close

Вход

Забыли?

вход по аккаунту

?

Tert-Butyl Ethers - A Comparison of Properties Synthesis Techniques and Operating Conditions for High Conversions.

код для вставкиСкачать
-
Tert-Butyl Ethers A Comparison of
Properties, Synthesis Techniques and
Operating Conditions for High
Conversions
M. G. Sneesby’, M. 0. Tad6 and R. Datta’
School of Chemical Engineering, Curfin University of
Technology, GPO Box U1987, Perfh 6001, Western Australia
Methyl tert-butyl ether (UTBE) is currently the industrially dominant ether
oxygenate although ethyl tert-butyl ether (ETBE) is growing in importance as an
alternative. ETBE has superior oxygenate properties (lower RVP and higher
octane) and can be made Pom renewable ethanol. The ETBE reaction system is
thermodynamically more restrictive than MTBE and the increased relative volatility
of the ethanol-isobutylene system compared with the methanol-isobutylene system
makes operation of a reactive distillation column more dificult. In this paper, the
advantages of reactive distillation for tertiary ether production are discussed.
Twenty different reactive distillation column conjigurations are identified with
different designs being preferred for ETBE and MTBE. The effects of eight key
operating criteria in reactive distillation systems are briefly discussed. Finally,
reactive distillation systems producing MTBE and ETBE are simulated using Proill
ver 4.0 to demonstrate that high conversions are possible for both products with
relatively simple equipment.
Introduction
In recent years oxygenates have become increasingly important due to their potential
to replace traditional lead-based octane enhancers, to reduce carbon monoxide
emissions from auto engines and lowering ozone levels. In the US, laws already
exist to force refiners to include at least 2.0% by weight oxygen in gasoline in
’
Department of Chemical and Biochemical Engineering, The Univeristy of Iowa,
Iowa City, Iowa, USA
Author for correspondence. (email: sneesbym@paravel.che.curtin.edu.au)
89
M.G.Sneesby, M.O. Tad6 and R. Datta
regions deemed to have low air quality. Similar laws are expected to develop in
other countries.
There are two main classes of oxygenates - alcohols and ethers. Of these, the
ethers would appear to have the more suitable characteristics, although ethanol is
currently the second most widely used oxygenate after MTBE. Commercial tests
have been done with several ethers to confirm their suitability as oxygenates for
gasoline blending. These tests have identified the tertiary ethers - methyl tert-butyl
ether (M'IBE), tert-amyl methyl ether (TAME),
ethyl tert-butyl ether (ETBE) and
terr-my1 ethyl ether (TAEE) - as the most important. Some of the key properties of
the tertiary ethers are given in Table 1. Each oxygenate has a high octane number
and relatively low Reid vapour pressure (RVP) whereas ethanol has a high octane
number but very high RVP (124 kPa). The oxygen contents vary from 18.3 weight
% for MTEIE to 13.8% for TAEE. More oxygen in the fuel allows the engine to
operate with a leaner fuel to air ratio, thereby reducing unburned fuel and carbon
monoxide emissions.
Table 1. Some key oxygenateproperties of the tertiary ethers [1-4J.
Common
Name
MTBE
ETBE
TAME
TAEE
Full Name
methyl tert-butyl ether
ethyl tert-butyl ether
tert-amyl methyl ether
tert-amvl ethvl ether
Hydrocarbon
Reagent
isobutylene
isobutylene
isoamylene
isoamvlene
Alcohol
Reagent
methanol
ethanol
methanol
ethanol
Octane
Number'
110
111
lo5
109
RVP
(kPa)
54
28
17
10
Oxygen
Content
(wt %)
18.3
15.7
15.7
13.8
Note: 1 (RON + MON)/2
In this paper, a comparison of reaction kinetics and synthesis methods of ETBE
and MTEIE is made using simulation studies. The advantages and disadvantages of
various column configurations are discussed together with key operating criteria.
Speculations on the emerging role of ETBE in future production of oxygenates are
also made.
90
Tert-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
Raw Materials for the Production of Ether Oxygenates
MTBE is currently the commercially dominant oxygenate, due primarily to the low
cost of its raw materials methanol and isobutylene. ETBE is also produced from
-
isobutylene but uses ethanol rather than methanol as the alcohol reagent. Although
ethanol is still sigruficantly more expensive than methanol, it is generally produced
from a renewable source (biomass) and is currently attracting subsidies in some
countries, including the US, making it more economically attractive. Methanol is
usually produced from natural gas but, in principle, it can also be produced from
biomass. As environmental pressure mounts to promote semi-renewable fuels and
more economical and viable technology emerges for producing renewable ethanol, it
is expected that the importance of ETBE and TAEE will continue to increase,
particularly in developing countries where large quantities of biomass are more
readily available. Concurrently, methanol prices are continuing to rise due to high
demands.
The main hydrocarbon feeds for ether production are isobutylene (used for MTBE
and ETBE)and isoamylene (used for TAME and TAEE), both of whch are found in
relatively h g h concentmuons in the product from catalytic craclung units. These
olefins can be used drectly in the gasoline pool but their h g h RVP restricts the
volume that can be added and may force the exclusion of other components. As a
gasoline blendstock, isoamylene has a much lower RVP and, therefore, higher value
than isobutylene Thus reduces the incentive to upgrade it to an ether, although the
ethers made from isoamylene have a higher intrinsic value than the ethers produced
from isobutylene.
Ethanol has a very hgh oxygen content (35% by weight) and is therefore an
efficient oxygenate but its other properties make it less desirable in the gasoline pool.
In particular, it has a low energy content in comparison to typical gasoline
components, and therefore, requires higher volumes of fuel to be burnt to achieve the
same power. Ethanol also has a tendency to attract water. Methanol is more
corrosive and has a lower energy content than ethanol and is, therefore, generally
considered unsatisfactory as a gasoline blend.
Conversion of isobutylene to MTBE or ETBE is desirable as a volatile, high RVP
component is converted to a hgh octane, low RVP product. Further, the conversion
of a light component allows other streams (for example, excess butane) to be added
to the gasoline pool, increasing overall gasoline production at the expense of LPG
91
M.G.Sneesby, M.O. Tad4 and R. Datta
and/or refinery fuel gas. Although conversion of isoamyleneto TAME or TAEE also
allows some upgrading of lighter components, the overall gain is smaller. However,
the main driving force for new ether production facilities remains strengthening
environmental legislation worldwide.
Thermodynamic and Kinetic considerations
All the oxygenate ether reactions are reversible and limited, to some extent, by
equilibrium at the temperatures encountered in the industrially relevant range.
Typical conversions for each of the ethers are given in Table 2, however, these will
vary with reaction compositions due to the effect of concentration on activity
coefficients. High MTBE conversions are attainable with mild conditions while the
ETBE reaction is more restricted by equilibrium. Figure 1 indicates the differences
in equilibrium constants between MTBE and ETBE.
In an industrial environment, conversions of hydrocarbon approaching 95 per
cent are required. Conversion can be enhanced by using a large excess of one of the
reactants in the reaction zone, as shnwn by Le Chatelier's principle. However, the
increase in conversion attainable using an excess of one reactant is subject to a
diminishing return and, even with an infinite excess, the maximum conversion can
still remain below that required. Alternatively, a series of reactors and separation
systems with recycle can be used to boost conversion. After each reactor, the product
is removed from unreacted feed which is then fed to the next reactor or recycled.
Theoretically, this type of system has unlimited conversion provided all reactant can
be removed from the product but is costly in terms of the equipment required and
energy demand. In practice, a combination of excess reactant, multiple reactors and
recycle have been used in conventional ether processes [9].
Table 2. Equilibrium conversionsfor various ethers and conditions [4-71.
Ether
MTBE
ETBE
TAME
TAEE
92
Equilibrium Conversion at
8d.C from Stoichiometric
Mixture of Reactants
88.7%
813%
70.2%
51.O%
EquilibriumConversion at
7d'C from Stoichiometric
Mixture of Reactants
90.7%
84.7%
74.8%
55.4%
Equilibrium Conversion
'at 70'C with 10%
Excess Reactant
94.3%
88.5%
78.7%
58.7%
Tert-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditionsfor High Conversions
180
160
140
120
100
80
60
40
20
0
40
50 60 70
80 90 100 110 120 130 140 150
Temperature (deg C)
Figure 1. MTBE and ETBE equilibrium constants [5,6].
For ETBE in particular, a slight excess of ethanol is used to promote the reaction
and reaction temperatures are controlled to ensure favourable equilibrium. Methanol
recycle is widely used in h4TBE production but ethanol recycle is less attractive for
ETBE due to the detrimental effect of high ethanol concentrations on reaction rates
[S]. The high boiling point of ethanol also increases the reaction zone temperature
and reduces the equilibrium constant, making high conversions harder to achieve.
Although, these factors increase the catalyst requirement, the increased conversion
attained as a result of introducing an Zxcess of reactant is the more sigmficant effect.
Isobutylene recycle is an alternative but there are associated problems. First,
isobutylene has a higher molecular weight and lower density than ethanol, thus,
much higher recycle flow rates are needed for the same recycle ratio. Second, the
isobutylene feed is normally supplied as a mixed butanes stream. Therefore, either
even higher recycle flow is required or a continuous separation of the inerts and
reactive isobutylene must be performed. Both alternatives are difficult and/or costly.
Third, high isobutylene concentrations favour the formation of the byproduct, diisobutylene (or iso-octane), which, although having a high octane and low RVP,
limits production of the ether by consuming the reactant.
Values of
methano1:isobutylene ratio of much less than unity have been shown to lead to
abnormal catalyst reactions in MTBE systems [ 5 ] , and a similar interaction between
ethanol and isobutylene is possible. A stoichiometric reacting mixture can be used
93
M. G. Sneesby, M.O.Tad6 and R. Datta
which simplifies downstream recovery by minimising the concentrations of reactants
in the ether product, although it is generally inadequate for high conversions of
hydrocarbon because of the equilibrium considerations already noted.
The literature suggests that the reaction mechanisms for MTBE and ETBE are
similar (both conforming to Langmuir-Hinshelwood-Hougen-Watsonmodel) but that
two active centres are present for MTBE and three for ETBE [5,8]. The cause of this
difference is not clear but it prevents a direct comparison of the accepted rate
constant expressions. However, data are available for ETBE using a model with two
active centres and agreement is very close to the accepted model. This data were
used to generate a rate constant expression for ETBE [8] that can be compared
directly with the preferred MTBE expression [5]. These rate constants are shown in
Figwe 2 and clearly demonstrates their similarity.
=
Y
g
4
0
200
180
160
140
120
100
80
60
0 4 0
4 2 0
a .
50
60
70
80 90 100 110 120 130 140 150
Temperature (deg C )
Rgure 2. MTBE and ETBE rate constants [5,6,8J.
The reaction rate is a function of both the species activities and the equilibrium
constant for the reaction. Lower activity coefficients for ethanol (compared with
methanol) and a lower equilibrium constant for ETBE results in the ETBE reaction
being sigruficantly slower than for MTBE. This is shown clearly by examining
actual reaction rates at low and high conversions in systems operated under similar
conditions and with the same final conversion (90%). Such data are presented for
MTBE and ETBE in Tables 3 and 4, respectively, and graphically in Figure 3. It has
94
Tert-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
also been found that ethanol has a significant retarding effect on the reaction rate for
the ETBE system 181. This is evident from Table 4 where increases in the excess
ethanol cause the reaction rates to fall, despite a continuously increasing rate
constant. The lower equilibrium constant, lower reaction rate, higher boiling point
of ethanol compared with methanol, and the detrimental effect of ethanol on the
reaction rate make satisfactory ETE3E reaction conditions more difficult to achieve
than for MTBE and reduce the final conversion attainable.
Table 3. Estimated reaction ratesfor 90% isobutylene conversion to MTBE via
methanol recycle [5].
Temp
PC)
.
I
Equilibrium
Constant
(W
40
50
60
70
80
90
100
150
__
94
60
39
26
17.1
11.6
Required
Me0H:IB in Feed
for 90% IB
Conversion
0.92
0.93
0.95
0.98
1.03
1.11
1.25
Rate
Constant,k
(kmoW
kgcat.hr)
0.03
0.07
0.20
Initial
Reaction Rate
(kmoll
kgcat.hr)
0.009
0.026
0.068
0.167
0.381
0.808
1.569
0.49
118
2.70
5.89
Note: A constant ratio of activity coefficients of 0.33 (ymE/(ykm
calculations.
Reaction Rate
at 80%
Conversion
(kmolkgcat.hr)
0.013
0.032
0.071
0.140
0.245
0.387
0.548
x 7,~))is used to simplify
Table 4. Estimated reaction ratesfor 90% isobutylene conversion to ETBE via
ethanol recycle [6,8].
Temp
(‘C)
Equilibrium
Constant
(16.)
40
74
50
44
60
27
17
11
7
5
70
80
90
100
Required
Et0H:IB in Feed
for 90% I6
Conversion
0.96
1.Ol
1.09
1.23
1.52
2.29
7.30
Rate
Constant,k
(km0U
kgcat.hr)
0.02
0.06
0.14
0.36
0.84
1.89
4.07
Initial
Reaction Rate
(kmoll
kgcat.hr)
0.010
0.025
0.058
0.120
0.208
0.270
0.144
Reaction Rate
at 80%
Conversion
(kmoVkgcat.hr)
0.010
0.053
0.071
0.080
0.075
0.055
0.018
Note: A constant ratio of activity coefficients Of 0.48( y ~ & ( y X~ y18))is used to simplify
calculations.
9s
M.G. Sneesby, M.O.Tade‘and R. Datta
-
1.6
1.4
3
1.2
=
1.0
9
z
S 0.8
3
0.6
5
0.4
0
a3
L
0.2
0.0
40
60
70
80
Temperature (deg C)
100
Figure 3. Estimated MTBE and ETBE reaction rates at high conversion
conditions in a plug flow reactor [.5,6,8].
Ted-Butyl Ether Synthesis
Conventional Systems
Small volumes of methanol or ethanol in the gasoline pool do not adversely affect its
quality. However, any s i m c a n t amount will start to attract water and form a
second liquid phase which is undesirable. Therefore, oxygenate production is
normally controlled in order to produce a high purity oxygenate which is relatively
free of alcohol (either methanol or ethanol). The other product from the oxygenate is
termed C4 or CSraffnate. It is also produced relatively free from alcohol, so that it
can be blended directly into the gasoline pool, and contains any unreacted
isobutylendisoamylene plus the majority of non-reactive hydrocarbons that were
present in the feed. As the alcohol reactant is expensive, this is an added incentive
to remove it from both products before blending.
A high conversion of the hydrocarbon reagent (isobutylene or isoamylene) is
considered desirable as it maximises the “value added” to the raw materials and
minimises the hydrocarbon feed for a fixed oxygenate product rate. The conversion
achleved is very dependent upon both the oxygenate (thermodynamics) and the
process selected. MTBE produced via conventional processes requires multiple
reactors and recycle loops to achieve very high conversions. However, reactive
distillation technology can readily achieve greater than 98% conversion of reactive
96
Tert-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
isobutylene. By comparison, ETBE produced via conventional methods results in 56% lower conversion than MTBE. Reactive distillation can lift this to greater than
95%.
A conventional system (without reactive distillation) uses a two-stage reaction
system. To minimise the total reactor volume required, the first reaction stage is
performed isothermally in a tubular reactor where most of the reaction takes place
and the temperature can be more easily controlled. The heat released during reaction
heats the mixture and eventually stops the reaction due to equilibrium
considerations. An intercooler is used prior to the second reaction stage which cools
the mixture sufficiently to allow the reaction to reach the desired conversion. The
second reaction stage is conducted adiabatically at low temperature. The reaction
rate is much slower due to the reduced temperature so that the reactor volume may
be similar to the first reaction stage although much less ether is being formed. The
second stage is commonly performed in a drum reactor [9].
Reactive Distillation Systems
Reactive distillation overcomes thermodynanuc and kinetic restrictions by utilising
the high degree of internal recycle (both methanoYethano1and isobutylene) already
present in a distillation system to create reaction conditions suitable for high
conversions. Other advantages include effective temperature control due to boil-up.
As an example of the change in composition that occurs within a distillation
environment, Table 5 shows the composition profile produced in distilling an
equilibrium mixture of ETBE, ethanol and isobutylene (calculated via Prom). The
column feed is a mixture of ETBE, ethanol, isobutylene and inerts such as might be
found in the product from a conventional reactor.
The stage-to-stage compositions are sufliciently changed from the feed
composition to promote further reaction, even though the reacting temperatures are
such that the equilibrium constant is relatively low. Apparent reaction rates can be
calculated from the temperature and composition and compared with what would be
achieved in a conventional reactor. However, as the reaction proceeds, the
composition profile will change and the reaction rate will slow. Thus, the data
presented in Table 5 are not real reaction rates but ‘initial’ reaction rates that are
somewhat equivalent to the initial reaction rates shown in Tables 3 and 4.
Furthermore, the reaction rates given in Table 5 are likely to be overstated due to
97
M. G.Sneesby, M.O. TadP and R. Datta
intraparticle diffusional limitations at high conversion and high temperature [ 5 ] .
Note that no reaction will occur on or below tray 7 since the ratio of reactants to
products (multiplied by stoichiometric indices) is higher than the equilibrium
constant K,) at these points. However, the reaction rate is high at only two trays
above the feed point.
Table 5. Compositions and potential reaction rates for ETBE in a distillation unit
[6.8,17/.
Tray
1
2
3
4
5
6
7
8
9
10
Temp
('C)
4 0
72
74
81
83
8 9
103
121
135
145
Mol% of
Isobutylene
16.3
15.8
14.8
12.3
11.4
9.3
5.8
2.8
1.2
0.5
Mot% of
Ethanol
Mol% of
ETBE
2.5
3.8
5.8
8.0
10.4
16.4
26.1
32.9
29.8
16.9
0.1
0.7
3.9
16.3
18.6
25.1
36.7
48.9
62.3
79.9
Equilibrium
Constant
(&a)
74
16
14
11
10
7.7
4.4
2.3
1.4
1.o
Products/
Reactants
0.03
0.2
0.8
3.4
3.7
5.0
10
27
Potential Initial
Reaction Rate
(kmoVkgcat.hr)
0.07
1.06
0.68
0.63
0.49
0.27
88
400
Notes: 1 Feed to tray 4 is 36% ETBE, 5% IB, 5% EtOH and 47% C, inerts.
2 Column overhead pressure is 900 kPag.
3 'Products/reactants" is given by (amE/(aEIOHx ale) and is directly Comparable to the
equilibrium constant.
4 No (orward reaction on trays 7-1 0 as "products/reactants" IGg.
Reactive Distillation Column Design
Basic Configuration
The design method for a reactive distillation (RD) column is complex and involves
several criteria not present in the design of either a conventional distillation column
or a reactor. These include the effects of the operating pressure on reaction rate and
equilibrium (as well as normal VLE effects), the location and size of the reactive
sections, the possible use of multiple feed points and the composition and
temperature of both the main and supplementary feeds, as well as the basic
configuration and recovery criteria. Although some papers are now appearing in the
literature which attempt to address RD design issues [ 10,111, a rigorous design will
inevitably involve trials of a range of configurations and operating conditions. The
optimal design will always be somewhat dependent on external factors and local
conditions and the development of a successll design will normally be a lengthy and
98
Ten-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditionsfor High Conversions
diflicult process. Nevertheless, conversions of greater than 95% can be shown to be
achievable (by simulations) in quite simple configurations.
In industrial systems, it is common to retain the first reaction stage from
conventional ether units so that the feed is partly reacted prior to entering the RD
column. This reduces the catalyst requirements inside the column and provides
some protection against catalyst damage as removing and reinstalling catalyst inside
a column is diflicult and costly. Only a low conversion is now required in the
reactor, and higher temperatures can be used and recycle can be eliminated in that
system. The extent of the reaction achieved before the RD column depends on the
relative economics of reaction in a conventional reactor and RD column. Eventually
a point will be reached where the reaction in a conventional reactor slows sufficiently
so that it is unattractive to continue further.
Five key questions must be answered when determining the basic configuration of
the RD column:
1. Is the system ternary (hvo reactants and one product present only) or
quaternary (two reactants, one product and one or more non-reacting
components present)?
2. Are the two reactants present in a stoichiometric ratio or is one reactant
used in excess?
3. If one reactant is used in excess, is it the lighter reactant or the heavier
reactant?
4.
Is the conversion of the limiting reactant complete or incomplete?
5.
Assuming the product is heavier than both reactants (generally true for
systems of the type A + B e C) and is therefore recovered as the
bottoms product (and the lightest reactant is recovered in the overheads),
is the intermediate component recovered in the bottoms or overheads
from the column, or as a side-stream?
Other questions must also be answered before proceeding with the column
design, includmg:
6. Are any azeotropes likely?
7. Should the reactants be introduced at different stages?
8. Where a catalyst is used, should it be restricted to only part of the
column?
The five key questions effectively determine the type of system, whereas the last
three questions essentially spec@ the particular configuration of the column and the
99
M. G.Sneesby, M.O. Tadiand R. Datta
extent of recovery equipment required. The first question can normally be answered
easily by considering the feeds available. At least one of the reactants is often only
available as a component in a mixture with one or more non-reacting species, and it
would be too costly to separate the reactant. The answer to the second question will
depend on the reaction thermodynamics. If the reaction equilibrium greatly favours
the product, then there is little or no incentive to supply a surplus of one reactant as
this only adds to the recovery load. However, ifthe reaction is equilibrium limited, it
is normally preferable to drive the reaction further towards the product by
introducing an excess of one reactant. It is, of course, desirable to use stoichiometric
ratios but if an excess is to be used, then one of the reactants must be selected (third
question). Here, reaction kinetics must be considered as the effects of a reactant
excess on reaction rate might be different for each reactant. The relative cost will
also be important. The fourth question can, again, be answered by considering the
reaction thermodynamics. The answer to this question helps determine the extent
and type of recovery/purification equipment required.
The fifth question is slightly harder to answer than the first four but is generally a
function of the relative volatility between the two pairs of components, light reactantheavy reactant and heavy reactant-product, and the azeotropic behaviour, if any. If
the relative volatility between the two reactants is much higher than the relative
volatility between the heavy reactant and the product, then the heavy reactant is
probably best recovered in the bottoms stream with the product. This arises from the
need to have both reactants present in the liquid phase in the reaction zone and the
desire to minimise product in the overheads. Forcing the heavy reactant and product
to opposite ends of the column when the relative volatility is low results in
intermediate conditions that drive the light reactant into the vapour phase and also
increase the amount of product in the overheads. Conversely, if the relative volatility
of the two reactants is low, then it is preferable to recover them together in the
overheads stream. Not only does this reduce the load on downstream purification
equipment but it ensures that both reactants occur in the liquid phase in the reactive
area of the column. If a minimum boiling point azeotrope is present, the overheads
product will contain at least the components of the azeotrope. Similarly, if a
maximum boiling point azeotrope is present, the bottoms product will contain at
least the components of the azeotrope.
The answers to these five questions are summarised in Table 6. Configurations 1
to 10 are all for ternary systems; configurations 1 to 4 and 11 to 14 are all for
Ten-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditionsfor High Conversions
stoichiometric systems and so on. In all, there are twenty possible configurations,
each with its own particular characteristics. The essential characteristics of each
system are summarised in Table 7. For example, configuration 1 (ternary,
stoichiometric system with complete conversion of both reactants with no excess
present) produces negligible overheads product from an RD column and has a
reaction zone that favours both reactants in the liquid phase. Generally only one
configuration is appropriate for a specific reaction system although several are
possible, and consideration must be given to the relative importance of separation
versus reaction for the system concerned.
Some modelling and simulation results have recently been published [ 121 on the
same system that was originally used to patent the catalytic distillation process for
MTBE synthesis [13]. That system is essentially quaternary (two reactants, one
product and several similar inerts), and has stoichiometric feed with incomplete
conversion (approximately 83%). Unreacted methanol, the heavier reactant, is
recovered in the distillate product. This system corresponds to configuration 12 in
Tables 6 and 7. The sigruficant features of this system are a medium-warm reaction
zone and the need to separate unreacted methanol from a mixed butanedbutylenes
stream for recycling.
Most industrial units currently producing MTBE with RD technology are also
essentially quaternary systems, and use a slight excess of methanol to promote higher
conversions (up to about 99.5%). Unreacted methanol is mostly recovered overhead
so that the bottoms product continues to be high purity MTBE, although as the
excess of methanol increases, more methanol is recovered in the bottoms stream.
This system corresponds to configuration 19 in Tables 6 and 7. The main
differences between this configuration and that used in the original MTBE patent, is
that the presence of inerts (with similar volatility to the light component) will
sigruficantly cool the reaction zone. With only a small excess of methanol, the extra
reactant will be almost completely reacted to MTBE. Systems with a large excess of
methanol are more likely to conform to configuration 20 in Tables 6 and 7 so that
both vapour-liquid t r a c and reaction zone temperatures are minimised.
101
M.G. Sneesby, M.O. Tad6 and R. Datta
102
L+H
L
N
N+L+H
N+L
N+L
N+H
N
med
larae
-
small
med
med
rned
large
large
med
large
large
8
9
10
11
12
13
14
15
16
17
18
Key:
20
19
large
L
large
7
med
med
med
large
med
rned
med
med
H + prod
med
prod
H +prod
prod
H + prod
prod
prod
prod
H + Drod
H + prod
Drod
Prod
H + prod
med
large
H + prod
prod
Prod
Bottoms
Composition
prod
Prod
Cool
med
cool
H + prod
H
L+H
L
v. cool
cool
rned
v. cool
med
warm
L
H + prod
H + prod
L+H
H
H + Drod
v. cool
med
L+H
L
med
cool
v. Cool
warm
Rxn Zone
Temperature
med
warm
H
L+H
L
H + prod
Rxn Components
in the Rxn Zone
L+H
H + prod
UH separation
Some product lost overhead
Hlprod separation
Low rxn temps reject H
UH separation
Unfavourablereaction equilibria
Hlprod separation
More separation stages required
H/N+L separation
UN+H separation
Hhrod separation
L& rxn temps reject H
UN separation
Low rxn temps reject H
HIN separation
Hlprod separation
Comments and/or
Potential Difficulties
Essentially one product only
Some product lost overhead
Unfavourable reaction equilibria
H/prod separation
Low rxn temps reject H
More stripping stages required
Unfavourablereaction equilibria
H/prod separation
UN+H separation
v. cool
UN separation
Low rxn temps reject H
large
N+ L+ H
med
prod
H + prod
med
H/N+L separation
med
N+L
large
H + prod
H
cool
Hlprod separation
1 Components: 'L' is light reactant; 'H' is heavy reactant; 'N' is non-reacting component; 'prod' is product.
2 Descriptions of flow Ate: 'smalr is 040%;'med' is 50-100%; 'large' is greater than 100%of the feed rate of the primary reactant.
N+ L+ H
N+L
med
L+H
nil
6
med
med
med
L
L
H
small
med
med
3
4
5
Bottoms
Rate
rned
med
L+H
Overbeads
Composition
Ovhds
Rate
nil
med
Configuration
Number (ex Table 6)
1
2
Table 7. Characteristics of the various reactive distillation confgurations.
%2.
25
2.
-la
2 s
3s3 .g.
G. y
?2 C'g
qs
O h
3 Y
-5
&S
95
23
g.2
Y O
fg
-
a
h
2
ir
9
z!?
3
2
M. G.Sneesby, M.O. Tadi and R. Datta
As with MTEIE, the light reactant in the ETBE reaction is generally only
available as a mixed stream, so that configurations 1 to 10 can be discounted for the
typical ETBE system. The ETBE reaction is equilibrium limited so that an excess of
one component is required to achieve high conversions (discounting configurations
11 to 16). Ethanol is the preferred reactant to use in excess as it can be obtained
more easily as a single component and high isobutylene to ethanol ratios have been
found to result in abnormal catalyst reaction (discounting conf&urations 17 and 18).
A minimum boiling point azeotrope, containing around 1% ethanol, forms between
ethanol and the r a n a t e so that for other than a very low ethanol excess, unreacted
ethanol will mostly be recovered in the bottoms product (discounting configuration
19). Increasing boil-up in an attempt to recover ethanol in the overhead stream
merely creates reaction zone conditions that tend to retard the ETBE reaction. Thus,
the configuration that appears to be most suitable for ETBE synthesis is number 20.
Considering Table 7 and configurations 17 to 20 in more detail, it can be seen
that configuration 17 requires a difficult and costly separation of isobutylene from
non-reactive butanes and butylenes (some slightly heavier, some slightly lighter than
isobutylene) and ethanol (heavier than isobutylene), and limits the ethanol excess
that can be used. Configuration 18 requires a similar difficult separation and
possibly also a separation of ethanol from ETBE, although it allows much higher
ethanol excesses. However, the reaction composition will be predominantly
isobutylene with little ethanol so that ethanol availability could still limit the
reaction. Configuration 19 must also be rejected as it creates a warmer reaction
zone, unfavourable reaction zone composition (too little of the light reactant) and
restricts the ethanol excess due to the azeotrope between ethanol and the r a n a t e .
This confirms that configuration 20 from Tables 6 and 7 is indeed the best choice for
ETBE systems. Configuration 20 has one further advantage over configuration 19,
the energy requirements and vapour-liquid traflic are lessened by recovering excess
ethanol in the bottoms stream.
The high relative volatility between ethanol and isobutylene makes it difficult to
achieve conditions that allow bcth reactants to be present in signiscant
concentrations in the liquid phase at the reaction zone and be recovered together in
the overheads, regardless of azeotropic behaviour. A low reboiler duty prevents
ethanol from reaching the rectification zone and recycles insufficient ethanol to the
reaction zone, thereby suppressing the forward reaction. A high reboiier duty drives
104
Tert-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditionsfor High Conversions
isobutylene from the reaction zone and raises the reaction zone temperature, thereby
reducing the equilibrium constant. The column pressure can be reduced to cool the
reaction zone and control the equilibrium constant but reboiler duty must still be
optimised and the effect of pressure on phase equilibrium must also be considered.
Compared with MTBE, the more restrictive thermodynamics for ETBE combined
with higher boiling points of the components and lower activity coefficient of ethanol
compared with methanol require more careful control of the reaction zone to achieve
the same isobutylene conversions, and reaction considerations may take precedence
over separation considerations. Although not conclusive, the evaluation presented
here demonstrates some of the interactions between phase and chemical equilibrium,
the range of possibilities for operating conditions, and the dependence of system
configuration on the reaction thermodynamics. Slight changes in the reaction
properties between the two systems may warrant a different operating configuration.
It follows that ETBE specific equipment is preferable for ETBE production, although
high conversions are still attainable in modified MTBE units.
Downstream Recovery Equipment
Assuming h4TBE is produced in a column that conforms to configuration 19, two
further columns are required to integrate the RD column into the refinery. The
overheads product from the RD column contains primarily methanol and rfinate in
a proportion close to the azeotrope (approximately 10% methanol) [14,17]. From
this mixture, methanol must be recovered to be recycled and raffinate must be
produced relatively free from methanol. The azeotrope between methanol and
raffinate prevents a second distillation column from being used to recover the
methanol. A liquid-liquid extraction column, using water to extract the methanol
from the raffinate product, is used instead. This then requires one further column (a
simple fractionation column as there is no azeotrope between methanol and water),
to remove water from the methanol to be recycled. If a small amount of methanol in
the MTBE product is not tolerable, a third recovery tower will be required to remove
methanol from MTBE.
With respect to an ETBE unit, the azeotrope between ethanol and raffinate
contains only around 1% ethanol [14,17]. Therefore, the overheads product from a
RD column based on configuration 20 contains littie ethanol (and virtually no
ETBE) so that it would not generally require any further processing before being
105
M. G.Sneesby, M.O.Tadiand R. Datta
used in the gasoline pool. However, if this stream is to be used in an alkylation unit,
then the ethanol concentration must be reduced to ppm levels requiring another
column. Either a water extraction column (similar to that used in a MTl3E unit) or a
molecular sieve can be used. Again, if a water extraction column is used to produce
high purity raflinate, a subsequent fractionation tower is required to remove water
from the ethanol to be recycled. Operating the ethanol-water fractionation tower at
low pressure can minimise the ingress of water to the process by reducing the water
content of the ethanol-water azeotrope. However, recycled ethanol makes up only
around 10% of the ethanol feed to the RD column so that some water in the
overheads of the ethanol-water fractionation tower is tolerable.
The bottoms product from the RD column contains around 9 mol% ethanol
(4wt%) but as ethanol is also an effective oxygenate, this is not necessarily a
problem. If high purity ETBE is desired, then another separation will be required.
This separation can be achieved in a simple fractionation column with low energy
requirements although the overhead product will be a binary, minimum boiling point
azeotrope between ETBE and ethanol [ 141. Therefore, some ETBE will be recycled
with circulating ethanol, thereby slightly increasing investment and operating costs.
However, as greater than 90% of ETBE fed to this column will leave with the
bottoms product (high purity ETBE) this shortcoming is considered acceptable.
In summary, a h4TBE unit will require at least two recovery towers, and possibly
three. An ETBE unit could conceivably operate with no recovery towers and also
requires a maximum of three columns to produce a pure ether and pure raflinate
product. A process water stream is also required for an MTBE unit but not
necessarily for an ETBE unit, unless a water wash is needed prior to reaction
(dependent on contaminant levels in the feed stream, especially salts).
Effects of Some Key Operating Variables
The following is not intended to be a complete analysis of the effects of the various
operating criteria for reactive distillation systems but to highlight some of the
interactions and considerations not present in conventional distillation or reaction
systems. Simulation studes are continuing to complete these analyses.
1. Pressure: The c o l m pressure signrficantly affects both chemical and vapour-
liquid equilibrium via the composition profile and the stage-to-stage temperatures.
106
Tert-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
As pressure (and, therefore, temperature) is increased, the composition and rate
constant become more favourable for reaction but the equilibrium constant becomes
less favourable. For mixed systems (e.g. hydrocarbon, ether, alcohol), the bubble
point c w e s are not parallel and can even cross. This is the case for both the MTEtE
and ETBE systems (see Figure 4). Another constraint exists in these cases, the
system must be operated at a pressure above the intersection of the bubble point
curves so that the ether can be concentrated in the bottoms product and not recycled
to the reaction zone. An optimum pressure must be determined taking into account
the overall conversion, reaction rates, composition profile and separation.
2. Multiple Feed Points: Since the column is in both chemical and vapour-liquid
equilibrium, reaction will be most favoured at one point in the column. Below thls
point, the VLE results in higher concentrations of product (heaviest component) and
a higher temperature. Thus,the equilibrium ratio of components will be higher but
the equilibrium constant will be lower. If the components are also close to chemical
equilibrium at that point, there will be essentially no reaction below that point.
Therefore, if reaction rates are high enough, it is preferable to conduct the reaction
on one equilibrium stage only. If this is not possible, multiple feed points should be
used to improve the distribution of reactants in the reaction zone.
180
160
G
0)
=2
al
120
3
* 100
2
a5
I-
80
60
40
100
200
300
400
500
600
700
800
900
loo0
Pressure (kPa)
figure 4. Bubble point curves for MTBE, methanol, ETBE and ethanol [ I 7J.
107
M. G.Sneesby, M.O. Tadk and R. Datta
3. Feed Temperatures: A low feed temperature increases conversion slightly by
cooling the reaction zone and keeping equilibrium constants high. However, a cold
feed will have a detrimental effect on separation and can even create a pinch point if
there are only a few separation stages. More catalyst will be required due to the
slower reaction rate.
4. Reflux Ratio: Reflux generally promotes conversion by recycling the lighter
reactant back to the reaction zone. Where the overheads rate is low, high recycle
ratios can be used to achieve overall conversions close to 100%. However, where the
overheads rate is high, increasing reflw can increase rectification separation. This
results in light reactant being rejected from the system and slightly heavier nonreactants being preferentially recycled to the reaction zone.
5. Reboiler Duty: The reboiler performs two main functions in the context of
reactive distillation, it controls the loss of unreacted reagents in the bottoms product
and it controls the reaction zone temperature (which, in turn, atrects the
composition). Where there is a large relative volatility between the two reactants, an
optimum must be established between loss of the lighter component in the bottoms
product and vaporisation of the lighter component in the reaction zone (thereby
reducing the concentration of the lighter reactant in the liquid phase). Where the
relative volatility of the two reactants is small, a much lower concentration of the
lighter reactant in the bottoms product can be tolerated.
6. Number of Separation Stages: The number of separation stages aEects the
composition of feed to the reaction zones. More stripping stages will reduce the
amount of product recycled to the reaction zone and, therefore, promote further
reaction. More rectification stages remove non-reacting light components from the
reaction zone in a quaternary system with light non-reactants, and also increase the
recycle of the reacting light component in a ternary system. (See comments under
Reflux Ratio for systems where the non-reactants are heavier than the light reactant.)
More separation stages also affect the vapour-liquid traffic in the column and,
therefore, change catalyst requirements in the reaction zone.
108
Ten-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
7. Supplementary Feed Rate: Increasing the supplementary feed rate increases the
concentration of reactants in the reaction zone and, therefore, promotes conversion.
However, where the supplementary feed is the heavier reactant, it has a secondary
effect in ‘quenching’ the tower and driving the lighter component out of the reaction
mixture. As the stoichometric ratio of reactants increases, more and more of the
excess reagent must leave the column in one of the products. This can create a
sigdicant downstream separation load as well as increasing the column hydraulic
loads.
8. Inerts in the Feed: The presence of inerts are important in both the MTBE and
ETBE systems as they reduce reaction zone temperatures and stabilise temperature
and composition changes through the column. Despite reducing the molar fractions
of the reactants, overall conversions are actually higher with some inerts in the
system due to favourable changes in stage-to-stage temperatures in the reaction zone,
thereby increasing the reaction equilibrium constant. A fourth component can also
offset the effects of azeotropes which can be detrimental in ternary systems.
However, above a certain point, comersions will be reduced by increasing inert
concentrations in the feed. The presence of a sigmficant volume of inerts in the
system also adds to energ requirements and increases the column diameter for the
same ether production rate.
Simulations of Reactive Distillation Systems
MTBE System
Simulation results are available for MTBE production using reactive hstillation
modules of both Aspen Plus [ 15) and Speedup [ 161. This system has been simulated
again on Pro/II version 4.0 [ 171 with results gwen in Table 8 as the base case. Best
estimates were used for missing data. Compared with Aspen Plus, ProAI predicted a
higher reboiler temperature, higher conversion to the byproduct, di-isobutylene
@IB), and
a lower temperature spke through the reaction zone. Each of these
properties appears to be closer to reported experimental results for the Pro/II
simulation (see Table 9). However, it is not clear whether the variations are due to
differences in the modelling algorithms between the two simulation packages, VLE
data correlations. or equilibrium expressions used. Other results were generally
similar from each of the three packages.
109
M. G.Sneesby, M.O. Tadi and R. Datta
The system in [9] was re-analysed on Pro/II, and produced a bottoms product
with 89% MTBE and an overall conversion of 88%. A second system is proposed
here, with some minor changes to the operating conditions. These changes increase
the isobutylene conversion and MTBE purity in the bottoms stream to 98.7% and
95.8%, respectively. The key changes from the original system are: (a) the overhead
pressure is increased to increase reaction rates which would otherwise be slow; (b)
the reaction zone is extended to cover three stages (this also allows more catalyst to
be used, if required); (c) both feed streams are split to allow more favourable reaction
mixtures to be maintained through the reaction zone; and (d) an excess of methanol
is used to drive the reaction towards the product. The optimised system is shown in
Figure 5 .
Table 8. Key operating conditions,feed and product data [5,6,8,17].
50%
50%
isobutylene,
50% inerts
isobutylene,
50% inerts
stage 6
stage 2
stages 3,4 8 5
stages 3,4 8 5
Optimised Case
with PreReaction
5% isobutylene,
5% methanol,
43% MTBE,
47% inerts
stages 3,4 8 5
stages 3,4 8 5
1-3
1-2
1-2
1-2
4
5-1 0
890
3-5
6-10
900
3-5
6-1 0
900
3-5
6-1 0
66
73-78
72-76
74-80
7%
7%
9%
Base
Case
Main
Hydrocarbon
Feed
Optimised
Case
Comnasitian
Main Feed Points
Supplementary
Feed Point
Rectification
Stages
Reaction Stages
Stripping Stages
Overhead
Pressure (kPag)
Reaction Zone
Temperature (‘Cf
MethanoVEthanol
ETBE
Case
9% isobutylene,
9% methanol,
36% ETBE,
46% inerts
stages 3,4 8 5
stages 3,4 8 5
900
EXCasS
Overall
lsobutylene
Conversion
MTBVETBE in
88%
97.7%
98.3%
97.7%
89%
94.8%
94.5%
89.7%
ImS
2.7%
2.84
3.77
3.60
IMw)
Reboiler Duty
2.86
(Mw)
Notes:
I10
I Stages numbered from condenser (1) to reboiler (10).
2 Condenserand reboiler duties are per 100 kmoVhr of main feed.
Tert-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditionsfor High Conversions
property
Bottoms
Temperature (‘C)
DIB in Bottoms
Product (%)
POINTS
Aspen Plus
[12,15]
Speedup
[lS,lS]
Pro/ll (ver 4.0)
[17
Experimental
[13]
114
147
126
127
- 1.0
- 1.0
2.7
- 6.0
SUPPLEMENTARY
FEED POINTS
reactive
stages
BOTTOMS
b
Figure 5. Reactive distillation qvstem for tert-butyl ether synthesis.
The third case investigates the second case with significant reaction in a prereactor (90% conversion of isobutylene). Results are generally similar to the second
case except that the reboiler duty is sigruficantly increased as less heat is generated
by the reaction. The final case shown in Table 8 is an ETBE system. This is
discussed in the next section.
M. G.Sneesby, M.O. TadP and R. Darta
ETBE System
In contrast to MTBE, few simulation results have been published in the literature for
ETBE synthesis, and these provide few comments concerning the differences
between their synthesis methods. While ETBE can be produced in the same
equipment that is used for MTBE, this does not necessarily represent the optimal
method of producing ETBE. As indicated above, the thermodynamics and kinetics
of the ETBE system are more restrictive than for MTBE and a slightly different
configuration is recommended, most notably, excess ethanol should be recovered in
the bottoms product.
A reactive distillation system designed specifically for ETBE is proposed here. A
quaternary system (ethanol, isobutylene, ETBE and light non-reactants) is assumed
with initial the concentrations of isobutylene and light non-reactants being equal.
An excess of the heavier reactant, ethanol, is used to promote the reaction and nonreacting ethanol is mostly recovered with ETBE in the bottoms product
(configuration 20 from Tables 6 and 7). Ten stages are used with catalyst present on
stages 3. 4 and 5 . as with the optimised MTJ3E cases given above. The feed to the
column is 36% ETBE (equivalent to 80% conversion in a pre-reactor). Similar to
the MTBE case where the reacting system is lughly non-ideal, the choice of VLE
correlations has a signrficant effect on the results and the UNIFAC model was used.
Compared with the MTBE system, the overhead pressure for an ETBE column
should be unchanged to counteract slower reaction rates for the ETBE system
(temperatures in an ETBE column will already be higher as individual components
are heavier). A slightly higher excess of ethanol is required to promote reaction
whch is more restricted by equilibrium than for MTBE. With these changes, a
similar isobutylene conversion is achieved. However, the bottoms purity is reduced
as unreacted ethanol is essentially recovered in the bottoms product rather than the
overhead.
Both the main and supplementary feeds are fed to the column at three separate
points (see Figure5). Multiple feed points for the supplementary feed helps to
maintain high ethanol concentrations in the reaction zone while a low ethanol feed
temperature (30°C) boosts conversion. The main feed was assumed to enter at the
equilibrium temperature of the pre-reactor (approximately 70°C). A relatively high
reflux ratio (RID = 10) was used to ensure satisfactory recycle of unreacted
isobutylene. The reboiler duty is set to minimise the amount of unreacted
112
Tert-Butyl Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
isobutylene lost in the bottoms product without causing excessive vaporisation of
isobutylene in the reaction zone. The overall conversion of isobutylene is 97.7%.
The temperature and composition profiles are shown in Table 10. Reaction rates are
not given as sufficient catalyst is assumed to be present for reactions to go to
equilibrium. This assumption is considered valid as both the MTBE and ETBE
reaction are fast and approach equilibrium with only small amounts of catalyst.
The reboiler and condenser duties are lower than for the optimised MTBE case
despite operating temperatures being slightly higher. This is because less energy is
required as latent heat, as excess ethanol is primarily recovered in the bottoms
product. Hydraulic loading of the column is also similar to the MTEiE system. DIB
formation is predicted to be slightly higher at 0.26% of the bottoms product.
However, this is not necessarily a disadvantage as DIB has a high octane and low
RVP. In fact, the reaction of isobutylene to form DIB or iso-octane is one of the key
reactions in allcylation.
Table 10. Composition Profile and Equilibrium Constants in an ETBE RD Column
16,I71.
Tray
1
2
3
4
5
6
7
8
9
10
Temp
(OC)
40
72
74
77
80
92
113
131
142
153
Mol%of
lsobutylene
1.8
1.7
1.6
2.1
2.6
1.8
0.9
0.3
0.1
0.05
Mol%of
Ethanol
1.7
2.2
3.0
5.2
10.7
21.8
33.4
33.8
22.1
8.7
MoI%of
ETBE
0.1
0.8
4.8
8.5
13.5
25.5
41.4
56.2
73.9
89.7
Equilibrium
Constant
74
17
15
13
10
6.8
3.0
1.6
1.1
0.8
(w
Ratio of
Reactants
0.3
3
15
13
10
23
69
300
1500
7700
Conclusions
Oxygenates are frequently described as the fastest growing chemicals of the nineties.
Now that the US Clean Air regulations have come into force, demand for
oxygenates, including the renewable and semi-renewable oxygenates, is expected to
increase further. ETBE has the potential to become a major oxygenate in the future,
mainly because it can be produced from a renewable resource but also because of its
higher octane number and lower RVP. The development of more economical
113
M . G.Sneesby, M.O. Tad6 and R. Datta
technology to produce renewable ethanol and continuing pressure on the methanol
market is already reducing the price differential between methanol and ethanol.
Combined with improvements in operating conditions, ETBE could conceivably
surpass MTBE in importance in the future.
Very high isobutylene conversions (to either MTBE or ETBE) are diflicult to
achieve in conventional reaction systems and require multiple reaction and
separation operations. Reactive distillation can be used to boost conversion over
more traditional processing schemes at a relatively low capital cost. A wide range of
reactive distillation configurations are possible and the choice of a configuration is
highly dependent on the reaction system and VLE properties.
RD technology has been used in MTBE production for several years. Although
similar systems can be used for ETBE, a slightly different operating configuration is
preferred to handle the more restrictive thermodynamics. Very high conversions (up
to 99%) are possible using expanded versions of the configurations presented here.
Further work is required to explore these configurationsin more detail.
Nomenclature
a
D
DIE
EtOH
IB
k
Kbs
MeOH
MON
PPm
R
RON
RW
Y
114
component activity
distillate rate
di-isobutylene (isosctene)
ethanol
isobutylene
reaction rate constant
equilibrium constant in terms of activities
methanol
motor octane number
parts per million
reflw rate
research octane number
Reid vapour pressure &Pa)
activity coefficient
Tert-Bury1Ethers - A Comparison of Properties, Synthesis Techniques and
Operating Conditions for High Conversions
References
1.
2.
3.
4.
5.
6.
7.
Piel, W.J. and Thomas,R.X. 1990. Oxygenates for Reformulated Gasoline.
Hydrocarbon Process., July, 68-73.
Rock, K. 1992. TAME:Technology Merits. Hydrocarbon Process., May, 86-88.
Peaff, G. 1994. Court Ruling Spurs Continued Debate Over Gasoline
Oxygenates. Chem. Eng. News, 26 September, 8-13.
Kitchaiya, P. and Datta, R. 1995. Ethers from Ethanol 2. Reaction Equilibria of
Simultaneous tert-Amy1 Ethyl Ether and Isoamylene Isomerisation. Ind. Eng.
Chem. Res., 34(4), 1092-1101.
Zhang, T. and Datta, R. 1995. Integral Analysis of Methyl tert-Butyl Ether
Synthesis Kinetics. Ind. Eng. Chem. Res., 34(3), 730-740.
Jensen, K.L. and Datta, R. 1995. Ethers from Ethanol 1. Equilibrium
Thermodynamic Analysis of the Liquid Phase Ethyl tert-Butyl Ether Reaction.
Ind. Eng. Chem. Res., 34(1), 392-399.
Rihko, L.K. and Krause, A.O.I. 1995. Kinetics of Heterogenously Catalysed
tert-Amy1 Methyl Ether Reactions in the Liquid Phase. Ind. Eng. Chem. Res.,
34(4), 1172-1180.
8.
9.
10.
11.
12.
13.
Fite, C., Iborra, M., Tejero, J., Izquierdo and J.F., Cunill, F. 1994. Kinetics of
the Liquid Phase Synthesis of Ethyl tertButy1 Ether (ETBE). Ind. Eng. Chem.
Res., 33(2), 581-591.
Brockwell, H.L., Sarathy, P.R. and Trotta, R 1991. Synthesize Ethers.
Hydrocarbon Process., September, 133-14 1.
Buzad, G. and Doherty, M.F. 1995. New Tools for the Design of Kinetically
Controlled Reactive Distillation Columns for Ternary Mixtures. Comput.
Chem. Eng., 19(4), 395-408.
Espinosa, J., Aguirre, P. and Perez, G. 1995. Some Aspects in the Design of
Multicomponent Reactive Distillation Columns Including Nonreactive Species.
Chem. Eng. Sci., 50(3), 469-484.
Abufares, A.A. and Douglas, P.L. 1995. Mathematical Modelling and
Simulation of an MTBE Catalytic Distillation Process Using Speedup and
Aspen Plus. Chem. Eng. Res. Des., 73(1), 3-12.
Smith, L.A. 1980. Catalytic Distillation Process and Catalyst. Eur. PatentAppl,
EP8860.
115
M. G.Sneesby, M.0.Tadi and R. Datta
14. Gmehling, J., Menke, J., Krafczyk, J. and Fischer, K. 1994. Azeotropic Data,
Part 1. VCH,Weinheim, Germany, 127, 136,360,375,698-700.
15. Aspen Technology Inc. 1993. The AspenPfus User’sManuaf,Cambridge,
Massachusetts,USA.
16. Aspen Technology Inc. 1993. The Speedup User’sManual, Cambridge,
Massachusetts, USA.
17. Simulation Sciences Inc. 1994. Pro/..I Keyword Input Manual, Brea, California,
USA.
Received: 21 June 1995; Accepted afrr revision: 31 August 1995.
116
Документ
Категория
Без категории
Просмотров
4
Размер файла
1 590 Кб
Теги
synthesis, properties, butyl, ethers, high, operating, tert, conditions, comparison, conversion, techniques
1/--страниц
Пожаловаться на содержимое документа