Separation and Purification Technology 86 (2012) 70–78

Contents lists available at SciVerse ScienceDirect

Separation and Purification Technology

journal homepage: www.elsevier.com/locate/seppur

Enhanced bioethanol dehydration by extractive and azeotropic distillation in

dividing-wall columns
Anton A. Kiss a,⇑, David J-.P.C. Suszwalak a,b
a
AkzoNobel Research, Development & Innovation, Process Technology ECG, Zutphenseweg 10, 7418 AJ Deventer, The Netherlands
b
Ecole Nationale Supérieure de Chimie de Mulhouse (ENSCMu), Mulhouse, France

a r t i c l e i n f o a b s t r a c t

Article history: The industrial production of anhydrous bioethanol requires energy demanding distillation steps to over-
Received 17 June 2011 come the azeotropic behavior of the ethanol–water mixture. In spite of the recent developments in per-
Received in revised form 7 October 2011 vaporation and adsorption with molecular sieves, the large scale production is still dominated by
Accepted 18 October 2011
extractive and azeotropic distillation as the separation technology of choice.
Available online 25 October 2011
This study proposes novel distillation technologies for enhanced bioethanol dehydration, by extending
the use of dividing-wall columns (DWC) to energy efficient extractive distillation (ED) and azeotropic dis-
Keywords:
tillation (AD). Notably, DWC is one of the best examples of proven process intensification technology in
Dividing-wall column
Azeotropic
distillation, as it allows significantly lower investment and operating costs while also reducing the equip-
Extractive distillation ment and carbon footprint.
Ethanol dehydration For both ED and AD cases a classic sequence of two distillation columns and the alternative based on
Energy savings DWC are optimized using the state of the art sequential quadratic programming (SQP) method. A mixture
of 85 mol.% ethanol is dehydrated using ethylene glycol and n-pentane as mass separating agents in an
extractive and azeotropic distillation setup, respectively. The results of the rigorous simulations per-
formed in Aspen Plus show that energy savings of 10–20% are possible for the novel process intensifica-
tion alternatives based on DWC, while using less equipment units as compared to the conventional ED
and AD configurations.
Ó 2011 Elsevier B.V. All rights reserved.

1. Introduction ethanol, and integrated lignocellulosic biomass-to-ethanol [3]. A

The shortage of fossil fuels is one of the current major concerns of
the modern society. Consequently, renewable sources of energy and
fuels are intensely investigated. In terms of biofuels, bioethanol is
considered as the most promising alternative on short and medium
terms and its use as a biofuel additive has rapidly increased
[50,3,21]. Additionally, the bioethanol production can be conve-
niently integrated with the biodiesel production [24]. A major
advantage of bioethanol over other fuel alternatives, such as hydro-
gen, is that it can be easily integrated in the existing fuel systems as a
5–85% mixture with gasoline that does not need any modification of
the current engines. The higher content of oxygen leads to a more
efficient combustion and thus reduces the carbon footprint. Also,
the raw materials for bioethanol production – such as corn, sugar
cane, wood – capture and convert the CO2 from atmosphere thus
making bioethanol a potential carbon neutral source of energy
[21,3].
Several processes are used at industrial scale to produce bioeth-
anol, such as: corn-to-ethanol, basic lignocellulosic biomass-to-
⇑ Corresponding author. Tel.: +31 26 366 9420; fax: +31 57 067 9125.
E-mail addresses: Tony.Kiss@akzonobel.com, tonykiss@gmail.com (A.A. Kiss).
common feature of these technologies is the production of diluted
bioethanol – in the range of 5-12 wt.% ethanol – that needs to be
further concentrated [48,17,14]. According to the current interna-
tional bioethanol standards, the maximum allowed water content
is 0.2 vol.% (EN 15376, Europe), 0.4 vol.% (ABTN/Resolução ANP
No. 36/2005, Brazil) or 1.0 vol.% (ASTM D 4806, USA).
Two energy demanding separation steps are required to reach
the purity target, mainly due to the presence of the well known
binary azeotrope ethanol–water (95.63 wt.% ethanol). The first step
is typically an ordinary distillation also called pre-concentration
stage that concentrates bioethanol up to the level of 92.4–94 wt.%
[50,48,17,14]. A recently proposed alternative is using cyclic distil-
lation for concentrating ethanol in an energy efficient way, with
low investments [31]. The second step is more complex and of
greater interest, as it requires further dehydration of ethanol up
to higher concentrations above the azeotropic composition. Several
alternatives are available and well described: pervaporation,
adsorption, pressure-swing distillation, extractive distillation
(ED), azeotropic distillation (AD), as well as hybrid methods com-
bining these options [22,30,29,9,17,21,48,14].
Pervaporation methods are energy efficient and have a modular
design that allows easy maintenance, as well as a smaller surface
area as compared to larger equipment such as distillation columns.

1383-5866/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.seppur.2011.10.022
 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78 71

Extractive Distillation

A B
S

B,S

A,B
A B

B,S
A,B

Distillation
Extractive DWC

Dividing Wall Column

Fig. 1. Path from conventional setup to extractive dividing-wall column (E-DWC).

Azeotropic Distillation
azeotrope E-rich
phase

azeotrope

E-rich
A,B B-rich
phase
phase

B A A,B
B-rich
phase

B A
Distillation
Azeotropic DWC

Dividing Wall Column

Fig. 2. Path from classic configuration to azeotropic dividing-wall column (A-DWC).

However, they do reach their limits in case of large scale separa-
tions [48,14]. Adsorption with molecular sieves became recently
more popular as it requires less energy than distillation. However,
the desorption step requires high temperature and/or low pres-
sure, thus leading to higher overall equipment costs [17,48,14].
Distillation methods, such as ED and AD, present relatively high
energy costs but in spite of this major drawback, they are still the op-
tion of choice in case of large scale production of bioethanol fuel
[48,14]. Usually, ED and AD are performed in a conventional
sequence of two columns, first of them separating ethanol while
the other splits water from the recovered mass separating agent
(MSA) that is recycled back.
An innovative solution to overcome the drawback of energy
intensive distillation is using advanced process intensification
and integration techniques, such as thermally coupled distillation
columns, dividing-wall columns, heat-integrated distillation col-
umns, or cyclic distillation [19,8,45,20,34–36,15,2,10–12,31]. The
Petlyuk configuration consisting of two fully thermally coupled
distillation columns [36] evolved to the practical implementation
in a dividing-wall column (DWC) that splits the middle section of
a single vessel into two sections by inserting a vertical wall in
the vessel, at an appropriate position [19,10,27]. DWC found great
appeal in the chemical process industry because it can separate
more components in a single distillation unit, thereby saving the
72 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78
Fig. 3. Residue curve map (left) and ternary diagram (right) of the mixture ethanol–water–glycol.
Ethanol
Solvent
DC1 DC2
Feed
DC1
BTM
Fig. 4. Aspen Plus flowsheet of the two-columns extractive distillation sequence.
cost of building two columns and cutting the operating costs by
using a single condenser and reboiler. In fact, using DWC can save
up to 30% in the CapEx and up to 40% in OpEx [41,18]. Several
excellent reviews and research papers were published on this to-
pic, covering the design, simulation, control, optimization and
applications of DWC [47,51,44,40,49,42,37,2,11,13,25,26,39,52].
Remarkable, the DWC technology is not limited to ternary separa-
tions alone, but it can be used also in azeotropic separations
[32,46], extractive distillation [7], and even reactive distillation
[33,23,16].
Fig. 1 illustrates the path from a classic distillation column to an
extractive DWC, while Fig. 2 shows the development from a con-
ventional system to an azeotropic DWC [52]. In this work, we
investigate the application of these novel E-DWC and A-DWC
configurations to the enhanced dehydration of bioethanol, from
85 mol.% (93.5 wt.% ethanol) to the required standard purity
(>99.8 wt.%). This particular feed stream is obtained after a pre-
concentration step by ordinary distillation that follows the produc-
tion of bioethanol in a fermentation reactor, which increases the
concentration from 5 to 93.5 wt.% ethanol. Ethylene glycol and
n-pentane are used here as mass separating agents (MSA) in an
extractive and azeotropic distillation, respectively. Unlike other
studies comparing a non-optimal classic sequence with an
optimized alternative, here we make a fair comparison between
an optimal conventional sequence and the optimized DWC alterna-
tive. Backed by a solid theoretical and computational foundation,
the sequential quadratic programming (SQP) method has arguably
become the most successful method for solving non-linearly con-
strained optimization problems [6,4]. Our results clearly show that
significant energy savings of 10–20% are possible while using less
equipment units.
2. Problem statement
Water
In order to be used as fuel or additive, bioethanol must have the
purity equal or higher than 99.0–99.8 wt.%, according to the inter-
national standards (EN 15376, ASTM D 4806). Most of the water
present in the diluted ethanol/water mixture (5–12 wt.%) from
the fermentation step is removed by simple distillation, but the
purity of the bioethanol product is limited to maximum
95.6 wt.% due to the presence of a binary azeotrope with water.
Cyclic distillation was recently described as an energy efficient
alternative for concentrating ethanol, but this method is also lim-
ited by the azeotropic composition [31]. The industrial processes
Ethylene currently used to remove water from ethanol involve pervapora-
glycol
tion, adsorption, pressure-swing distillation, extractive and azeo-
tropic distillation, or a combination of these.
The problem of all these methods is either the high energy
requirements or the high equipment costs leading to penalties in
the operating costs and long payback times. Considering the high
demand of bioethanol, new alternatives are needed to reduce these
costs. To solve this problem, we propose here two novel extractive
and azeotropic distillation alternatives based on DWC that allow
over 10% and 20% energy savings in case of E-DWC and A-DWC,
respectively. Remarkably, the energy requirements are reduced
while using a lower number of equipment units as compared to
the conventional ED and AD systems.
3. Results and discussion
This section describes the results on the conventional and DWC
alternatives for extractive and azeotropic distillation. Aspen Plus
simulations were performed using the rigorous RADFRAC unit with
RateSep (rate-based) model, and explicitly considering three phase
balances. NRTL and UNIQUAC property methods can be used due to
the presence of a non-ideal mixture containing polar components.
Both methods were successfully used in the past, practically lead-
ing to very similar results [5,28].
Note that in the pre-concentration step, the diluted ethanol
stream (5 wt.%) obtained by fermentation is conveniently distilled
to the near-azeotropic composition of 93.5 wt.% ethanol. This is
typically performed by an ordinary distillation column that re-
quires significant thermal efforts of up to 2.6 kW h/kg bioethanol,
due to the large amount of water that needs to be separated. How-
ever, the pre-concentration step is out of the scope of this paper,
since this ordinary distillation is quite trivial [14]. The feed stream
considered here is the one obtained from the pre-concentration
stage of bioethanol, and consists of a mixture of 85 kmol/h
(3915.9 kg/h) ethanol and 15 kmol/h (270.2 kg/h) water thus hav-
ing a near azeotropic composition (93.5 wt.% ethanol). The target
 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78 73

Table 1
Design parameters of an optimal two-column extractive distillation sequence.

Design parameters Value Unit

Reflux ratio DC1 0.3 kmol/kmol
Number of stages DC1 17 –
Feed stage DC1 13 –
Feed stage of extractive agent 3 –
Reflux ratio DC2 1.15 kmol/kmol
Number of stages DC2 16 –
Feed stage DC2 7 –
Feed flowrate of ethanol 85 kmol/h
Feed flowrate of water 15 kmol/h
Feed flowrate of extractive agent 190 kmol/h
Heat duty DC1 1402.43 kW
Heat duty DC2 605.39 kW
Total heat duty 2007.82 kW
Operating pressure DC1 1 bar
Operating pressure DC2 1 bar
Ethanol recovery 99.85 %
Water recovery 99.10 %
Ethylene glycol recovery 99.93 %
Purity of bioethanol product 99.80/99.85 wt.%/mol.%
Purity of water by-product 98.00/99.16 wt.%/mol.%
Purity of ethylene glycol recycle 99.98/99.94 wt.%/mol.%

[53]. The ternary mixture ethanol–water–glycol presents a single

Fig. 5. Temperature and composition profiles in DC1 and DC2 (two-columns
extractive distillation sequence).
purity for the end product was selected to be over 99.8 wt.% etha-
nol in order to comply with each and every bioethanol standard.
All the classic and novel alternatives described hereafter were
optimized in terms of minimal energy demand using the state of
the art sequential quadratic programming (SQP) method available
in Aspen Plus [1,4]. This can be linked to the minimization of the
heat duty of the sequence, constraint by the required purity and
recovery of the bioethanol product, using sensitivity analysis and
the SQP optimization tool from Aspen Plus. Several optimization
variables are used: total number of stages, feed-stage location, side
stream location, recycle streams location, solvent flowrate, reflux
ratio, liquid and vapor split.
3.1. Extractive distillation
Extractive distillation performs the separation in the presence
of a miscible, high boiling, relatively non-volatile component that
forms no azeotrope with the other components in the mixture.
Ethylene glycol (EG) remains the most common entrainer used in
the extractive distillation of ethanol–water [5,28,38], although
hyperbranched polymers [43] and ionic liquids were also proposed
binary azeotrope and no liquid phase splitting, as conveniently
illustrated by the residue curve map (RCM) and the ternary dia-
gram shown in Fig. 3.
The conventional direct sequence presented in Fig. 4 consists of
two column shells, two condensers and two reboilers. Ethylene
glycol is a high boiling solvent hence it is being added on a stage
(3) higher than the feed stage (13) of the ethanol–water mixture.
Due to the presence of the EG solvent the relative volatility of
ethanol–water is changed such that their separation becomes pos-
sible. Pure bioethanol is collected as top distillate product of the
first distillation column (DC1), while the bottom product – a
mixture of ethylene glycol and water – is being fed to the second
distillation column (DC2). High purity water by-product and ethyl-
ene glycol are obtained, respectively as top distillate and bottom
product of DC2, respectively. The EG solvent recovered here is typ-
ically recycled back to the first distillation unit (DC1). Fig. 5
presents the temperature and composition profiles along the two
distillation columns, while Table 1 lists the main designs parame-
ters for the optimal sequence.
Since both distillation columns shown in Fig. 4 operate at atmo-
spheric pressure, the use of a DWC seems to be an attractive alter-
native. Fig. 6 (left) shows the conceptual design of the proposed
E-DWC. Such a column is usually called a split shell column with di-
vided overhead section and common bottoms section [52]. In this col-
umn the solvent is separated as single bottom product, while two
distillate products are collected on each side of the wall – ethanol
and water, respectively. Since there is no off-shelf DWC unit in the
currently available process simulators, two coupled RADFRAC units
were used in Aspen Plus, as the thermodynamically equivalent of
the E-DWC. Fig. 6 (right) illustrates this decomposed flowsheet,
consisting of two column shells, two condensers but only one
reboiler. The Aspen Plus model of the direct sequence is used as
the starting point of the E-DWC simulation. The results of the
direct sequence simulation provide in fact the initial estimates
for the number of trays, feed tray locations, liquid and vapor split.
Fig. 7 plots the temperature and liquid composition profiles in
the E-DWC, while the key parameters of the optimal design are
presented in Table 2. Remarkable, the temperature difference
between the two sides of the wall is very low, less than 20 °C –
such conditions being easily achievable in the practical application.
Moreover, high purity and recovery is obtained for all three prod-
74 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78

Ethanol Water
Ethanol Water
Solvent Solvent
Prefrac DWC
Feed
Feed Vap1

Liq1

Ethylene Glycol Ethylene Glycol

Fig. 6. Scheme of the proposed E-DWC (left). Thermodynamically equivalent, decomposed flowsheet of E-DWC suitable for simulation (right).

Table 2
Design parameters of the optimized E-DWC.

Design parameters Value Unit

Reflux ratio pre-fractionator 0.27 kmol/kmol
Number of stages pre-fractionator 16 –
Feed stage pre-fractionator 13 –
Feed stage of extractive agent 3 –
Reflux ratio DWC 0.2 kmol/kmol
Number of stages DWC 20 –
Stage of the interconnection Liq1 14 –
Stage of the interconnection Vap1 14 –
Interconnection liquid flow 291 kmol/h
Interconnection vapor flow 86 kmol/h
Feed flowrate of ethanol 85 kmol/h
Feed flowrate of water 15 kmol/h
Feed flowrate of extractive agent 190 kmol/h
Heat duty pre-fractionator 0 kW
Heat duty DWC 1819.52 kW
Total heat duty 1819.52 kW
Operating pressure pre-fractionator 1 bar
Operating pressure DWC 1 bar
Ethanol recovery 99.80 %
Water recovery 99.10 %
Ethylene glycol recovery 99.90 %
Purity of ethanol recovered 99.80/99.84 wt.%/mol.%
Purity of water by-product 97.64/99.07 wt.%/mol.%
Purity of ethylene glycol (recycled) 99.98/99.93 wt.%/mol.%

Table 3
Comparison condenser and reboiler duty for extractive distillation.

Fig. 7. Temperature and composition profiles in the E-DWC (dashed line used for Direct sequence ED E-DWC
the pre-fractionator side, while continuous line is used for the main DWC side).
DC1 DC2 Total PreFrac DWC Total
Condenser 1200.62 367.29 1567.90 1175.01 205.09 1380.10
ucts of the extractive dividing-wall column: ethanol and water as duty (kW)
Reboiler 1402.43 605.39 2007.82 0.00 1819.52 1819.52
top distillates, and EG solvent as recovered bottom product.
duty (kW)
It is important noting that for the optimal design a different
number of stages was obtained on both sides of the dividing-wall,
as they are actually manipulated variables and no restriction was
applied with regard to having the same number of stages on both 3.2. Azeotropic distillation
sides of the wall. Moreover, a dividing-wall with symmetry on both
side of the wall does not systematically represent the best scheme Azeotropic distillation is carried out by adding other chemicals
to achieve the required separation [7]. to generate a new, lower-boiling azeotrope that is heterogeneous –
Table 3 presents a comparison of the reboiler and condenser thus producing two, immiscible liquid phases. In some meanings,
duties for the conventional two-column sequence versus the pro- adding an entrainer is similar to extractive distillation. However,
posed E-DWC. The specific energy requirements are calculated at in case of extractive distillation, a high-boiling mass separating
0.51 kW h/kg for ED and 0.46 kW h/kg bioethanol for E-DWC, agent is used leading to lower energy demands as compared to
respectively. Energy savings of around 10% are possible with the azeotropic distillation, as the high-boiling solvent does not have
E-DWC, as compared to ED. These savings are lower than the to be evaporated [52].
usually reported values of 25–40%, due to the fact that the conven- One of the best entrainers for bioethanol dehydration by azeo-
tional ED sequence was also optimized in this study. tropic distillation is n-pentane, as it forms a low-boiling ternary
 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78 75

Fig. 8. Residue curve map (left) and ternary diagram (right) of the mixture ethanol–water–pentane.

Dec-org
Top2

Top1

DC1 DC2
Feed
Dec-aqua

Ethanol Water

Fig. 9. Flowsheet of the conventional two-column azeotropic distillation sequence.

azeotrope with ethanol and water [14]. Note that cyclohexane was
also successfully used as an alternative ([46]). The mixture etha-
nol–water–pentane presents three binary azeotropes, one ternary
heterogeneous azeotrope, and a significant liquid phase split enve-
lope, as clearly illustrated by the residue curve map (RCM) and the
ternary diagram shown in Fig. 8.
Similar to the previous case, a conventional two-column
sequence (Fig. 9) was simulated and then optimized accordingly.
Bioethanol is separated as bottom product of the first column
(DC1), while the ternary azeotrope ethanol–water–pentane is col-
lected as top distillate that is afterward sent to a decanter. The
decanting step allows the split of the ternary azeotrope in two
liquid phases, following the LL equilibrium tie lines. The organic
phase, containing mostly pentane, is recycled back to the first
column (DC1). The aqueous phase is fed to a second distillation col-
umn (DC2), where high purity water is collected as bottom prod-
uct, while the top stream of DC2 is also recycled back to the first
distillation column (DC1).
Fig. 10 plots the temperature and composition profiles in the
two distillation columns, while Table 4 lists the optimal design
parameters. These results confirm that pentane is a feasible entrai-
ner allowing an excellent separation of ethanol–water. However,
the solvent flowrate is larger as compared to the previous extrac-
Fig. 10. Temperature and composition profiles in DC1 and DC2 (two-columns
tive distillation case, thus leading to larger equipment size in order azeotropic distillation sequence).
to satisfy the purity and recovery requirements. Moreover, the
solvent has to be completely evaporated in this case due to the The extension to the A-DWC model follows the same simulation
low-boiling azeotrope that is formed, thus making the AD alterna- and optimization procedure as for the previously described E-
tive less energy efficient than extractive distillation. DWC. Fig. 11 (left) shows a schematic representation of the pro-
76 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78

Table 4
Design parameters of an optimal two-column azeotropic distillation sequence.

Design parameters Value Unit

Reflux ratio DC1 0.12 kmol/kmol
Number of stages DC1 25 –
Feed stage DC1 (main feed stream) 11 –
Feed stage of recycle organic phase 9 –
Feed stage of recycle DC2 top stream 8 –
Reflux ratio DC2 2.14 kmol/kmol
Number of stages DC2 29 –
Feed stage DC2 1 –
Feed flowrate of ethanol 85 kmol/h
Feed flowrate of water 15 kmol/h
Feed flowrate of extractive agent 751.57 kmol/h
Heat duty DC1 5948.76 kW
Heat duty DC2 994.02 kW
Total heat duty 6942.78 kW
Operating pressure DC1 1 bar
Operating pressure DC2 1 bar
Ethanol recovery 99.70 %
Water recovery 98.50 %
Purity of bioethanol product 99.90/99.70 wt.%/mol.%
Purity of water by-product 96.30/98.50 wt.%/mol.%

posed A-DWC configuration. This is also known as a split shell col-

umn with common overhead section and divided bottoms section [52].
This alternative setup consists of a single shell, two reboilers and
only one condenser. Consequently, two bottom products are col-
lected: bioethanol on the feed side and water on the other side.
The azeotropic top stream is fed to a decanter from which the
organic phase is recycled back to the feed side, while the aqueous
phase is returned to the other side of the A-DWC column.
A thermodynamically equivalent model using a post-fraction-
ator was implemented in Aspen Plus, using two RADFRAC units
thermally coupled. Fig. 11 (right) illustrates the simulated flow-
sheet. A-DWC was initialized using the design parameters of the
direct sequence, and then it was optimized using the same meth-
odology, as previously described.
It is worth noting that the presence of a high amount of pentane
that has a lower boiling point as compared to water and ethanol
creates a higher temperature difference in the DWC design. The
temperature profile, plotted in Fig. 12, shows a larger difference
between the temperatures on the two sides of the dividing-wall,
as compared to E-DWC. Nevertheless, the temperature difference
is below 50 °C so it can still be implemented in practice without
the need of insulating the dividing-wall [11]. Larger temperature
differences would require more attention, as equivalent configura-
tions used for the simulation are thermodynamically equivalent to
a DWC only if the heat transfer across the wall can be neglected.
Table 5 provides the optimal design parameters for the A-DWC
system. The results clearly show that all products can be obtained
Fig. 12. Temperature and composition profiles in the A-DWC (dashed line used for
the pre-fractionator side, while continuous line is used for the main DWC side).
with the required purities and recoveries. Table 6 gives a compar-
ison of the energy requirements for the conventional two-column
sequence versus the proposed A-DWC. The specific energy require-
ments were calculated at 1.78 kW h/kg for the conventional AD
and 1.42 kW h/kg bioethanol for the A-DWC alternative, respec-
tively. Remarkable, although both cases were optimized, the
A-DWC still allows over 20% energy savings as compared to the
conventional AD configuration.
Note that a very recent study [46] showed that A-DWC is feasi-
ble for ethanol dehydration when using cyclohexane as entrainer,
leading to a specific energy demand of 2.19 kW h/kg bioethanol
for the optimized A-DWC. Considering the similar feed conditions
used in this work and the value of 1.42 kW h/kg bioethanol
obtained here, it can be concluded that n-pentane is actually a bet-
ter entrainer than cyclohexane for the ethanol–water separation,
leading to significant savings of up to 35%.

Dec-org Solvent Dec-org

Solvent
make-up make-up

Dec-aqua
Top1 Top1

Liq1
DWC
Vap1
Feed Feed
Dec-aqua Postfrac

Ethanol Water Ethanol Water

Fig. 11. Schematics of the proposed A-DWC (left). Equivalent A-DWC flowsheet (right).
 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78 77

Table 5 financial support given by AkzoNobel to David Suszwalak (ENSCMu,

Design parameters of the optimal A-DWC. France) during his MSc internship is also gratefully acknowledged.
Design parameters Value Unit
Reflux ratio post-fractionator 0.53 kmol/kmol
Number of stages post-fractionator 25 – References
Feed stage post-fractionator
(aqueous phase from decanter) 10 – [1] Aspen Technology, Aspen Plus, User guide – Vol. 1 and 2, 2010.
Reflux ratio DWC 0.93 kmol/kmol [2] N. Asprion, G. Kaibel, Dividing wall columns: fundamentals and recent
Number of stages DWC 35 – advances, Chemical Engineering and Processing: Process Intensification 49
(2010) 139–146.
Feed stage DWC (recycle organic phase) 15 –
[3] M. Balat, H. Balat, C. Oz, Progress in bioethanol processing, Progress in Energy
Feed stage DWC (main feed) 15 –
and Combustion Science 34 (2008) 551–573.
Stage of the interconnection Liq1 10 – [4] M. Bartholomew-Biggs, Nonlinear optimization with engineering applications,
Stage of the interconnection Vap1 10 – Springer Optimization and Its Applications 19 (2008) 1–14.
Interconnection liquid flow 116.1 kmol/h [5] C. Black, D.E. Ditsler, Dehydration of aqueous ethanol mixtures by extractive
Interconnection vapor flow 156.6 kmol/h distillation, Advances in Chemistry 115 (1974) 1–15.
Feed flowrate of ethanol 85 kmol/h [6] P.T. Boggs, J.W. Tolle, Sequential quadratic programming, Acta Numerica 4
Feed flowrate of water 15 kmol/h (1995) 1–51.
Feed flowrate of extractive agent 851.9 kmol/h [7] C. Bravo-Bravo, J.G. Segovia-Hernández, C. Gutiérrez-Antonio, A.L. Duran, A.
Heat duty post-fractionator 1475.2 kW Bonilla-Petriciolet, A. Briones-Ramírez, Extractive dividing wall column:
Heat duty DWC 4062.9 kW design and optimization, Industrial and Engineering Chemistry Research 49
Total heat duty 5538.1 kW (2010) 3672–3688.
Operating pressure post-fractionator 1 bar [8] H. Becker, S. Godorr, H. Kreis, Partitioned distillation columns – why, when &
how, Journal of Chemical Engineering (2001) 68–74.
Operating pressure DWC 1 bar
[9] C.A. Cardona, O.J. Sanchez, Fuel ethanol production: process design trends and
Ethanol recovery 99.74 %
integration opportunities, Bioresource Technology 98 (2007) 2415–2457.
Water recovery 97.40 %
[10] A.C. Christiansen, S. Skogestad, L. Liena, Complex distillation arrangements:
Purity of bioethanol product 99.90/99.70 wt.%/mol.% Extending the Petlyuk ideas, Computers and Chemical Engineering 21 (1997)
Purity of water by-product 93.60/97.40 wt.%/mol.% 237–242.
[11] I. Dejanović, L. Matijašević, Ž. Olujić, Dividing wall column – a breakthrough
towards sustainable distilling, Chemical Engineering and Processing: Process
Intensification 49 (2010) 559–580.
[12] I. Dejanović, L. Matijašević, H. Jansen, Ž. Olujić, Designing a packed dividing
wall column for an aromatics processing plant, Industrial and Engineering
Table 6 Chemistry Research 50 (2011) 5680–5692.
Comparison of condenser and reboiler duties for azeotropic distillation. [13] R.C. Diggelen van, A.A. Kiss, A.W. Heemink, Comparison of control strategies
for dividing-wall columns, Industrial and Engineering Chemistry Research, 49
Direct sequence AD A-DWC (2010), 288-307.
DC1 DC2 Total DWC PostFrac Total [14] A.K. Frolkova, V.M. Raeva, Bioethanol dehydration: state of the art, Theoretical
Foundations of Chemical Engineering 44 (2010) 545–556.
Condenser 7165.75 955.88 8121.63 12500.0 0.00 12500.0 [15] J. Harmsen, Process intensification in the petrochemicals industry: drivers and
duty (kW) hurdles for commercial implementation, Chemical Engineering and Processing
Reboiler 5948.76 994.02 6942.78 4062.9 1475.20 5538.1 49 (2010) 70–73.
duty (kW) [16] S. Hernandez, R. Sandoval-Vergara, F.O. Barroso-Munoz, R. Murrieta-Duenasa,
H. Hernandez-Escoto, J.G. Segovia-Hernandez, V. Rico-Ramirez, Reactive
dividing wall distillation columns: simulation and implementation in a pilot
plant, Chemical Engineering and Processing 48 (2009) 250–258.
[17] H.J. Huang, S. Ramaswamy, U.W. Tschirner, B.V. Ramarao, A review of
4. Conclusions separation technologies in current and future biorefineries, Separation and
Purification Technology 62 (2008) 1–21.
This study successfully demonstrated the potential use of novel [18] R. Isopescu, A. Woinaroschy, L. Draghiciu, Energy reduction in a divided wall
distillation column, Revista de Chimie 59 (2008) 812–815.
distillation alternatives based on extractive and azeotropic DWC,
[19] G. Kaibel, Distillation columns with vertical partitions, Chemical Engineering
for enhanced bioethanol dehydration. The innovative processes Technology 10 (1987) 92–98.
proposed here are technically feasible, allowing the separation of [20] B. Kaibel, H. Jansen, E. Zich, Z. Olujic, Unfixed dividing wall technology for
high purity bioethanol (over 99.8 wt.%) according to the current packed and tray distillation columns, Distillation and Absorption 152 (2006)
252–266.
standards (EN 15376, ASTM D 4806). [21] W. Kaminski, J. Marszalek, A. Ciolkowska, Renewable energy source –
The use of E-DWC and A-DWC configurations leads to over 10% dehydrated ethanol, Chemical Engineering Journal 135 (2008) 95–102.
and 20% energy savings, respectively. Remarkably, only one [22] P. Kanti, K. Srigowri, J. Madhuri, B. Smitha, S. Sridhar, Dehydration of ethanol
through blend membranes of chitosan and sodium alginate by pervaporation,
column shell is used in both cases while a reboiler or a condenser Separation and Purification Technology 40 (2004) 259–266.
is spared, as compared to the conventional extractive (ED) and aze- [23] A.A. Kiss, H. Pragt, C. van Strien, Reactive dividing-wall columns – how to get
otropic (AD) distillation configurations. Moreover, in case of more with less resources?, Chemical Engineering Communications 196 (2009)
1366–1374
extractive distillation, the use of a high-boiling solvent leads to [24] A.A. Kiss, Separative reactors for integrated production of bioethanol and
lower energy requirements as compared to the azeotropic distilla- biodiesel, Computers and Chemical Engineering 34 (2010) 812–820.
tion case, since the high-boiling solvent does not have to be [25] A.A. Kiss, C.S. Bildea, A control perspective on process intensification in
dividing-wall columns, Chemical Engineering and Processing: Process
evaporated. Intensification 50 (2011) 281–292.
Based on these results, the use of E-DWC and A-DWC for etha- [26] A.A. Kiss, R.R. Rewagad, Energy efficient control of a BTX dividing-wall column,
nol dehydration is particularly interesting in case of building new Computers and Chemical Engineering, 2011, doi:10.1016/
j.compchemeng.2011.03.024.
bioethanol plants. However, in case of revamping existing plants
[27] B. Kolbe, S. Wenzel, Novel distillation concepts using one-shell columns,
the savings in the operating costs might not entirely justify a Chemical Engineering and Processing 43 (2004) 339–346.
new investment, but a revamping to the equivalent Petlyuk config- [28] B. Kotai, P. Lang, G. Modla, Batch extractive distillation as a hybrid process:
uration should be certainly considered. separation of minimum boiling azeotropes, Chemical Engineering Science 62
(2007) 6816–6826.
[29] W.S. Ling, T.C. Thian, S. Bhatia, Process optimization studies for the
dehydration of alcohol–water system by inorganic membrane based
Acknowledgments pervaporation separation using design of experiments (DOE), Separation and
Purification Technology 71 (2010) 192–199.
[30] Y. Ma, J.H. Wang, T. Tsuru, Pervaporation of water/ethanol mixtures through
We thank Eugeny Kenig (TU Paderborn, Germany) for the very microporous silica membranes, Separation and Purification Technology 66
helpful discussions on the advanced applications of DWC. The (2009) 479–485.
78 A.A. Kiss, D.J-.P.C. Suszwalak / Separation and Purification Technology 86 (2012) 70–78
[31] V.N. Maleta, A.A. Kiss, V.M. Taran, B.V. Maleta, Understanding process
intensification in cyclic distillation systems, Chemical Engineering and
Processing: Process Intensification 50 (2011) 655–664.
[32] S. Midori, S.N. Zheng, I. Yamada, Azeotropic distillation process with vertical
divided-wall column, Kagaku Kogaku Ronbunshu 27 (2001) 756–760.
[33] I. Mueller, E.Y. Kenig, Reactive distillation in a Dividing Wall Column–rate-
based modeling and simulation, Industrial and Engineering Chemistry
Research 46 (2007) 3709–3719.
[34] Z. Olujic, B. Kaibel, H. Jansen, T. Rietfort, E. Zich, G. Frey, Distillation column
internals/configurations for process intensification, Chemical and Biochemical
Engineering Quarterly 17 (2003) 301–309.
[35] Z. Olujic, M. Jodecke, A. Shilkin, G. Schuch, B. Kaibel, Equipment improvement
trends in distillation, Chemical Engineering and Processing 48 (2009)
1089–1104.
[36] F.B. Petlyuk, V.M. Platonov, D.M. Slavinskii, Thermodynamically optimal
method for separating multicomponent mixtures, International Chemical
Engineering 5 (1965) 555–561.
[37] G.P. Rangaiah, E.L. Ooi, R. Premkumar, A simplified procedure for quick design
of dividing-wall columns for industrial applications, Chemical Product and
Process Modeling, 4(1) (2009) 7.
[38] M.A.S.S. Ravagnani, M.H.M. Reis, R. Maciel Filho, M.R. Wolf-Maciel, Anhydrous
ethanol production by extractive distillation: a solvent case study, Process
Safety and Environmental Protection 88 (2010) 67–73.
[39] R.R. Rewagad, A.A. Kiss, Dynamic optimization of a dividing-wall column using
model predictive control, Chemical Engineering Science, 2011, doi:10.1016/
j.ces.2011.09.022.
[40] B.G. Rong, I. Turunen, A new method for synthesis of thermodynamically
equivalent structures for Petlyuk arrangements, Chemical Engineering
Research and Design 84 (2006) 1095–1116.
[41] M.A. Schultz, D.G. Stewart, J.M. Harris, S.P. Rosenblum, M.S. Shakur, D.E.
O’Brien, Reduce costs with dividing-wall columns, Chemical Engineering
Progress (2002) 64–71.
[42] J.G. Segovia-Hernandez, E.A. Hernandez-Vargas, J.A. Marquez-Munoz, Control
properties of thermally coupled distillation sequences for different operating
conditions, Computers and Chemical Engineering 31 (2007) 867–874.
[43] M. Seiler, D. Kohler, W. Arlt, Hyperbranched polymers: new selective solvents
for extractive distillation and solvent extraction, Separation and Purification
Technology 29 (2002) 245–263.
[44] M. Serra, A. Espuña, L. Puigjaner, Control and optimization of the divided wall
column, Chemical Engineering and Processing 38 (1999) 549–562.
[45] P.B. Shah, Squeeze more out of complex columns, Chemical Engineering
Progress (2002) 46–55.
[46] L.Y. Sun, X.W. Chang, C.X. Qi, Q.S. Li, Implementation of ethanol dehydration
using dividing-wall heterogeneous azeotropic distillation column, Separation
Science and Technology 46 (2011) 1365–1375.
[47] C. Triantafyllou, R. Smith, The design and optimisation of fully thermally
coupled distillation columns, Transactions IChemE, 70A (1992), 118–132.
[48] L.M. Vane, Separation technologies for the recovery and dehydration of
alcohols from fermentation broths, Biofuels, Bioproducts and Biorefining 2
(2008) 553–588.
[49] S.J. Wang, D.S.H. Wong, Controllability and energy efficiency of a high-purity
divided wall column, Chemical Engineering Science 62 (2007) 1010–1025.
[50] O.P. Ward, A. Singh, Bioethanol technology: developments and perspectives,
Advances in Applied Microbiology 51 (2002) 53–80.
[51] E.A. Wolff, S. Skogestad, Operation of integrated three-product (Petlyuk)
distillation columns, Industrial and Engineering Chemistry Research 34 (1995)
2094–2103.
[52] O. Yildirim, A.A. Kiss, E.Y. Kenig, Dividing wall columns in chemical process
industry: a review on current activities, Separation and Purification
Technology 80 (2011) 403–417.
[53] G. Yun, Z. Lianzhong, X. Yuan, W. Geng, J. Ji, Selection of ionic liquids as
entrainers for separation of (water + ethanol), The Journal of Chemical
Thermodynamics 40 (2008) 1248–1252.