COLOSO, Ma. Jinelle V.
ChE511-Equipment Design
5ChE-B
November 5, 2017
SIEVE TRAY DISTILLATION COLUMN
INTRODUCTION
Distillation is one the most common unit operations used in the industries. It is a process wherein the desired component is separated from the initial mixture by selective condensation or evaporation. This separation process could be in partial or complete depending on the type of distillation use [1]. One of the types of distillation and is one of the main process in the propylene glycol industry is the use of flash distillation, also known as equilibrium distillation. This happens when a definite fraction of the liquid feed is vaporized and is in equilibrium with the residual liquid. The process also involves reducing the pressure suddenly to lower the boiling point of the component (hence, it is called flash distillation) [2]. Sieve tray distillation column are only used for continuous/ flash distillation processes. Sieve trays are design to look like the usual sieves use in mechanical separations. The sieve trays have metal plates with holes wherein the vapor passes straight upward through the liquid on the plate. The effectiveness of the tray depends on the number and size of the holes and the arrangement of it [3].
Figure 1. Sieve tray mechanism
Compared to the other types of trays found in columns, sieve trays have the highest efficiency ranging from 0.7 to 0.9, needs low maintenance, low fouling tendency and has the lowest cost [4].
In the propylene glycol industry, a 99.6% final product purity of propylene glycol from glycerol, which is harvested directly from biodiesel productions, will be the main concerned. The crude glycerol that can be used as the feed, cannot be fed directly to the reactor because it contains chloride and sulfate salts. If they were to enter the reactor, these salts would cause corrosion. Therefore, removing these salts using a distillation column is a necessity [5].
The distillation column is designed to process the feed from the heat exchanger which contains 15,000 kg/hr crude glycerol, with a composition of 51.6% water and 1.7% methanol (Usual composition of crude glycerol found in the market) at its dew point. The objective of the equipment is to completely separate the methanol from the glycerol and reduce the water content to 20% of the original (bottoms: glycerol-20% of the water from the feed; distillate: methanol-80% of the water from the feed).
INPUT CONDITIONS
Crude glycerol is being processed in the column to recover pure glycerol in the bottoms. In this process, the following assumptions were made:
1. Since methanol has a lower boiling point than that of water, the two mixtures could be considered as one system. Once the water hits its boiling point, methanol had already reached its boiling point, therefore, complete separation of methanol had already occurred. 2. The efficiency was estimated using the O’Connell Equation. (Mc Cabe-Thiele method cannot be use because of the distance of the equilibrium curve from the diagonal) 3. The vapor feed enters at its dew point which is at 113.5 oC.
4. The crude glycerol contains 46.7%(w/w) glycerol, 51.6%(w/w) water and 1.7%(w/w) methanol 5. 15-20% by weight weight of the original water water content will be retained in the bottoms. This is the optimal concentration to be used in the trickle bed reactor due to viscosity concerns. 6. The vapor-liquid diagram with its data was obtained from www.vle-calc.com [Appendix A]. TEMPERATURE
x 1 (water)
y1 (water)
°C
mol/mol
mol/mol
299.075
0
0
273.803
0.01
0.416587
254.717
0.02
0.690761
239.457
0.03
0.8304
226.967
0.04
0.900621
207.635
0.06
0.959137
193.233
0.08
0.979931
181.988
0.1
0.988869
162.088
0.15
0.996434
148.903
0.2
0.998467
139.501
0.25
0.999209
132.51
0.3
0.999537
127.188
0.35
0.999701
123.089
0.4
0.999791
119.93
0.45
0.999844
117.514
0.5
0.999877
115.699
0.55
0.999898
114.377
0.6
0.999911
113.456
0.65
0.99992
112.854
0.7
0.999924
114.498
0.75
0.99991
114.498
0.8
0.99991
114.498
0.85
0.99991
111.87
0.9
0.999923
111.713
0.92
0.999924
111.502
0.94
0.999927
111.221
0.96
0.999937
111.046
0.97
0.999945
110.842
0.98
0.999957
110.603
0.99
0.999974
109.593
1
1
Phase diagram of Water-Glycerol Mixture
Figure 2 Phase Diagram of Water-Glycerol Mixture
FLOW DIAGRAM and MATERIAL BALANCES:
Material Balances: Overall Material Balance:
F=D+B
Solute Balance:
Fx f = = DxD + BxB
Other balances:
V=L+D L = L + qF V = V + (1-q)F
EQUIPMENT DESIGN THEORETICAL CALCULATIONS
Table 1 shows the composition of the feed at its dew point. This were the basis for all calculations. Table 1. Feed Composition Feed (kg/hr)
Molecular Weight
Feed (mol/hr)
% by mole
Glycerol
7,005
92
76.141
14.81
Water
7,740
18
430
83.64
Methanol
255
32
7.96875
1.55
The dew point of the feed was obtained using Figure 2 wherein the composition of the water (0.8364) was used.
Table 2 shows the composition of the distillate and the bottoms. Take note that the water in the bottoms satisfy the 15-20% by weight range. Table 2. Distillate and Bottoms Composition Distillate
Distillate (% by
Bottoms
Bottoms (%
(mol/hr)
mole)
(mol/hr)
by mole)
Glycerol
0
0
76.1413
46.96
Water
344
97.7
86
53.04
Methanol
7.96875
2.3
Trace
Trace
TOTAL
351.96875
162.1413
The flowrates of L and V in the rectifying section was also computed. The values obtained were 211.18125 mol/hr and 563.15 mol.hr, respectively. L and V was also computed. The values obtained were 211.18125 mol/hr and 49.0452 mol/hr, respectively (See Appendix B for the calculations of Table 2 and the flowrates) .
The minimum number of stages was calculated using the Fenske-Underwood Equation [4].
logg(()) = log∝g∝
Equation 1
Where: xD = mole fraction of light key component in the distillate (0.977) xB = mole fraction of light key component in the bottoms (0.5304) ά = relative volatility (37.60896) * See Appendix C for on how to get this. Substituting all values from Equation 1, the minimum number of stages calculated was 0.999999 or 1 stage. The minimum reflux ratio for the system was also calculated cal culated using Equation 2. This equation is use when feed is at the dew point [4].
1 = ∝−∝
Equation 2
Where: D = Distillate, mol/hr (351.96875 mol/hr) F = Feed, mol/hr (514.1098 mol/hr) ά = relative volatility (37.60896) The calculated minimum reflux ratio from the above equation was 0.5. As a rule of thumb, economically optimum reflux ratio is about 1.2 times the minimum. Therefore, the reflux ratio is 0.6. In getting the theoretical number of stages, the Erbar and Madd graph was used where [4]:
= + = 0.375;
= + = 0.333
Figure 3 Plates reflux correlation of Erbar and Madd
this to
=
Plotting the z and y-axis, a 0.45 value for the x-axis was obtained. Equating , the number of theoretical stages will be 2.22 or 3 stages.
For the tray efficiency, efficiency , the O’Connell correlation w as used [6].
Where:
= 51.4∝ −.
Equation 3
ά = relative volatility (37.60896) μ = viscosity of the feed (12.3578) [9]. Substituting the values to Equation 3, the efficiency was 12.828%. Using the efficiency, the actual number of trays can be calculated using the formula:
= ℎ The actual number of trays is 23.39 or 24 plates.
Equation 4
EQUIPMENT SPECIFICATIONS
All calculations calculations for the equipment design design is seen in Appendix D. The calculations calculations were based on Seader and Henley and Flooding considerations using the Handbook assuming the column is a plate column. Material of Construction
The material of construction to be used is A S TM A 285 G rade A . It is a type of carbon steel pressure vessel plate which are common in distillation columns. In general, carbon steels are most common than stainless steel unless there is something in ASME or the pressure directive forces the design to use a higher specification and more expensive steel.
SIEVE TRAY DISTILLATION COLUMN Function: To separate glycerol from water-methanol in order to purify reactor feed
stream. Operation: Continuous Design Data:
Number of Theoretical trays: 3
Molar Reflux Ratio: 0.6
Pressure of Column: 20 psia
Tray Spacing*: 24 in.
Design Pressure: 30psig
Design Temperature: 123.5 oC
Functional height: 18.859m.
Weir height*: 2in.
Tray holes*: 0.0127m.(diameter) 0.0127m.(diamet er)
Column diameter: 2.232m
Column thickness: 0.375in.
Tray efficiency: 12.828%
Tray Thickness*: 0.134in
Number of Actual trays: 24
Pressure drop*: 747 Pa
Downcomer clearance*: 1.5in.
Net Area: 3.899m 2
Flood velocity: 0.129 m/s
Weir length*: 1.674m
Hole pitch*: 1.1in
Liquid flow pattern: Single Pass
Active Area: 3.869m 2
Fractional Hole Area*: 0.387m 2
Number of Holes/tray: 128
Skirt height*: 4m
Vessel weight: 51437.3748 kg
RENDERED MODEL
Figure 4 Column Internals, Sieve trays
Figure 5. Cross sectional view of the column
Figure 6 Top view of the column
Figure 7 Front view of the column
CONCLUSION AND RECOMMENDATION RECOMMENDATION
A distillation distillation column was successfully successfully design to completely separate the watermethanol mixture from glycerol wherein the feed originally contains 83.64% by mole water. The desired product is at the bottoms which contains 18.1% of the original water and 100% recovery of glycerol. In order to successfully design a column, knowledge on heat and mass transfer must be mastered. Vacuum distillation is much more effective than the usual distillation for it separates the volatile compound more efficiently. efficient ly. Under vacuum conditions, a powerful pump should be considered in the process so as to maintain the vacuum. It is also recommended that before entering the distillation column, the feed should be heated further with medium pressure steam so that the duty requirement of the reboiler will be reduced.
REFERENCES
[1] Distillation. (2017, September 19). Retrieved September 22, 2017, from https://en.wikipedia.org/wiki/Distillation [2] Jaffar, N. (1970, January 01). Chemical Engineering Knowledge. Retrieved September 22, 2017, from http://chemknowhow.blogspot.com/2014/10/differentialdistillation-flash.html Sapp, K., Lawson, Lawson, J. (2007). Batch Distillation Theory. USA. University of Florida. Retrieved September 22, 2017, from http://www.che.ufl.edu/ren/course/4404L/BD/BD%20Theory%20Handout.pdf [3] Process & Operation Engineer, Egypt. Follow. (2015, February 25). Types and design of the towers trays. Retrieved November 03, 2017, from https://www.slideshare.net/MohamedSalah69/types-and-design-of-the-towers-trays (n.d.). Retrieved November 03, 2017, from http://www.separationprocesses.com/Operations/POT_Chp02b.htm [4] Brannan, C. (n.d.). Rule of Thumb for Chemical Engineers (3rd ed.). Gulf Professional Publishing. [5] Chatterjee, K., Hall, K., & Tell, S. (n.d.). Glycerol to Propylene Glycol [Abstract]. University of Pennsylvania: Scholarly Scholar ly Commons. Retrieved November 3, 2017, from http://repository.upenn.edu/cgi/viewcontent.cgi?article=1025&context=cbe_sdr [6] SALUNKE, D. B. (2006). AN O’CONNE O’CONNELL LL TYPE TYPE CORRELAT CORRELATION ION FOR PREDICTI PREDICTION ON OF OVERALL EFFICIENCY OF VALVE TRAY COLUMNS (Unpublished master's thesis). Pune University . McCabe –Thiele –Thiele method. (2017, September 03). Retrieved September 22, 2017, from https://en.wikipedia.org/wiki/McCabe%E2%80%93Thiele_method Kim, K., Diwekar, U. (2005, April 27). Chapter 5: Batch Distillation. USA. p.112-114. Wired Chemist. (n.d.). Retrieved September 25, 2017, from http://www.wiredchemist.com/chemistry/instructional/laboratory-tutorials/distillation Vogelpohl, A. (1974). The Fundamental Equation of Distillation. Germany. Technical University of Clausthal. What is the Difference Between Reflux & Distillation? (n.d.). Retrieved September 25, 2017, from http://sciencing.com/difference-between-reflux-distillation5953384.html
Albarede, F., Hofmann, A. A. W., & Condomines, M. (n.d.). The Rayleigh Rayleigh distillation equation. Geochemistry,220-221. doi:10.1017/cbo9781139165006.017 Th: Define and use K values, relative volatility, and x-y diagrams. (n.d.). Retrieved November 03, 2017, from http://pillars.che.pitt.edu/student/slide.cgi?course_id=12&slide_id=72.0 (n.d.). Retrieved November 03, 2017, from http://www.separationprocesses.com/Distillation/DT_Chp04p1.htm Douglas, James. Conceptual Design of Chemical Processes. McGraw-Hill, Inc., 1988 Towler, G., &Sinnot, R. (2008). Chemical Engineering Design.Elsevier. Perry, R. H. (2008). Perry's Chemical Engineers' Handbook (8th ed.). McGraw-Hill. Chuang, K.T., & NandaKumar, K. (2000). Tray Columns: Designs. University of Alberta. (n.d.). Retrieved November 03, 2017, from http://facstaff.cbu.edu/rprice/lectures/distill7.html#stage Feed Entry Pitfalls in Tray Towers. (2010). Distillation Troubleshooting, 97-110. doi:10.1002/9780471690726.ch5 (n.d.). Retrieved November 03, 2017, from http://seperationtechnology.com/distillation-column-tray-selection-1/
APPENDIX A
VLE-Calc is a free on-line application which provides both vapor-liquid, liquid-liquid equilibrium data of different compounds. This application also plots phase-diagrams and even can solve common distillation problems. CAS Registry Number is a Registered Trademark of the American Chemical Society. APPENDIX B
Distillate composition:
1 = 7740 7740 0.80 = 6192 18 255 1 32 = . 1 = 7740 7740 0.20 = 1548 18 7005 7005 921 = . = = 0.6 = = = . = Bottoms composition:
Calculating the L, V, L and V.
L = 211.18125 211. 18125mol/ mol/hr hr
q = 0; Dew point vapor feed
mol/hr
V = 49.0452mol/hr
V= 563.15mol/ 563. 15mol/hr hr
APPENDIX C
Calculating for the relative volatility [12]
Where:
∝= 11
Y = mole fraction of the light component in the vapor (0.977) X = mole fraction of the light component in the liquid (0.5304) Substituting the values given, the relative volatility of the mixture is 37.60896. APPENDIX D * All data were collected from Heuristics [4]. 1. Tray S pacing pacing : For reasons of accessibility, tray spacings are made 20-24in. 2. Weir Heig ht: ht: Weir heights are 2 inches. 3. Tray Holes and Hole A rea: Sieve tray holes 0.25-0.50 in. dia, hole area being 10% of the active cross section 4. Tray Thickness: In general, tray thickness is about gauge 10 (0.134 in; 3.40 mm) for carbon steel 5. Press ure Drop: Pressure drop per tray tra y is of the order of 747 Pa (3 ( 3 in. water) or 689.5 Pa (0.1 psi).
owncomer C leara learance nce: It is where the liquid is discharged from the bottom of 6 . D owncomer the downcomer onto the tray below, should be 0.5 in (1.25 cm) smaller than the outlet weir height to ensure a positive downcomer seal. 7 . Weir Leng th: The chord length will normally be between 0.6 and 0.85 of the column diameter.
Hole Pi tch: Normal practice is to use a hole pitch to hole diameter ratio between 8. Hole 2.2 to 3.8. 9. Fr actional Hole Ar ea: 8 -12% of bubbling (Active) area 10. Sk irt heig heig ht: ht: See references [5].
**ALL VA LUES LUES USE IN MATHCAD MATHCAD A R E IN SI UNIT UNITS S OTHER OTHER WISE STATE D. Flooding velocity and Column diameter:
Thickness of the column:
Weight of the column:
Height of the tower:
Active area, area, Fractional Fractional hole area and Number Number of of holes:
Weeping velocity: Umin,op > Umin. Therefore, no weeping could occur.
For Pressure Vessel: Double Welded butt joints and fully radiographed; E = 1 Corrosion Allowance for non-corrosive feed = 3 mm Max possible operating Temp = 113.5 oC Pressure = 20psia (5psig) Therefore: MAWP = 5 + 25 psig = 30psig
Design Temp. = 113.5 oC + 10oC = 123.5 oC Other concerns: Pressure vessel head: Both are hemispherical, nozzles for feed, bottoms and distillate product, and skirt thickness. APPENDIX E
This section displays the detailed parts of a sieve tray distillation column with its nomenclature and definition.