Int.J.

Applied Thermodynamics, ISSN 1301-9724

Vol.4, (No.2), pp.85-92, June-2001

Exergy Analysis of a Reactive Distillation MTBE Unit∗

R. RIVERO# & M. GARCIA

Instituto Mexicano del Petróleo
Eje Central Lázaro Cárdenas 152
México D.F. (MEXICO)
Tel: (5) 333-8427, Fax: (5) 333-8067
E-mail: rrivero@imp.mx

Abstract
In this paper application of exergy analysis to the reactive distillation system of a
Methyl TerButyl Ether (MTBE) production unit of a crude oil refinery is presented. In a
refinery, the MTBE is obtained from methanol, and butanes (isobutylenes) produced in
the fluidized catalytic cracking (FCC) unit. The reactives (isobutylenes) after
purification are introduced to the main reaction system, the products of which are sent
to the reactive distillation system to complete the reaction. A top rectification section, a
bottom stripping section and a medium reaction section, compose the reactive column.
The results of the exergy analysis of the unit indicate that the main exergy losses (about
63%) of the MTBE plant occur in the reactive distillation system, particularly in the
distillation column itself and in its associated condenser. A detailed exergy analysis of
the system, is presented in this paper.
Key words: exergy analysis, reactive distillation, MTBE, improvement potential

1. Introduction operation costs due to high reaction yields, and

The objective of the work presented in this
paper is to establish the optimal operating
conditions of the reactive distillation column of a
Methyl TerButyl Ether (MTBE) production unit,
and to evaluate its main exergy performance
parameters in order to determine the operating
conditions minimizing exergy losses in the
reactive distillation system.
Gasoline mixtures have been reformulated
incorporating ethers such as MTBE, which
increases the octane number, and reduces
pollutant gases emissions. The MTBE production
process is a relatively new process, which uses
reactive distillation columns. Distillation and
chemical reaction occur simultaneously in a
packed and tray column. This combination
shows important advantages over the packed-bed
reactor and over the distillation system, including
the use of the heat of reaction for the separation
of products, a relatively easy control of the
temperature profile in the catalytic section, low
low capital costs due to reduced equipment
items.
Process simulation studies help in
determining the influence of operating
parameters such as the column feed location,
methanol flowrate, and the reboiler and
condenser heat duties.
With the exergy analysis of the simulated
process it is possible (1) to establish the optimal
operating conditions producing a higher amount
of high quality products, (2) to establish the
critical equipment items with the highest exergy
losses, and (3) to evaluate the improvement
potential of the reactive distillation system.
2. Process Description
In the refinery scheme the MTBE unit is
located after the Fluid Catalytic Cracking (FCC)
unit, the obtained products are the MTBE which
is sent to the gasoline pool and the raffinate
which is sent to the alkylation unit.

∗
This paper was presented at the ECOS2000 Conference in Enschede, July 5-7, 2000
#
Author to whom all correspondence should be addressed.

Int.J. Applied Thermodynamics, Vol.4 (No.2) 85

 Section 1: Isobutylenes (C4´s) Section 3: Reactive distillation
stream washing

Methanol
FA-1

DA-2 EC-2
Washing water FA-2
to TAME unit GA-1
DA-1

EA-6

GA-3

EA-2

EA-5
EA-3

Section 4: Raffinate washing

EA-1
FA-3
Washing C4´s Return Decanted
water stream water water

DC-1 DA-3

EA-4

Section 2: Main reaction

EC-1
GA-2

MTBE to Washing Methanol- water Raffinate to

gasoline pool water to TAME unit Alkylation unit

Figure 1. Process flowsheet of the MTBE unit.

RA
15 EC-2
FA-2
3 4

GA-3
14
7
AA W1 6

DA-2 Rectification zone W2

5 8
Methanol
2 Reaction zone

Methanol + iC4´s + MTBE 1

Stripping zone Decanted Raffinate-Methanol-Water
water
13

EA-2 17 Condensate

10
16
Steam
12

MTBE to 11
cooling

Figure 2. Reactive distillation scheme

86 Int.J. Applied Thermodynamics, Vol.4 (No.1)

 The process flowsheet of the MTBE unit is
presented in Figure 1. This unit is composed by
four sections: (1) The isobutylenes or ‘reactives’
(C4’s) stream washing, in which catalyst
contaminating impurities are eliminated in
column DA-1; (2) the main reaction section in
which most of the conversion of isobutylene to
MTBE is achieved in reactor DC-1; (3) the
reactive distillation section, in which the
conversion to MTBE is completed and the
MTBE is separated in column DA-2; and (4) the
raffinate washing section in which unreacted
isobutylenes are recovered and the methanol-
water stream is separated in column DA-3 for
sending it to the methanol recovery section of the
TerAmyl Methyl Ether (TAME) unit.
Product MTBE is obtained in the reactive
distillation section, cooled and sent to the
gasoline pool. The reactive distillation system is
presented in Figure 2.
3. Exergy Analysis
The main reaction of MTBE formation is
(Yuxiang and Xien, 1992):
CH2 CH3
CH3OH + CH3 C CH3 CH3 C O CH3
CH3
Methanol Isobutylene MTBE
The formation rate is:
r MTBE = k + C IB − k − C MTBE (1)
where:
⎛ 6844 ⎞
k+ = 2.5152 * 107 exp ⎜ −
⎝
⎟ (2)
T ⎠
⎛ 11381⎞
k− = 8.2680 *1011 exp ⎜ − ⎟ (3)
⎝ T ⎠
This equation considers that there is a methanol
excess and the temperature range of the reaction
is 60 - 80 °C.
The simulation of the plant has been
conducted using the Aspen Plus simulation code
and the exergy analysis subroutines developed
by the IMP Exergy Group (Rivero, 2000a). The
dead state conditions in all calculations are T0
= 25°C and P0 =1 atm and X0 = Szargut model,
(Szargut et al., 1988).
A complete exergy analysis of the naphtha
reforming unit has been conducted in the frame
of a research project (Rivero, 2000a) using the
general methodology presented in previous
papers (Rivero et al., 1989).
Figure 3 shows the block diagram of the
reactive distillation system with all streams
involved.
Int.J. Applied Thermodynamics, Vol.4 (No.1)
Irreversible Exergy Losses (exergy
destruction) represent energy degradation in a
quantitative manner; they are calculated simply
as the difference between the total exergy input
(Exti) and the total exergy output (Exto) or as the
difference between the net exergy supplied (Exns)
and the net exergy produced (Exnp):
Irr = Exti – Exto (4)
Irr = Exns – Exnp (5)
with:
Ex ti = Ex1 + Ex 2 + Ex14 + Ex16 + W1 + W2 (6)
Ex to = Ex5 + Ex8 + Ex11 + Ex15 + Ex17 (7)
Ex ns = (Ex16 − Ex17 ) + W1 + W2 + ∇Exph (8)
Ex np = (Ex15 − Ex14 ) + ∆Ex ch (9)
It is important to mention that in this process
there is an increase in the chemical exergy of the
process streams which is a part of the net exergy
produced and a decrease in the physical exergy
of the process streams which is a part of the net
exergy supplied:
∇Ex ph = (Exph1 + Ex ph 2 ) − (Ex ph5 + Ex ph 8 + Ex ph11)
(10)
∆Exch = (Exch5 +Exch8 + Exch11) − (Exch1 + Exch2)
(11)
In addition to the Irreversible Exergy
Losses, there are Effluent Exergy Losses. This
waste is simply the sum of all exergy streams
rejected to the environment:
Efl = Ex15 (12)
Total Exergy Losses, Pex, represent total
energy degradation both internally and
externally:
Pex = Irr + Efl (13)
In order to determine how well is the
desired effect of the system accomplished, the
Effectiveness is calculated as the ratio of the net
exergy produced to the net exergy supplied:
Ex np
ε= (14)
Ex ns
Total Exergy Losses and Effectiveness are
the quantitative and qualitative measures of
energy degradation, and they can be combined to
have a more complete parameter of the
performance of the system. This parameter is
called the Improvement Potential (Rivero et al.,
1989) which provides a hierarchy of all systems
or blocks composing a process for optimization
purposes:
Pot = Irr(1 − ε) + Efl (15)
87
 Condensate Return air

17 15

Methanol 11 MTBE to cooling

2
REACTIVE
8 Raffinate-Methanol-Water
DISTILLATION
Methanol + iC4´s + MTBE SYSTEM
1 Decanted water
5

W1
16 14
W2

Steam Atmosferic air Power

Figure 3. Block diagram of the reactive distillation system.

4. Results Figure 6 gives the distribution of the

improvement potential of the different sections
Figure 4 shows the distribution of the total
of the MTBE unit..
exergy losses of the different sections of the
MTBE unit. The critical section of the plant indicated by
the exergy losses is confirmed to be the reactive
70
distillation section. It is then very important to
Effluents
Effluents 8.9 further analize this section through a parametric
60 Irreversibilities
Irreversibilities
study.
% Exergy losses

50

40
30
30 26
54.5
0.3
25
20 22
% Effectiveness

10 19.8 20
14.5
1.9 15
0 15
Isobutylenes
Isobytilenes Main
Main reaction Reactive
Reactive Raffinate
Raffinate washing
stream
stream washing
washing reaction distillation
distillation washing 10
Section
5
2
Figure 4. Exergy losses by sections 0
Isobutylenes
Isob yt ilenes Main
Mai n reacti on Reactive
React ive Raffinate
Ra ffinate

The main exergy losses of the MTBE unit stream

str eam washing
was hing reaction distillation
dist lil ation washing
was hing

Se ct i on
(about 63% of the total) are located in the
reactive distillation system, particularly in the
distillation column itself (32%) and in its Figure 5. Effectiveness in the sections.
associated condenser (36%) (Rivero, 2000a).
The main reaction section is clearly not the 70
most important one from the exergy losses 60.5
60
viewpoint, even if this section has the lowest
% Improvement potential

50
effectiveness as shown in Figure 5.
40
The reason why the main reaction section is
not the most important from the exergy losses 30
23.3
viewpoint steems from the fact that in the 20 14.6
reactive distillation section a high energy 10
1.6
degradation occurs between the reboiler and the 0
condenser for separating the products. Isobytilenes
Isobutylenes Main reaction
Main Reactive
Reactive Raffinate
Raffinate
stream
streamwashing
washing reaction distillation
distillation washing
washing
By combining the imperfection aspects
Section
measured by both the exergy losses and the
effectiveness one obtains the improvement
potential. Figure 6. Improvement potential in the sections.

88 Int.J. Applied Thermodynamics, Vol.4 (No.1)

 The reactive distillation column consists of 0.2284
58 stages; the rectification zone is located from 0.2281
stages 2 to 14 (the condenser is stage 1), the

MTBE Production
(kg/ kg reactives)
0.2278
catalytic zone is located between the 15th and
20th stages. It has been designed with three 0.2275

possible stages for methanol feed: stage 13 in the 0.2272

rectification zone and stages 15 and 17 in the 0.2269
catalytic zone.
0.2265
The main operating conditions of the MTBE
2 4 6 8 10 12 14 16 18 20 22
unit (base case) are shown in the TABLE I.
Methanol feed location
A parametric analysis of the effect of the
methanol feed location, methanol flowrate, and Figure 7. Effect of the methanol feed location.
reboiler heat duty, on the performance of the
system has been conducted in order to find the As mentioned, this feed location corresponds
optimal conditions both from the exergy also to the maximal MTBE production, but there
parameters viewpoint (exergy losses, are two interesting points in Figure 8. When the
effectiveness and improvement potential) and methanol feed is located in stage 13, a minimum
from the MTBE production viewpoint. of exergy losses is obtained but also a near-to-
the-minimum MTBE production. However,
TABLE I. MTBE UNIT OPERATING when the feed is located in stage 19, a near-to-
CONDITIONS (BASE CASE). the-minimum exergy losses is obtained and a
Parameter Stream (kg/kg of near-to-the-maximum MTBE production; stage
reactives) 19 would be an adequate feed location from both
Reactive column feed flow 1 1.1118 points of view.
rate Figure 9 shows that all possible feed tray
Methanol flow rate to the 2 0.010 locations have improvement potentials of 127.1
reactive distillation column kJ/kg of reactives with a very low fluctuation.
MTBE production 11 2.7125 When methanol is introduced at tray 20 exergy
Raffinate-Methanol-Water 8 0.8506 losses are greater than 647 kJ/kg of reactives
production with an improvement potential of 641.7 kJ/kg of
Methanol feed location in stage 14 reactives; it is at this possible feed location that
the reactive distillation irreversible exergy losses are the greatest of the
column process.
Reboiler heat duty (kJ / kg 440.09
of reactives) 137,9286

137,9284
(kJ/kg of reactives)

Condenser heat duty (kJ / kg 540.0 137,9282

Irreversibilities

137,9280
of reactives) 137,9278

Reflux ratio 9/8 0.9312 137,9276

137,9274

137,9272

4.1 Effect of the methanol feed location 137,9270

137,9268
P10 P11 P12 P13 P14 P15 P17 P18 P19
The effect of the location of the methanol Methanol feed location
feed to the reactive column on the amount of
MTBE produced is shown in Figure 7. When the Figure 8. Irreversible exergy losses for
feed is located below the catalytic zone, (stages different methanol feed locations.
20 to 22) the amount of MTBE increases and
then remains constant; this can be explained by a 127,1020

better contact between the reactives and the

127,1015
catalyst.
Improvement Potential
(kJ/kg of reactives)

127,1010
The increase in MTBE production is
127,1005
important from stages 10 to 20, so the exergy
performance parameters were evaluated for this 127,1000

range of feed locations. Figure 8 shows the 127,0995

Irreversible Exergy Losses for stages 10 to 19. 127,0990

P10 P11 P12 P13 P14 P15 P17 P18 P19
Stage 20 is not included in the figure because the
Methanol feed location
exergy losses increase dramatically (647 kJ/kg of
reactives vs. 138 kJ/kg of reactives in average for
all other feed locations). Figure 9. Improvement potential for
different methanol feed locations.

Int.J. Applied Thermodynamics, Vol.4 (No.1) 89

 In most of the possible feed locations the The improvement potential of the reactive
effectiveness is 25% as shown in Figure 10. At column represents 78% of the system’s total
tray 20, not shown in the figure, the effectiveness exergy losses. By changing the methanol flow
falls to 3.15%. rate to the reactive column, these losses could be
reduced by a better choice of operating
25,0130 conditions as shown in Figure 13.
25,0129

25,0128

25,0127
% Effectiveness

25,0126 0.495
25,0125
0.492

Improvement Potential
25,0124

(kJ/kg of reactives)
25,0123 0.488
25,0122
0.485
25,0121

25,0120 0.482
P10 P11 P12 P13 P14 P15 P17 P18 P19
0.478
Methanol feed location
0.475

Figure10. Effectiveness for different 0.470

methanol feed locations. 0.0033 0.0044 0.0054 0.0065 0.0076 0.0087 0.0098 0.0109 0.0111 0.0119 0.013

Metanol flow rate (kg/kg of reactives)

4.2 Effect of the methanol flow rate
The second variation was to change the
Figure 13. Improvement potential for
methanol flow rate supplied to the column,
keeping fixed all other design parameters and different methanol flow rates.
obtaining the following results. By varying the Figure 14 shows that when methanol flow
methanol flow rate to the reactive column rate increases, the effectiveness is reduced from
(Figure 11) the flow rate to get the maximal
25.6 to 23.8 %. Since this reduction is not very
MTBE yield is found; it is in the range of 0.011
to 0.013 kg/kg of reactives. important the average effectiveness can be
considered to be 25 %.
0.250

26,0
MTBE Production
(kg/kg of reactives)

0.245
25,5
0.240
% Effectiveness

25,0

0.235
24,5

0.230
0.0033 0.0044 0.0054 0.0065 0.0076 0.0087 0.0098 0.0109 0.0111 0.0119 0.013
24,0

Methanol flow rate (kg/kg of reactives)

23,5
0.0033 0.0044 0.0054 0.0065 0.0076 0.0087 0.0098 0.0109 0.0111 0.0119 0.013

Figure 11. Change of the methanol flow rate. Methanol flow rate (kg/kg of reactives)

As shown in Figure 12, system

irreversibilities increase with the increase of the Figure 14. Effectiveness for different
methanol flow rate because the MTBE methanol flow rates.
production also increases.
4.3 Effect of the reboiler heat duty
0.530 It is important also to establish the maximal
heat duty to be supplied to the distillation system
(kJ/kg of reactives)

0.528
at the reboiler EA-2. A parametric analysis of the
Irreversibility

0.524
reboiler heat duty gives the results shown in
0.521 Figure 15.
0.518 The MTBE yield remains constant when the
0.515
heat duty is greater than 422.3 kJ/kg of reactives
0.0033 0.0044 0.0054 0.0065 0.0076 0.0087 0.0098 0.0109 0.0111 0.0119 0.013 so it is not necessary to increase the heating
Methanol flow rate (kg/kg of reactives)
steam supply. This can also be observed from the
Figure 12. Irreversibility for different irreversibility losses point of view (Figure 16).
methanol flow rates.

90 Int.J. Applied Thermodynamics, Vol.4 (No.1)

 0.238 26

0.235 24
(kg/kg of reactives)
MTBE Production
0.231

% Effectiveness
22

0.228
20
0.225
18
0.221
16
0.218
14
0.215 381.5 395.1 408.7 422.3 436.1 463.3 476.9

381.5 395.1 408.7 422.3 436.1 449.8 463.6 476.9 Q (kJ/kg of reactives)
Q (kJ/kg of reactives)

Figure 18. Effectiveness for different

Figure 15. Parametric analysis of the reboiler heat duties.
reboiler heat duty.
5. Conclusions
The parametric analysis of the system shows
144
that the optimal methanol feed location from the
142 posible locations available is stage 17. However
if the column could be modified the optimal feed
(kJ/kg of reactives)

140
location would be stage 19.
Irreversibility

138
The optimal heat duty of the reboiler is
136
422.3 kJ/kg of reactives and the optimal
134 methanol flowrate is 0.0109 kg/kg of reactives.
132
The reactive distillation section of the
130
381.5 395.1 408.7 422.3 436.1 463.3 476.9
MTBE unit is used to increase the conversión of
Q (kJ/kg of reactives)
the reaction from 95% in the reaction section to
99% after the reactive distillation. The exergy
Figure 16. Irreversibility losses for different losses of the unit are almost doubled to obtain
reboiler heat duties. this additional 4% conversion. The convenience
of finding an alternative to the use of reactive
The improvement potential represents near distillation in the plant is presently being studied
80% of the total irreversibility losses as shown in (Rivero, 2000b).
Figure 17. Nomenclature
For a heat supply equal or lower than 422.3
C concentration
kJ/kg of reactives the effectiveness of the system
Efl effluent exergy losses
is 20% or lower which is lower than the 25%
Ex exergy
value of previous cases. However this lower
Irr irreversible exergy losses
effectiveness is still acceptable since the MTBE
k rate constant of reaction,
yield remains practically constant for any heat
(m3/kg catalyst h)
supply. On the other hand, if operation is fixed to
P pressure
have a 25% effectiveness, the total and
Pex total exergy losses
irreversible exergy losses would be very high, as
Pot improvement potential
can be deduced by comparing Figure 18 and
Q heat duty
Figure 16.
rMTBE MTBE formation rate,
(kmol/kg catalyst h)
131,5
T absolute temperature, (K)
Improvement Potential

131,0
W work
(kJ/kg of reactives)

130,5

130,0
X composition
129,5
Greek symbols
129,0

128,5 ∇ decrease
128,0
∆ increase
ε
127,5
381.5 395.1 408.7 422.3 436.1 463.3 476.9 effectiveness
Q (kJ/kg of reactives)
Subscripts
0 dead state
Figure 17. Improvement potential for
+ forward reaction
different reboiler heat duties.
- reverse reaction
Int.J. Applied Thermodynamics, Vol.4 (No.1) 91
ch chemical Rivero R., González, G., García, V., Pulido, R.,
np net produced Escárcega, C., 1989, “Exergy Analysis of a
ns net supplied Crude Oil Atmospheric Distillation Unit.”, In:
ph physical R.X. Cai & M.J. Moran, editors. International
ti total input Academy Publishers - Pergamon Press. Beijing.
to total output pp. 506-511.
References Szargut, J., Morris, D. R., Steward, F.R., 1988,
“Exergy Analysis of Thermal, Chemical and
Rivero, R, 2000a, “Optimización Exérgica de un Metallurgical Processes”, New York:
Esquema de Refinación”, Proyectos E0A-7708, Hemisphere Publishing Corporation.
E0A-7709, E0A-7442, P.00257, F.00257, Grupo
de Exergia - Instituto Mexicano del Petróleo. Yuxiang, Z., Xien, X., 1992, “Study on Catalytic
Distillation Processes, Part II. Simulation of
Rivero, R, 2000b, “Investigación y Desarrollo Catalytic Distillation Processes Quasi-
Tecnológico de Proceso de Destilación Homogeneous and Rate-Based Model”, Trans
Diabática” Grupo de Exergia - Instituto IchemE., 70(A): pp. 465-470.
Mexicano del Petróleo.

92 Int.J. Applied Thermodynamics, Vol.4 (No.1)