CM ) D(T R ) '

 CM

  CM  ( CT   CM ) D(T R )

Further, with adjustments to SRK and PR applied in PRO/II we get the BIP„s between the components:

   k ij  1   ' 1/ 3 1/ 3 '  ( CMi   CMj ) / 2  1/ 3 ' 1/ 3 '  CMi  CMj

n3


n

(the best fit with experimental data)

Notation a, b.......constants in Redlich Kwong EOS D(TR )...correction function to the critical region k ij .........BIP's for interaction parameters P...........total pressure T...........temperature TR ..........reduced temperature V......... ..total volume of liquid mixture  ............molar

volume of liquid or liquid mixture

 k .......... partial molar

volu me of component k in liquid phase

 CM ........ pseudocritical

volume of mixture

'

 CM .......corrected pseudocritical  CT .........true critical

volume of mixture

volume of mixture

x......... ....mole ffraction in liquid phase

 a , b ....dimensionl ess parameter in Redlich Kwong EOS  ............acentric factor 8EDL_SDA_PRO/II_Fill-Options

PRO/II PFD- one stage Typical one stage liquid –liquid extraction unit:


Pseudo component feed Vacuum Residue (VR) d 15  0.9669 kg/kg


Pseudo component properties


PRO/II- K-value – Fill Options


EOS for „Fill –Options“ The following EOS are awailable for the Fill –Option“: - Modified Chueh Prausnitz Hydrocarbon Fill • Soave Redlich Kwong • SRK –Panagiotopoulos-Reid • SRK –Modified Panagiotopoulos-Reid • SRK –SIMSCI • Peng –Robinson • PR –Panagiotopoulos-Reid • PR –Modified Panagiotopoulos-Reid • Predictive Peng Robinson 78

• Tacite The best fit with practical results we achieved with the SRK –Modified Panagiotopoulos Reid thermodynamic !


BIPs calculation BIPs calculated from the output report:


Stream list

DAO: 28.5 %, Asphalt : 71.5 % 15EDL_SDA_PRO/II_Fill-Options

VR-DAO-Asphalt TBP


VR-DAO-Asphalt distribution


Property list As shown in the property list special data as S, V, Ni, Fe, N and CCR can be controlled in the DAO, Asphalt via a distribution in the f eed (VR):


Bench scale tests The theoretical results were compared to bench scale tests in a autoclave:

Autoclave

Conditions: subcritical pressure from 30 to 35 bar and temperature about 50°C Solvent: Propane (100%) 19EDL_SDA_PRO/II_Fill-Options

Bench scale tests Results: DAO and Asphalt from the bench scale unit Asphalt

DAO DAO…..Deasphalted oil AS……..Asphalt (Pitch)

*…SRK -Mod Panag. -Reid **…PR -Mod Panag. -Reid

The SRK Modified Panagiotopoulos -Reid method gives the best fit to the practical results !


Study of thermodynamic models


Solvent impact

l i O e d u r C

Oils

Resins

e u d i s e R c i r e h p s o m A

e u d i s e R m u u c a V

e t a t i p i c e r P e n a p o r P

e t a t i p i c e r P e n a t n e P

e t a t i p i c e r P e n a t p e H

Asphaltenes

The choise of the solvent or solvent mixture is very important ! 22EDL_SDA_PRO/II_Fill-Options

Simulation studies of Solvent mixtures


Ternary plots -Temperature influence-

Binodal curves

Ternary plot from literature:

Solvent to Feed ratio (tie lines): 1…0,5 2…3,0 3…8,0 1


2

3

Ternary plot – Pressure influence -

Binodal curves Solvent to Feed ratio (tie lines): 1…0,5 2…3,0 3…8,0

1


2

3

PRO/II simulation of multiple stage extraction

Heater on stage 1 to precipitate Asphalt Controller for precipitation


Ternary plot – Pressure influence The number of trays are not so important !

For the design the HTU, NTU concept must be applied ! HTU…Height of Transfer Unit NTU…Number of Transfer Unit


Summary

The Solvent Deasphalting (SDA) is a key technology processing heavy residues to valuable oils (DAO, Deasphalted Oil) and asphalt (Pitch) for the Bitumen production. The process will be more and more interesting also in fuel oil refineries as a “Bottom to Barrel” technology with low costs. The calculation of the extraction process with NPB components is difficult because of the lack of BIPs between the components. Other ways to build up the data from the UNIFAC model are not applicable. The model from CHUEH and PAUSNITZ, published in the AIChE Houston Meeting in 1967 is applicable to estimate the BIPs between the NBP components and the solvents, finally to calculate the extraction of the SDA for high pressure.


Summary The model application in the PRO/II simulation program under the Equation of State (EOS) with the VLE K values –“Fill Option” is the basis for the complete process simulation of the whole SDA process.

Experimental results have shown that the model can be applied to design the SDA process with different solvents (Propane, Butane etc.) and solvent mixtures for the sub- and supercritical extraction processes.

Further investigation should be done to apply the model also to other extraction technologies e.g. solvent extraction with furfural or NMP.

Thank you for your attention !


process-simulation-of-solvent-deasphalting-plants-with-proii.pdf

Recommend Documents