close

Вход

Забыли?

вход по аккаунту

?

Optimisation of Methyl Tert-Butyl-Ether (MTBE) Synthesis Processes using Aspen-Plus.

код для вставкиСкачать
Dev Chem. Eng Mineral Process., lO(1R).p p 89-103.2002
Optimisation of Methyl Tert-Butyl-Ether
(MTBE) Synthesis Processes using AspenPlus
S. Kim and P.L. Douglas"
Department of Chemical Engineering, University of Waterloo,
Waterloo, Ontario N2L 3G1, Canada
Four diflerent configurations for the production of MTBE were simulated and optimised using AspenPlus All process conjigurations were simulated and costed using
AspenPlus The total annual cost (TAC), i e the sum of the individual annualised
capital and operating costs, was calculated using models consistent with the conceptual design level Optimisation of each process configuration using AspenPlus simulation models was performed to determine the optimal design and operating variables
The optimised configurations were compared on an economic basis
It was possible to achieve iso-butylene conversions in the range of 90 to 97%
with a single conventional reactorfollowed by a distillation tower However, by using
a reactive distillation column it was possible to achieve iso-butylene conversions
> 99% The optimal configuration was found to be a combination of an isothermal
reactorfollowed by a catalytic distillation column The optimal TAC of this configuration was 22% lower than its nearest competitor, the catalytic distillation column
It was found that the net MTBE reaction rate remains high until a substantial
amount of iso-butylene is converted, and then decreases quickly as the reacting liquid
composition approaches the reaction equilibrium Therefore, it is not surprising that
the best means of synthesis is to carry out rhe bulk of the initial conversion using a
conventional reactor and to apply reactive distillation to the reactor effluent that is
close to equilibrium at the exit temperature
Introduction
The purpose of this research was to optimise and compare four MTBE (Methyl-TertButyl-Ether) process alternatives that involve conventional and reactive distillation
process technology For the various MTBE configurations studied, feed composition
and process operating conditions were taken fiom [ I ] and are presented in Table 1.
The hydrocarbon feed for the MTBE process should be rich in isobutene. In practice,
this comes from various fluidised catalytic cracking (FCC) units (1535% isobutene),
a stream cracking unit (40-55%), or an isobutane dehydrogenation unit (40-55%) [2].
In this paper, it was assumed that the hydrocarbon feed comes from an FCC unit containing 35 6% isobutene Process operating conditions are described below.
Authorfor correspondence (email pdouglas@cape uwaterloo ca)
89
S Kim and P L Douglas
MTBE product puritv. The MTBE product purity is normally in the range of 97%
to 99% However, the production of high purity MTBE (99%) is not important since
the by-products, TBA (tertiary butyl alcohol) and DIB (di-isobutene), also have good
octane blending values and are widely used in gasoline [3]. The MTBE purity required
for this research is set to 98%.
MTBE production rate. The MTBE product stream flow rate was fixed at 197
mol/s [4].
Isobutene conversion: Although the range of conversion examined is from 96% to
98%, a high conversion is normally required if pure n-butene is to be fractionated
from the MTBE process raffinate, or if the raffiate is fed to other chemical processes.
In most cases > 98% conversion is necessary [3].
Table I . Feed stream specifications
FIOW rate
Temperature
(a
Pressure
(arm)
215.5
320
11
Isobutene
195.3 (35.6%)
350
11
n-butene
353.7 (64.4%)
350
11
Methanol
c4
stream
(mo//s)
Based on the process conditions above, an AspenPlus simulation complete with
capital and operating costs was developed, and then economic optimisation was performed using the built-in SQP optimisation routine in AspenPlus. The economic optimisation problem is a non-linear, steady state, constrained optimisation problem. The
objective hnction consists of the sum of equipment and utility costs as shown in
Equation (I).
subject to g (x) 2 0
h (x) = 0
where x
=
=
CtYpp
Curr~,rres=
=
g (x)
=
h (x)
90
decision variables (see Table 2);
total annualised equipment cost ($/year);
total annual utility costs ($/year),
the set of process inequality constraints;
the set of equality constraints represented by the process model.
Optimisation of MTBE Synthesis Processes using AspenPlus
In addition to the equality constraints represented by the process model, an inequality constraint, that the MTBE product purity be L 98%, is required. The decision
variables can be divided into two groups, one for the column and reactor design specifications and the other for the operation of column and reactor. AspenPlus requires
two specifications to define the operation of a rigorous distillation coIumn module.
Once the bottom product flow rate is set to 197 mol/s, the column is left with only one
more specification. In our case the reflux ratio was manipulated for various sets of
column specifications such as the number of stages and feed locations. The decision
variables used in the optimisation problem are given in Table 2.
Table 2. Decision variables used in the optimisation problems
I
I
I
Reactor
Column
0
0
0
0
Length (m)
Operating temperature (K)
0
0
I
I
Number of stages
Number of stages of each section in a
reactive distillation column
Feed stage locations
Reflux ratio
I
Theory
Simulafion and Optimisation using AspenPlus
The simulation and optimisation was performed using the AspenPlus commercially
available flowsheeting package, licensed from Aspen Technology Inc. [ 5 ] . The four
process configurations are:
0
a single stage conventional reactor followed by a distillation tower,
0
a two stage conventional reactor followed by a distillation tower,
0
a reactive distillation column combining reaction and distillation in one unit, and
0
a combination of conventional reactor followed by a reactive distillation column.
The four flowsheets to be modelled and optimised are shown in Figure 1. The methanol recovery unit, which requires additional capital and utility costs, was not considered.
The RPLUG rigorous plug flow reactor model was used to simulate the chemical
reactors [ 5 ] . We assumed that there was no mixing in the axial direction, but perfect
mixing in the radial direction. Both adiabatic and isothermal operation can be specified in the RPLUG model.
The DSTWU and RADFRAC models were used to simulate distillation and catalytic distillation columns [ 5 ] DSTWU is a shortcut model using the Fenske-GillilandUnderwood models. It calculates the number of stages and the required reflux ratio,
given the product recovery specifications and the ratio of the desired reflux ratio to the
minimum reflux ratio. The results from DSTWU give good starting points for the
rigorous RADFRAC model, which is a rigorous tray-by-tray model for simulating all
types of vapour-liquid fractionation operations and also reactive distillation. The
91
S Kim and P L Douglas
RADFRAC model is based on ideal equilibrium stages and can handle chemical reactions on the trays using either chemical equilibrium or chemical kinetics.
Conjiguration I
Configuration 2
A two-stage conventional process
1
Conjiguration 3
A reactive distillation column
-
Conjiguration 4
Combining a primary reactor and
reactive distillation column
Figure 1 MTBE process flowsheets to be modelled and designed
(Is cooler required in conflgurution 2? r f so, place it correctly)
The reaction kinetics represented by Rehfinger and H o f h a n n [6, 71 were supplied
to AspenPlus through a subroutine in both the RADFRAC and RPLUG models.
RADFRAC solved the kinetic models and VLE equations associated with each stage
92
Optimisation of MTJ3E Synthesis Processes using Aspen Plus
in the reactive distillation column. The liquid and vapour phases were assumed to be
well mixed; and the physical properties were calculated at these mixed conditions. In
the case of kinetically controlled reactions, the rates of reactions are also calculated
with these average conditions IS]. The reaction rate for the MTBE synthesis reaction
is represented by Equation (2) [6]
where r is the reaction rate per unit catalyst mass (kmoVs/kg catalyst) for the reactive
distillation or reaction rate per reactor length (kmoVm-s) for the reactor, q is the catalyst activity (4.9 equivkg or 3.3E+03 equiv/m3 ), k, is the kinetic constant, KO is the
equilibrium constant, and a is the component activity. The temperature dependencies
of the reaction rate and equilibrium constant are presented in Table 3. The reaction
rate is a function of component activities, and is described using a user supplied
FORTRAN subroutine The liquid hold-up was assumed to be 1000 kdstage.
Table 3 Reaction rate constant and thermodynamic equilibrium constant
as afunction of temperature
k r = 3 67x10 12,-11 llO/T
Ka = 284 x exp[f(T)]
+C4(T2 -T,2)+C5(T3 -T2)+C6(T4 - T : )
T,=298 15K
C1= -1493 K;
c2=-77.4;
C3= 0.508 K ' ;
C4=-0.913x 10-'K; Cs=l I I X I O ~ K -c6=
~ ; -0 6 2 8 ~ 1 0 - ~ K ~
The UNIQUAC model was selected to model the liquid and gas-liquid interactions, while gas-phase interactions were modelled using the well-known RedlichKwong equation of state. The UNIQUAC parameters used by Rehfinger and Hoffmann [6] to describe the kinetics of the MTBE reaction are presented in Table 4.
General thermodynamic data were obtained from the DIPPR databank [5].
Design Considerations
To simulate and optimise each MTBE process, various parameters were fixed to values found in the literature. The feeds to the process and output specifications were
described above.
93
S. Kim and P L Douglas
Table 4. Coeflcients to calculate the UNIQUAC binary interactionparameters,
zIi = exp(A, / T )
~
~~
A,”
Value
A,”
Value
A12
35.38 K
88.04 K
-706.34 K
A23
-5220 K
-468.76 K
24.63 K
A13
A2 I
MeOH
=
I , isobutene
= 2,
A31
A32
MTBE
=3
The reaction temperature should be low enough (3 10-365 K), to increase the isobutylene equilibrium conversion and prolong catalyst life [9]. Typically, in the case of
the two-stage conventional process, industrial experience has shown a typical catalyst
life of more than one year in the first reactor, and longer for the second reactor [lo].
During the optimisation studies, the catalyst life was assumed to be two years. Most
catalytic reactions have a preferred range of temperature for which the catalyst can be
used most effectively. Lower reaction temperatures also decrease the formation of
by-products The preferred range of catalyst temperature overlaps the preferred temperature and pressure range for the separation of MTBE from the reactants and inerts
by distillation [9]. This then makes the MTBE process an excellent candidate for
reactive distillation. The reactive section is positioned in the middle of the reactive
distillation column. If the bottom temperature is > 400 K, then the preferred temperature of reactive section can be obtained at a pressure of 11 atm. The temperature profiles calculated by Jacobs and Krishna [ 11 in the reactive distillation column were reproduced using AspenPlus in this work and are shown in Figure 2. For the MTBE unit
simulations, the system pressure was set to 11 atm and the temperature range was limited to 3 10-365 K for the reaction.
440
h
420
sa 400
L-
3
.w
E 380
8
E 360
?
--*
4
b
:
act‘W p
_
.’ - - --
:
34 0
320
I
1
I
I
2 3 4
TOD
I
I
I
I
I
I
I
I
I
I
5 6 7 8 9 10 1 1 12 13 14 15 16 17
Column stage
Figure 2. Predicted column temperature profile at I I atm
94
I
Bottom
Optirnisation of MTBE Synthesis Processes using Aspen Plus
Thermodynamic equilibrium conversion varies with the methanol to isobutylene
ratio [ 111 Conversion increases with the methanol to isobutylene ratio at constant
temperature However, excess methanol increases the methanol recovery costs. The
latter would dictate a methanol content to be limited by the azeotrope composition of
the MTBE column overhead. Too large an excess of methanol is harmhl to MTBE
purity, while a sub-stoichiometric amount of methanol is detrimental to the catalyst
life. Thus a compromise needs to be struck between increased conversion and cost.
The ratio of methanol to isobutylene was set to 1.1 in designing the MTBE process for
this research.
The goal for this research is to compare four MTBE process alternatives at their
optimum design conditions. We must calculate the capital and operating costs for the
various configurations of the MTBE synthesis process at the optimised design conditions. In the design of MTBE processes, the state of the feeds to the process and the
output specifications were given above and the optimal equipment sizes and configurations were determined by adjusting the various design parameters such as reflux
ratio, the number of stages, etc. The optimisation problem is simplified by eliminating
all but the most important design variables and by including only the dominant cost
functions.
Cost Models
There are several levels of engineering design and cost estimates; the American Association of Cost Engineers (AACE), has proposed three classifications of cost estimates
as shown in Table 5 . Each classification has a degree of accuracy associated with it.
The AACE defines “order-of-magnitude’’ estimates as those made without detailed
engineering data. “Budget” estimates are those from preliminary flowsheets, layouts,
and equipment details; “defmitive” estimates are those prepared in considerable detail.
Since ‘‘conceptual design” focuses on the structure of the process, and not on the detailed design of the equipment [121, the simplified “order-of-magnitude” estimate level
is used for the cost models. Equipment cost correlations have been published by Peters
and Timerhaus [ 131, Chilton [14], Happel and Jordan [15], and Guthrie [16].
Accuracy
Type
Budget
Definitive
- -30%
+30% - -15%
+50%
Order of magnitude
I
+15%
-5%
It should be emphasised that the simulation and optimisation calculations associated
with screening are not rigorous All that we require is sufficient accuracy, acceptable
for the engineering purposes to compare each configuration. Therefore, the design
analysis using a simulation model is based on simple flowsheets and economic
95
S Kimand P L Douglas
models. We can use a simulation model to optimise the design of the process by making a series of case studies to ensure that the plant will work properly over a wide
range of operating conditions.
The cost models for the processes were developed using a simplified version of
Guthrie’s correlation [16]. The normal material (the base costs assume carbon steel)
and pressure correlation factors are used to estimate the purchased cost, but the most
conservative cost factor was used to estimate the installed costs. These correlations
were updated using the M&S (Marshall and Swift) index, which is updated monthly in
Chemical Engineering [ 171. The M&S index and the utility costs are based on the year
1993. The 1993 M&S index from Chemical Engineering [ 171 is 965 0
Guthrie’s correlation [ 161 for pressure vessels, columns, and reactors is:
1.066 x H0*’O2 x (2.18 + Fc)
Installed cost ($) =
(3)
where D is the diameter (fi), H is the height (ft), and Fc is the correction factor assumed to be one. For preliminary design purposes, the tower height is given by Douglas [12] as H = 3(1 15)N/E0 where Eo is the plate efficiency and is assumed to be one,
and a 3 ft tray spacing was used. The recommended tray spacing depends on the
tower diameter. If the tower diameter has a range of 12 to 24 ft., then the recommended tray spacing for petrochemical distillation columns is 3 ft [18]. Sieve trays
were assumed for all simulation studies. U.S Patent No. 4,439,350 by Jones [19],
discloses “the contact catalyst structure that consists of closed porous containers containing the solid particulate catalyst and a clip means for holding and spacing containers as a unit for installation above a distillation tray for the use of the reactive distillation”. These perforated or porous metal or plastic containers are supported above a
conventional distillation tray so as to allow liquid flow through and past them, and
also to allow vapour passage through the liquid. According to the present invention of
Jones [ 191, catalyst structures are positioned above a perforated sieve tray.
The installed equipment costs were annualised by multiplying them by a capital
charge factor of 113 year for petrochemical processes [ 121:
- x 101.9 x D1.066
Annual c o s t ( $ / y r ) = (;s)
H0.802
x(y)
(4)
Guthrie’s correlation for heat exchangers.
Annual cost($/yr.) = ( -M
) x&I S0 1 . 3 ~ . 4 ~ * ~ ~
280
x(
2.29 + Fc
)
(5)
where A is the area (ft2) The correction factor, Fc, was assumed to be one. The heat
exchanger area is computed from the following equation:
A T =-
96
Q
UATL
Optimisation of MTBE Synthesis Processes using AspenPlus
where, AT is the total heat transfer area, U is the overall heat-transfer coefficient, Q is
the rate of heat-transfer in the entire exchanger, and ATfdis the logarithmic mean temperature difference.
It was assumed that the overall heat-transfer coefficient for the condenser, Uc,
was 100 Btu/hr fi2 OF [ 121 The temperature of cooling water was assumed to be
30°F. The temperature driving force in the reboiler must be constrained to be less than
about 30 to 45°F to prevent film boiling. We expect to obtain a high value of the
overall heat-transfer coefficient in the reboiler because we have heat transfer between
a condensing vapour and a boiling liquid. Thus we expect that there will be a limiting
heat flux in the reboiler, of U,ATL = 1 1,250 Btu/hr ' ft2 .
If the cooling water costs $0.16/MMBTU [20], the annual cost of cooling
water is:
Annual cos t ($ / yr) =
= 0.004453xQc
(7)
For the case of 24 atm saturated steam pressure (where AHs = 2800.4 kJ/kg), and the
cost is $0.017/10001b [20], then the annual steam cost is:
Annual cos t ($ / yr )
= 0.00039267 x Qr
(8)
The price of Amberlyst 15 was obtained from an engineer at LG Caltex in Korea [21]
and is %4000/ton.The catalyst life was assumed to be two years.
Results and Discussion
This process although reasonably simple has received a lot of attention primarily due
to its previous industrial significance, and because it was one of the most significant
processes to use catalytic distillation columns The process can be carried out in any
number of ways involving from 1 to 4 pieces of equipment. The optimum design and
operating conditions for each of the MTBE synthesis process configurations are presented in Table 6. High conversion can be achieved using:
0
a single stage conventional process,
0
a reactive distillation column, and
0
a combination of conventional reactor and reactive distillation column configurations.
97
4
Conversion 96%
Product Flowrate(191.251 moVs)
A single stage conventionalprocess
Ccmversim98.89%
A sin& stage amvcniiond process
Configuration
Utilily
Distillation
column
reactor
Adiabatic
exchanger
Heat
reactor
Utility
Distillation
column
reactor
Adiabatic
exchanger
Heal
reactor
I
I
I
1
I
I
1
Mole-RR
Inlet Temp.
(317.79)
( 0.7856)
( 19”: 4.349m)
Stages (a),
Feed (13)
Mole-RR
Max. dia.
( 0.7869)
Volume
(27.79 m3)
Volume
Stages (20),
Feed (13)
Max. dia.
(19Ih:4.898m)
Volume
I(635.02m3)
Volume
I
I
1
I
1
81,098
1
119.948
41,839
48,513
160,415
$
149,550
I
I
$
714.668
-
1.72
11.25
@ Pmductlon rate I 197 molls, MTBE pu y P 98%
I
1’76E07
-1.20E07 $ 354,018
$
-2.43E06 $
1
1
-1.76E07 $
$
1-98E07
-1.15E07 $ 375,568
-1.33E06 $
Relative
TAC
Table 6. Optimum Design and Operating Conditions for each of the MTBE Synthesis Process Configurations
Conversion : 99.33 ?4
Conversion : 99.15%
E
Configuration
~
Utility
Catalytic
distillation
column
Isothermal
reactor
Utility
column
distillation
Catalytic
Equipment
1-
Table 6. Continued.
Max. dia.
(17" : 4.076m)
Feed (lllh)
RfIRIS
( O W
Stages (18)
Volume
(1.26 m3)
Feed (12)
Max. dia.
(19Ih :4.230m)
(011I n )
RfIWS
Stages (20)
Optimal Spec.
I
1
Mole-RR
(1.6026)
Temp.
(366.35K)
Mole-RR
(2.978)
Operation
I
$ 145,977
8 Production rate o 197 n
-I
I
1'67E07 $ 261,624
-1.67E07
-1.46E07 $ 7,752
$ 118,241
1*097E07
-2.559E07 $ 389,281
y = 98%
-
1
1.22
Relative
TAC
-
Two-age conventianalprocess
Configuration
utility
Distllafkn
column #2
reactor
Adiabatic
lieat
exchanger
I
I
I
Stages 6)
Feed (9 )
(15Ih:3.861)
I
(0.495)
Mole-B
(20.87)
Temp.
1334.92 -
gzbg:)
Distillalion
column
#l
Max. Dia.
Mole-RR
(0.707)
Mole-B
(176.13)
Optima'
Operation
condition
Temp
(351.16K)
spec.
Optimal
Volume
(4.94m3)
Equipment
Isothermal
reactor
Table 6. Continued.
I
I
I
$186,329
$220,810
$ 30,514
$ 11,657
298,950
I
I
$773,922
TAC
(Slyr..)
@ Productlon rate = 197 mUs, MTBE put
9.83E06
-9.63M6
-8.14E5
-t t
I
1.52E07
-1.25E07
$ 25,662
(waul
-1.67E07
AC
($syr.)
Heat duty
1
:
98%
-
1.86
Relative
TAC
-
Optimisation of MTBE Synthesis Processes using Aspenplus
A single-stage conventional process achieved conversions over 98%, however, the temperature
in the second reactor in adiabatic mode was very low resulting in a large reactor and proportionately large cost of $4,671,573 Therefore, to achieve reasonable costs a single-stage MTBE
process should be used only for lower conversions of w 96%
A combination of a conventional reactor and a reactive distillation column was
found to be the optimal arrangement for producing MTBE. This general conclusion
has been reported by others but without quantitative results [22]. The total annual cost
for the combination process is almost 50% less than the conventional process, and
20% less than its nearest competitor, a single reactive distillation column.. The size of
a reactor in a combination process is smaller than in the conventional process, and the
column size is smaller than when a reactive distillation column only was used. With
the lowest cost, the combination process achieves the highest conversions. In addition,
the unconverted reactant concentration is low enough to eliminate a recycling unit,
which would require additional investment
When the reactive distillation technology is applied, the required heat duty is decreased It is important to note in Table 6 that the number of rectification stages of a
reactive distillation column at the optimal condition is zero. That is when rectification
stages are added, no improvement is obtained, therefore, no rectification stage is
needed. This shows that the function of the reactive section in a reactive distillation
column is not only to provide a site for the reaction but also to purify the top product.
As the number of reactive and stripping stages increase, the attainable isobutene conversion increases; however, continual addition of those stages results in diminishing
benefits.
If one looks at the optimal cost as a h c t i o n of the number of pieces of equipment, no clear trend emerges. Conventional wisdom was that a CD column would reduce costs when compared to a process involving reactors followed by separators (i.e.
I-unit versus 3-units). The results in Figure 3 show that although this may be true, the
idea that a 2-unit process would be cheaper than a 1-unit process is not intuitively obvious Although this trend cannot be generalised it is expected that these non-intuitive
results will be found in other CD processes.
Conclusions
Aspenplus is a suitable process simulation system for performing steady-state
simulation and optimisation of MTBE synthesis process alternatives.
2 The combination of a conventional reactor and a reactive distillation column was
found to be the optimal configuration with an equivalent TAC z 22% lower than
the next closest competitor, namely the reactive distillation column.
3. The intuitive conclusion that a process with fewer process units is more economical than one with more process units was shown to be incorrect in the case of the
MTBE process.
1.
101
S. Kim and P L Douglas
2
I
3
4
#dRmessW
Figure 3 Normalised optimal annualised costs vs number of process units
References
Jacobs, R ,and Krishna, R 1993 Multiple solutions in reactive distillation for Methyl tert-Butyl ether
synthesis, Ind Eng Chem Res ,32(8) 1706-1709
Miracca, I , Tagliabue, L and Trotta, R 1996 Multitubular reactors for esterifications, Chem Eng
Sci , 5 1 (10) 2349-2358
Bitar, L S , Hazbun, E A and Piel, W J 1984 MTBE production and economics, Hydrocarbon Processing, 63(10) 63-66
Hearn, D April 9, 1985 Transesterification method, US patent No 4,510,336
Aspen Technology Inc , Cambridge, Massachusetts, USA November 1995 ASPEN PLUS Users manual, release 9 2 edition
Rehfinger, A and Hoffmann, U 1990 Kinetics of the Methyl Tertiary-Butyl Ether liquid phase synthesis catalyzed by ion exchange resin 1 Intrinsic rate expression in the liquid phase activities,
Chem Eng S c i , 45(6) 1605-1617
Rehfinger, A , and Hoffmann, U 1990 Kinetics of Methyl tertiary butyl ether liquid phase synthesis
catalyzed by ion exchange resin 2 Macropore diffision of methanol as rate controlling step, Chem
Eng Sci, 45(6) 1619-1626
Velo, E , Puiglaner, L , and Recasens, F 1988 Inhibition by product in the liquid-phase hydration of
isobutene to tert-butyl alcohol kinetics and equilibrium studies, Ind Eng Chem Res , 27(12) 2224223 1
DeGarmo, J L , Parulekar, V N , and Pinjala, V 1992 Consider reactive distillation, Chem Eng
Prog ,88(3) 43-50
10 An~llotti:F , Pescarollo, E , Szatmari, E and Lazar, L 1987 MTBE from butadiene-rich C ~ SHydro,
carbon Processing, December 50-53
11 Brockwell, H L , Sarathy, P R and Trotta, R 1991 Synthesize ethers, Hydrocarbon Processing, September 133-141
-
102
Optimisation of MTBE Synthesis Processes using AspenPluA
12 Douglas, J M 1988 Conceptual Design of Chemical Processes, McGraw-Hill, New York
13 Peters, M S and Timmerhaus, K D 1968 Plant Design and Economics for Chemical Engineers,
McGraw-Hill, New York, Chapters 13 to 15
14 Chilton, C H 1949 Cost data correlated, Chem Eng 56(6) 97
15 Happel, J and Jordan, D G 1975. Chemical Process Economics, Marcel Dekker, New York, Chapter
5
16 Guthrie, K M 1969 Capital cost estimating, Chem Eng , 76(6) I14
17 Chemical Engineering Magazine December 1993 M&S (Marshal and Swift) index
18 Pham, H N and Doherty, M F 1990 Design and synthesis of heterogeneous azeotropic distillations I1 Residue curve maps, Chem Eng Sci ,45(7) 1837-1843
19 Jones, E M 1989 Contact structure for the use in catalytic distillation, US Patent 4,439,350
20 Eldarsi, H S and Douglas, P L 1998 Methyl-Tert-Butyl-Ether catalytic distillation column - Part I1
Optimization, Trans IChemE, 76(5) 5 17-524
21 L G Caltex Co Ltd 1997 Personal communication with Mr K Kim - Engineer, Yeochun, South
Korea
22 Subwalla, H and Fair, J R 1999 Design guidelines for solid catalyzed reactive distillation systems,
Ind Eng Chem Res ,38,3696-3709
Received 30 January 200 1 ;Accepted aster revision: 1 1 June 2001.
103
Документ
Категория
Без категории
Просмотров
3
Размер файла
665 Кб
Теги
methyl, using, synthesis, butyl, plus, ethers, processes, aspen, tert, optimisation, mtbe
1/--страниц
Пожаловаться на содержимое документа