Energy 156 (2018) 468e480

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Improvements in fermentation and cogeneration system in the

ethanol production process: Hybrid membrane fermentation and heat
integration of the overall process through Pinch Analysis
Carolina Lopez-Castrillon a, Juan Alvaro Leon a, Milagros Cecilia Palacios-Bereche a,
Reynaldo Palacios-Bereche a, *, Silvia Azucena Nebra a, b
a
Centre of Engineering, Modelling and Social Sciences, Federal University of ABC (CECS/UFABC), Rua Santa Adelia 166, CEP 09210170 Santo Andre, SP, Brazil
b
Interdisciplinary Centre of Energy Planning, University of Campinas (NIPE/UNICAMP), Rua Cora Coralina 330, PO Box 6166, CEP 13083896 Campinas, SP,
Brazil

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

Article history: The incorporation of an alternative technology for the fermentation process, and an improved cogene-
Received 5 September 2017 ration system in the first-generation ethanol production process were evaluated; and a heat integration
Received in revised form analysis was performed as well. The hybrid membrane fermentation, the alternative technology, was
27 April 2018
considered as a non-conventional operation, and its impact on the overall energy consumption of the
Accepted 13 May 2018
Available online 14 May 2018
integrated process was estimated. The improved cogeneration system contemplated a supercritical cycle,
and the increase in surplus electricity was also evaluated assuming a mixture of bagasse and sugarcane
straw as fuel. The heat integration of the process streams was performed applying the Pinch Analysis in
Keywords:
Ethanol
order to determine the targets for minimum external heating and cooling. Furthermore, the integration
Heat integration of the multiple-effect evaporator of sugarcane juice, to the process, was optimised based on the heat
Hybrid fermentation integration analysis results. Moreover, the results showed a possibility of generating a surplus electricity
Sugarcane of 138 MW (275 kWh/t of cane) with the new technologies applied.
© 2018 Elsevier Ltd. All rights reserved.

1. Introduction and the bioelectricity produced from cogeneration systems, are

A sustainable future, regarding energy utilisation and resource
conservation, depends on the increase of the participation of
renewable energy in the global energy matrix, especially in
developing countries. Such increased participation can help in
extending fossil fuel reserves, besides helping with the process of
reducing greenhouse gas emissions, and enable better energy se-
curity on a global scale. The use of environmentally sustainable
technologies is necessary to achieve these goals; however the
biggest challenge of these technologies is that they also need to be
economically feasible. Within the economic development scenario,
which is part of a sustainable development, the concepts of energy
efficiency and process optimisation are fundamental in the pro-
cesses that use renewable energies [1].
In this context, the sugarcane has been one of the main crops
used to produce energy vectors from renewable sources in the last
years. In this industry, the alcohol fuel (ethanol), an alternative fuel;
* Corresponding author.
products with great capacity of expansion. Although this sector
used to be low-efficient in terms of energy use, the concern
regarding energy efficiency, as well as the treatment and disposal of
waste, is growing nowadays [1].
It is a fact that the traditional ethanol production process has
many energy deficiencies that result in an excessive energy con-
sumption, mainly, in the integration of reaction and separation
areas. These areas are mainly composed of three operations: (i)
fermentation, (ii) simple distillation and (iii) dehydration. However,
these energy deficiencies can lead to a greater opportunity for
optimisation and reduction in energy consumption [2].
In ethanol distillation, the energy consumption depends
explicitly on the ethanol concentration in the feed wine. Conse-
quently, there is an effect on the energy consumption due to the
way in which fermentation process is conducted. Still, there are
several factors that affect the alcoholic fermentation, such as the
ethanol produced in the fermentative medium, which has an
inhibitory effect on the cell growth of yeast. In this context, one
possible solution is to partially change the structure of the

https://doi.org/10.1016/j.energy.2018.05.092
0360-5442/© 2018 Elsevier Ltd. All rights reserved.
 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
fermentation and separation areas, in the ethanol production pro-
cess, so that the ethanol is continuously extracted as it is produced
in the fermentation. This can be achieved by implementing per-
vaporation membrane technology in the conventional fermenta-
tion process, which would result in a positive impact on the energy
consumption in the process.
Taking into account the sugarcane plantations, in recent years,
the sugarcane industry has undergone changes such as a drastic
reduction of harvesting burning of sugarcane plantations, and a
proportional increase of mechanised harvesting. Before the mech-
anisation, the sugarcane straw, comprised of sugarcane tops and
dried leaves, was burnt in the field to facilitate the cultivation
practices of rattoons, however, with the mechanised harvesting
increase; these residues remain on the soil surface. Some studies
and management technologies indicate that part of the straw
should be left in the cane fields with the objective of assisting in soil
conservation and water retention. The amount of straw left will
depend on the type of soil. Currently, with the research advances
and the interest in its energy potential, a part of the straw is
recovered from the field and mixed with the bagasse to be used as
fuel in the industry.
Since the bioelectricity became another important product of
the mill, the evolution of cogeneration systems became a necessity
as well, in order to increase the efficiency of electricity generation.
In this scenario, the normal way of improving cogeneration systems
consists in reducing the steam consumption of the processes by
means of adopting more efficient technologies and heat integra-
tion; besides raising the pressures and temperatures in the steam
generation process, and in the use of condensing-extraction steam
turbines, as in the case of the supercritical cycle, which represents
the most efficient systems for electricity generation based on the
Rankine Cycle. Nowadays, the cogeneration plant Avedøre II located
in Denmark, which is already in use, works with a supercritical
steam cycle using different fuels, including biomass. This fact shows
the possibility of improvement and optimisation of the cogenera-
tion systems in sugar-ethanol plants with the inclusion of super-
critical steam cycles [3].
The heat integration of the process, on the other side, allows the
improvement of the energy utilisation of the process streams. This
analysis can minimise the process steam consumption and, at the
same time, minimise the need for cooling water. It is important to
note that the use of process integration techniques, even in similar
processes, may require different designs to meet the maximum
energy recovery (MER). Each process should be studied separately,
considering local operating costs, as well as operating conditions.
At present, there are different studies related to the reduction in
energy consumption in the ethanol and sugar production process,
such as Moncada et al. [4], where the authors studied different
conversion paths from the distribution of raw materials in the
sugarcane mill, analysing the variation in the production of surplus
electricity and by-products. On the other hand, Ensinas et al. [5]
analysed the reduction in steam demand, in a sugarcane plant,
applying heat integration; and evaluated different configurations of
cogeneration systems including a BIGCC (Biomass Integrated
Gasification Combined Cycle).
Regarding process integration research and multi-objective
optimisation, several studies can be found in the literature. For
instance, Ensinas et al. [6] performed a thermoeconomic optimi-
sation in the sugarcane juice evaporation system and heater
network in the plant. Morandin et al. [7] applied optimisation
techniques to evaluate the potential for energy savings in the
sugarcane conversion process and in the on-site electricity pro-
duction, assuming bagasse as fuel in the cogeneration system. In
addition, Ensinas et al. [8] applied a multi-objective optimisation
technique using evolutionary algorithms, setting the electricity
469
production and the produced ethanol as objective functions to be
maximized in the production process of first- and second-
generation ethanol from sugarcane. Pina et al. [9] evaluated the
reduction of the process steam demand and the water usage
through heat integration. Regarding alternative technologies in the
ethanol production process and its integration through Pinch
Analysis, the studies of Palacios-Bereche et al. [10], where double-
effect distillation is assessed; Palacios-Bereche et al. [11], who
evaluated the introduction of cooling fermentation and vacuum
extractive fermentation; and Palacios-Bereche et al. [12] who
evaluated different extraction systems for sugarcane juice, are
worth mentioning.
Furthermore, the literature presents several studies that, even
though not related to the ethanol production from sugarcane, are
related to the process integration problem assessed in this study.
Regarding the study of heat integration of multiple-effect
evaporators and heat pumps, the study of Sharan and Bandyo-
padhyay [16] presented a methodology to minimise the overall
energy consumption of a system, by thermally integrating a
multiple-effect evaporator, thermo-vapour compressors, into the
background process. The method focused on the optimal location of
the thermo-vapour compressor (TVC) suction position. The authors
evaluated two cases; the first one being a corn glucose concentra-
tion system, while the second one presented a desalination system.
In the same research topic, Oluleye and Smith [18] carried out a
study where different technologies of heat upgrading, such as
mechanical heat pumps, absorption heat pumps, and absorption
heat transformers, are integrated to an existing process through a
Mixed Integer Linear Program (MILP). This procedure takes into
account the interactions of the aforementioned technologies with
the associated cogeneration system, temperature of process
streams, and quantity of heat sources, as well as process economics
and the potential to reduce carbon dioxide emissions.
Furthermore, in the field of Combined Heat and Power, Total Site
Integration is presented as a method for the targeting and inte-
gration of individual processes with heating and cooling re-
quirements linked by the same utility system. Several studies have
been developed in this area, for instance, Ren et al. [14] presented a
method for targeting the cogeneration potential of Total Site utility
systems from the pressure values and heat loads of the steam
mains, using the commercial software Aspen Plus. The simulator
supplies temperature values, flow rates, and shaft powers of steam
turbines; furthermore, this study presents a procedure to estimate
capital cost, GHG emissions, and the water footprint. In another
study, Ong et al. [15] developed a method to visualise and solve
Total Site Mass, Heat, and Power Integration, using process inte-
gration and process graph techniques (P-graph), which is an
approach to visualise multi-dimensional network problems using
Mathematical Programming Optimisation. These authors applied
the method to evaluate the incorporation of alternative processes,
such as gasification of black liquor, hydrothermal liquefaction,
simultaneous scarification and co-fermentation of biomass, into an
existing Kraft Pulp Mill in New Zealand, together with a combined
heat and power system on-site. In the same field, Tarighaleslami
et al. [17] presented an alternative method for utility selection and
optimisation of Total Site Heat Integration, using cost and exergy
derivative analysis. This method can optimise both non-isothermal
and isothermal utilities in the same procedure, being the main
target the minimisation of the total annualised cost. Finally,
Walmsley [13] presented a new approach for a Total Site Heat
Integration method for the design of integrated evaporation sys-
tems including vapour-recompression based on a hybrid Total Site
Profile, which can be applied to optimise industrial case studies,
minimising the energy use and costs.
The present work aims at evaluating the incorporation of hybrid
470 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
fermentation with pervaporation membranes, as well as its heat
integration through Pinch Analysis, into the ethanol production
process, in combination with improvements in the cogeneration
system; which include the use of sugarcane straw recovered from
the field, and a supercritical steam power cycle using steam re-
heating and regenerative heat exchangers, in order to evaluate
the surplus electricity.
The novelty of this work resides in the study of the introduction
of pervaporative fermentation, and its heat integration through
Pinch Analysis, into the overall ethanol production process; eval-
uating the utility consumption and impacts on the cogeneration
system. The literature has already presented some studies related
to pervaporative fermentation, though, neither of them analyses
this technology in an integrated way with the rest of the process.
Moreover, although the state of the art already presents studies
evaluating supercritical cycles in cogeneration, together with the
ethanol production process, none of them have evaluated the use of
sugarcane straw in these systems, becoming this use of sugarcane
straw another novelty of this work.
2. Theory: technological aspects
2.1. Ethanol production process
Fig. 1 shows a flow sheet of the conventional ethanol production
process [19]. The process begins with the cleaning of sugarcane,
followed by the preparation and juice extraction system, the latter
assumed to be performed by mills. The bagasse, fibrous residue of
the sugarcane, is sent to the cogeneration system, while the raw
juice extracted is sent to the treatment stage. After that, the juice
concentration takes place in an evaporation system. The final su-
crose content in the concentrated juice should be appropriate for
the fermentation process, and it is usually around 17%e19% (19e21
Brix). During the fermentation step, ethanol is produced in a bio-
logical process lead by Saccharomyces cerevisiae yeast. The alcoholic
fermentation is affected by several factors, such as the ethanol
produced in the fermentative medium, which has an inhibitory
Fig. 1. General diagram of an ethanol production plant from sugarcane [19].
effect on the cell growth of yeast. For this reason, it is not possible to
feed the fermentation process with a substrate that is too
concentrated in sugars. Most of the Brazilian plants use the fed-
batch fermentation with cell recycle, or Meille-Boinot process;
however, there are few plants using the continuous fermentation
process. In the conventional process, the wine obtained in the
fermentation process has an ethanol content of around 6% wt.
Following that, the wine is sent to a distillation step where hy-
drated ethanol is produced with an ethanol content of around
93.5% wt. Optionally, anhydrous ethanol with an ethanol content of
99.4% can be produced in a dehydration step.
This work assumed an ethanol production plant coupled with
two new technologies: (i) pervaporation in the fermentation stage,
and (ii) supercritical cycle in the cogeneration stage. These tech-
nologies are presented in the following items.
2.2. Pervaporation coupled to the fermentation process
In the distillation step, the energy consumption depends
explicitly on the ethanol concentration in the feed wine. Conse-
quently, the interconnection between fermentation and distillation
affects the energy consumption in these two stages of the process.
Since the yeast suffers inhibition by the ethanol produced in the
fermentative medium, one possible solution would be the extrac-
tion of this ethanol as it is formed. This allows the fermentation
medium to keep a relatively low ethanol concentration, and, as a
result, an operation with a must with higher sugar concentration
(approximately 25 Brix) can be achieved.
The aforementioned solution could be achieved by imple-
menting the pervaporation membrane technology in the
conventional fermentation process [20]. Pervaporation
(permeation þ evaporation) is a type of membrane operation
applied to the separation of mixtures in liquid phase It is considered
as a technology that could be appropriate to be used in the
fermentation stage, increasing the performance of the fermenta-
tion, consequently reducing the steam consumption in the distil-
lation stage. During the pervaporation process, the substance
transferred through the membrane (permeate) changes from liquid
to vapour phase. In the case of extractive fermentation, one part of
ethanol and water (ethanol-rich portion) is removed from the so-
lution, while the other portion remains in the fermenter (reten-
tate). After the separation, the permeate is condensed for recovery
[20]. Fig. 2 shows the pervaporation scheme adopted.
According to [22], the mass transfer through the membrane is
governed by the chemical potential gradient in the two phases. In
the case of pervaporation, the chemical potential gradient is
influenced by the partial pressures of the substances in each phase.
The mass flow of the chemical species and the separation factor
Fig. 2. General diagram for hybrid fed-batch membrane fermentation H-FBMF arrange-
ment [21].
 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480 471

depend also on the affinity (solubility) of the chemical species for
the membrane material (PDMS1).
A membrane adapted to the fermenter removes the ethanol that
is generated in the reaction, shifting the biochemical equilibrium
(inhibition conditions) towards the fermentation products [23]. The
improvement of the fermentation performance has two direct
consequences on the energy efficiency of the process: (i) the
fermenter will need less cooling water per kg of ethanol produced;
and (ii) the liquid product of the fermenter will be a wine (distil-
lation feed) with a higher ethanol content. This phenomenon is
directly linked to the energy consumption in distillation. In such a
way, the distillation feed is considered as an ethanol-water solution
with an ethanol concentration higher than the one in the conven-
tional process. This condition in the incoming stream promotes a
decrease in the energy requirements in the distillation columns’
reboilers. Besides, a significant reduction in energy consumption in
the distillation was demonstrated by applying membranes in
fermentation, in comparison with the traditional process [20].
Regarding the operation, the permeate is constantly removed
and it by-passes the first set of distillation columns (A, A1 and D),
directly feeding the B-B1 column. The retentate is then recycled to
the fermentation vat. The liquid phase is discharged from the
fermentation vat at the end of the fermentation time and is nor-
mally fed at the first set of columns (column A1).
Owing to the fermentation process being operated in batch
mode, several fermenter vats are used in parallel, though, in lagged
operation. In addition, the wine from all fermentation vats is dis-
charged to a standardisation holding tank, before being sent to the
distillation sector.
previous treatments that are of great importance. For this reason,
techniques need to be used in order to remove impurities, mainly
soil that is carried with the cane and straw; and, in some cases, the
reduction of the ash content and the control of the moisture con-
tent are also necessary. Different authors affirm that straw burning,
without an appropriate treatment, represents several problems for
boilers, such as corrosion, abrasion, and deposits of molten ash.
Two of the most studied problems in the burning of straw in
cogeneration are corrosion and incrustations; however, different
authors affirm that cleaning treatments before the burning, such as
separation and cleaning operations, can mitigate these problems
[26].
The aforementioned treatments are exposed in Fig. 3a and b:
3. Methods
3.1. Modelling and simulation of the conventional ethanol
production process
Mass and energy balances in each stage of the conventional
ethanol production process were accomplished using the Aspen
Plus software. A cane rate of 500 t/h for anhydrous ethanol pro-
duction was assumed. Regarding the distillation process, a con-
ventional configuration with two sets of columns was assumed. The
first set considered the columns A, A1, and D for wine stripping,
while the columns B-B1 for phlegm stripping were considered in
the second set. This configuration can be observed in Fig. 7a.
Further information regarding process specifications and results is
available in Ref. [9].

2.3. Cogeneration systems

3.2. Modelling and simulation of the hybrid fed-batch membrane
The cogeneration system supplies the plant with electricity and fermentation and its integration with the distillation process
thermal energy in the form of process steam. It is traditionally
composed of a steam cycle that uses sugarcane bagasse as fuel. The simulation of the hybrid fed-batch fermentation H-FBMF
According to [24], Brazilian mills have an electricity self- was accomplished using the Matlab software, using the kinetic
consumption of 12 kWh/t of cane, and a mechanical power con- parameters from Atala et al. [27] for fermentation; and the
sumption of 16 kWh/t of cane that is used in the sugar extraction Maxwell-Stefan equation for pervaporation, assuming a PDMS
system and cane preparation devices. In addition, approximately membrane. Further information regarding pervaporation mem-
330 kWh/t of cane of thermal energy [24], in the form of process brane coupled to fed-batch fermentation is available in Ref. [21].
steam, are used in the production process. A simulation of the distillation process was performed in the
The parameters of live steam, in the boilers of cogeneration Aspen Plus software, introducing the results obtained from the
systems in Brazilian mills, used to present values around 22 bar and kinetic modelling of H-FBMF [21]. The UNIFAC-DORMUNT method,
300  C. However, modern mills, as well as old ones that have in Aspen Plus, was selected for properties calculation, assuming an
substituted their boilers in the last years, have adopted devices
with pressures in the range of 65 and 67 bar and temperatures
between 480  C and 520  C. Additionally, there are some few mills
with boilers at 100 bar and 530  C.
In this work, some configurations of commonly used cogeneration
systems, in the sugar and ethanol industry, were analysed; as well as
new technologies that can be used to increase the efficiency of electric
power generation, such as the supercritical steam power cycle.
The supercritical systems can be considered as an alternative to
the cogeneration in sugarcane plants. By way of illustration, the
supercritical steam cycle of the Avedøre II plant shows very satis-
factory results in biomass burning, which allow this option to be
closer to the development of the cogeneration of sugarcane mills.
A thermoelectric plant based on a steam cycle can be classified
by the temperature and pressure level at which it works. That being
so, they are usually classified according to their operating pressure
as subcritical, supercritical, and ultracritical [25].

2.4. Sugarcane straw utilisation

The use of the straw as fuel in cogeneration systems requires Fig. 3a. System flow chart for cleaning and grinding of straw.
472 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
Fig. 3b. Theoretical flow chart of the system for cleaning and grinding of straw.
Fig. 4. Conventional five-effect evaporation scheme of sugarcane juice.
Fig. 5. Configuration of the cogeneration system with condensing-extraction steam
turbines. Configuration A.
ethanol-water binary mixture.
In order to integrate the fermentation with the distillation, the
permeate was assumed to be recovered as a low-pressure (vacuum)
vapour. The permeate temperature is close to room temperature
(i.e. 298 K). Then, the low-pressure vapour is compressed and
eventually condensed [21].
Moreover, it is worth mentioning that the permeate is
constantly removed from the fermenter; so, the collection and
storing of the permeate in a holding tank, during the fermentation
time, was assumed. Then, this permeate and the wine (liquid
phase) from the fermenter are simultaneously sent to the distilla-
tion stage. In this study, the collection time of the permeate was
assumed to be 8 h (the membrane is activated 2 h after the begin-
ning of fermentation, and it is deactivated at the end of a ten-hour
fermentation). For this reason, the mass flows of wine and
Fig. 6. Configuration of the supercritical cogeneration system. Configuration B.
Fig. 7(a). Distillation scheme for the production of hydrated ethanol. Case study FBF-
200.
permeate, which enter the distillation stage, are defined as the
mass accumulated in fermenters and the accumulated mass of
permeate divided by the discharging time of the fermenter (1 h).
These calculations were performed for a reference volume of a
fermenter, and extrapolated to the available mass flow of must for
fermentation of the conventional process [9]. It was assumed that
the permeate is pressurised, condensed and fed to the B-B1 column,
while the wine is pre-heated and fed to the A column, according to
Fig. 7b [20].
3.3. Heat integration through Pinch Analysis
The first step in the heat integration procedure was the data
extraction. In this study, data for the conventional ethanol pro-
duction process were extracted from Ref. [9]; while for the case that
assumes the hybrid fermentation, data from Ref. [21] were
extracted and adjusted for this case study.
The next step was the determination of the energy targets,
which was performed through the Heat Cascade method. After that,
the construction of the Composite Curves (CCs) and the Grand
Composite Curve (GCC) was carried out.
The Grand Composite Curve (GCC) is a graphical tool for visu-
alising the heat integration problem. It is the graphical represen-
tation of the Heat Cascade and is obtained by plotting the values of
the temperature boundaries versus the heat flows across said
boundaries [28]. One of the main advantages of the GCC is the
 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
Fig. 7(b). Distillation with a permeate feed recovered from the hybrid fermentation
with membranes. Case study H-FBMF-240.
identification of the appropriate temperature level of utilities to be
used, minimising the irreversibilities of the heat exchange process
and allowing the evaluation of the use of lower cost utilities [24].
Furthermore, this tool is useful for the next step, which is the
appropriate placement of the multiple-effect evaporation system.
This analysis was performed by means of the EXCEL software.
The value of DTmin was assumed to be 10  C for process streams,
and 4  C for evaporation system streams.
Optimisation of the juice evaporation process using heat
integration:
The separation process to obtain the concentrated juice, or
syrup, occurs in a multiple-effect evaporation system that uses
exhaust steam from the cogeneration system as a heat source in the
first effect (see Fig. 4). The vapour generated in the separation
process, called vegetal vapour, is used as heat source for the sub-
sequent effects, providing a saving of steam, when compared to
simple evaporation in single stage [24].
The operating conditions of the juice evaporation process were
extracted from Ref. [9], being the main parameters the mass flow of
the clarified juice (141.5 kg/s), the initial temperature (98  C), and
the initial Brix (Brix ¼ 15). The final Brix was assumed at 21 for the
Base case, and at 25.7 for the H-FBMF case (A purity of 86% was
assumed in all cases).
In this analysis, the pre-heating of the clarified juice, prior to its
entry into the evaporation system, was included. This way, the
clarified juice temperature is raised to the saturation point of the
first effect. The simulation of the juice concentration was per-
formed using the calculations implemented in the EES® software,
and employing mass and energy balances based on real data ob-
tained from the literature.
Mass balance for the first effect:
: : :
mc1  mv  mc2 ¼ 0
: :
mc1 , Brix1 ¼ mc2 , Brix2
Energy Balance for the first effect: Hypotheses DEc ¼ 0 and
DEp ¼ 0
: : :

mc1 , hc1  mv , hv þ mc2 ,hc2 ¼ Q1
: :
m1 , hve  m2 , hcve ¼ Q1
Owing to the existence of a multiple-effect evaporation system,
the heat integration procedure was carried out by steps. The first
step adopted was the heat integration of process streams excluding
473
the streams from the evaporation system.
The second step was the accomplishment of mass and energy
balances in each effect of the evaporation system, and the ther-
modynamic profile examination of the evaporator against the
background process, that being so, each effect of the evaporator is
represented as a horizontal bar in the GCC, trying to achieve an
appropriate placement (opportunities for integration without
increasing the utility consumption).
After the placement of the effects, the mass and energy balances
in the evaporation system should be recalculated in order to update
vapour bleedings and condensate streams.
3.4. Cogeneration systems
This study assumed that 50% of the produced straw (total po-
tential) is used as fuel in the cogeneration system, while the other
50% is left in the field. It was also assumed that the straw has
different treatments before the combustion and mixing with
bagasse.
It was assumed, in this work, that the cane preparation devices
and the mills of the juice extraction system are driven by electric
motors, instead of the direct drive steam turbines commonly used
in old mills. Steam cycle options, with different pressure and
temperature levels, together with the use of back-pressure and
condensing-extraction steam turbines, were evaluated. The sys-
tems were simulated on the condition that the demand of steam
and electricity of the process is covered by the cogeneration system
that consumes a mixture of bagasse and straw as fuel for the gen-
eration of steam.
3.4.1. Steam cycle with condensing-extraction steam turbines e
configuration A
This configuration is characterised by the fact that all the
available bagasse (discounting a part for reserve and the bagasse for
filters) is burnt. The straw was assume to be used, in mixture with
bagasse, as additional fuel in the cogeneration system, as it would
allow a greater production of steam in the boiler and, consequently,
a greater amount of surplus electrical energy that could be com-
mercialised in the network.
The scheme of this cogeneration system configuration is shown
in Fig. 5. This configuration comprises a biomass boiler, steam
turbines, an electric generator coupled to a steam turbine, a
deaerator, electric pumps of boiler feed and condensate return, and
a desuperheater to control the temperature reduction of the
exhaust steam used in the process.
3.4.2. Supercritical steam cycles. Configuration B
Fig. 6 shows the configuration of the supercritical steam cycle
assumed in this study. It was adopted so as to use a mixture of straw
and bagasse as fuel, including the use of steam reheating and
regenerative heat exchangers for heating the returning condensate.
(1) This type of configuration does not yet exist in the sugar-alcohol
Brazilian industry, though; it represents the most efficient system
for electricity generation based on the Rankine Cycle [25].
(2)
Since a supercritical cycle operates at pressures and tempera-
tures above the critical point, a reheating of the steam leaving the
high-pressure turbine (TAAP in Fig. 6) is required to increase the
cycle efficiency and decrease the percentage of moisture in the last
(3) stage of the turbine. This cycle also requires a series of regenerators
to increase the temperature of the water before entering the boiler;
increasing, consequently, the system efficiency [25]. The scheme of
(4)
the cogeneration system shown in Fig. 6 has condensing-extraction
steam turbines to take advantage of all the available fuel (bagasse
and straw), representing the straw 23% (wet basis) of the total
amount of fuel. Considering the highest values practiced, the
474 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
efficiency of the boiler is set at 86% (PCI base) while the steam is
generated at 300 bar and 600  C, being reheated to 600  C after the
expansion up to 62 bar that takes place in the TAAP turbine, [25].
3.4.3. Study of the pressure and the temperature increase of the
steam generated in the boiler
The influence of the steam parameters at boiler outlet on the
cogeneration system efficiency (steam power cycle efficiency) was
also studied. Thus, the increase of the pressure and temperature of
generated steam was evaluated for three conditions. The first
condition (C1) was established as the steam generation parameters
traditionally used by the sugar-alcohol industry in Brazil, with
boilers that use only bagasse as fuel. The second condition (C2)
corresponds to the pressure and temperature used in most of the
new projects of cogeneration systems that have been introduced in
the sector. Such new projects present more efficient boilers and
energy recovery systems from the exhaust gases, such as air pre-
heaters and economisers. Different sugar and alcohol production
plants already possess this type of technology. The third condition
(C3) refers to a more advanced system, which works with steam
and reaches supercritical conditions. It is not yet used in the sugar-
alcohol sector; however, it may provide an increase in the future
electricity generation potential, as well as an increase in efficiency.
These three conditions can be observed in Table 1.
4. Case study introduction
Two cases of study were evaluated in this work. The first one
assumed a conventional ethanol production process with a fed-
batch fermentation process (FBF-200), while the second one
assumed the incorporation of hybrid fed-batch membrane
fermentation (H-FBMF-240). In the FBF-200 case, the substrate
concentration at feed (sugar concentration in the must for
fermentation) was 200 kg/m3 (21.7 Brix); and in the H-FBMF case,
the substrate concentration at feed was 240 kg/m3 (25.7 Brix).
Both cases assumed a mill capacity of 2,000,000 t of cane/y; a
crushing rate of 500 t of cane/h; season operating hours of 4000 h/
y; and a bagasse production of 277 kg/t of cane [9]. Additional data
regarding the main streams are presented in Table A1 of Annex.
Leon et al. [17] carried out a study to evaluate the variation of
steam consumption, in the distillation stage, by implementing the
pervaporation in the fermentation process. According to the
distillation process studied by these authors, Fig. 7a shows the
conventional distillation process, and Fig. 7b shows the distillation
process coupled to the hybrid fermentation with a permeate feed in
the B-B1 column.
From the aforementioned previous studies, [9] for conventional
process and [21] for hybrid fermentation; streams available for heat
integration were extracted and are shown in Table 2. In order to
simplify the analysis of the heat integration, process streams with a
heat load below 1000 kW were not included.
After the calculation of the evaporation system, the table of
streams for heat integration is updated with the streams of juice
pre-heating (before the first effect of the evaporation system),
vapour condensates, juice for sterilisation, and the streams of the
evaporation system (juice and vapours). Table A2 in the Annex
Table 1
Operating conditions depending on the pressure and temperature of the steam
generated.
Steam conditions at boiler outlet Pressure [bar] Temperature [ C]
C1 65 480
C2 120 520
C3 300 600
presents the final streams for heat integration for cases FBF-200
and H-FBMF-240.
Since the fermentation is operated in batch mode, the heat
released in the fermentation, as well as the heat from the heater H1
and the cooler C1 from the pervaporative process (Fig. 3), were not
taken into account. In addition, since heat is released from the
fermentation at low temperature (approximately 35  C), the op-
portunities of heat integration at this temperature level are scarce.
In such a way, this analysis focuses mainly on the effects of the
hybrid fermentation on the distillation step, and on the potential of
surplus electricity; taking into account the overall process.
4.1. Cogeneration system
For the cogeneration system, the following parameters were
assumed (Table 3):
Sugarcane straw is understood as green leaves and straw that
currently remain on the land after harvesting. When staying in the
field, its function is to serve as protection for the soil in the process
of nutrient replenishment and in pest control. However, it would be
possible to use some of this material as fuel in the boilers, without
impairing its function in soil recovery [26]. In this work, it is
considered that 50% of the total straw produced is left in the field,
while the remaining 50% is used in the cogeneration, which rep-
resents 23% in the bagasse-straw mixture used as fuel. At present,
there are plants that use this percentage of straw as fuel in mixture
with bagasse.
According to Figs. 5 and 6, the mass and energy balance simu-
lations for the cogeneration system were performed with the EES®
software, to calculate the surplus electricity for each case study
proposed in Table 4.
For all proposed case studies, the flowchart exposed in Fig. 3a,
representing the cleaning and grinding system of straw prior to the
mixing with bagasse, was assumed.
5. Results
The Grand Composite Curve (GCC) was constructed for each case
study. Fig. 8 shows the GCC without considering the streams of the
evaporation system.
According to [28], one of the main advantages of the GCC is the
identification of the appropriate temperature level of utilities to be
used for each case study; minimising the irreversibilities of the heat
exchange process, and allowing the evaluation of the use of lower
cost utilities.
5.1. Heat integration with the evaporation process
According to [28], the integration of evaporation systems should
be performed taking into account the overall process, with the help
of the GCC. In order to accomplish an appropriate placement of the
multiple-effect evaporation system, from the results of the evapo-
ration system simulation, each effect of the evaporation system is
represented in the GCC as a horizontal bar [28]. The length of the
bars represents the heat exchanged in each effect, while its place-
ment in the vertical axis is determined by the operation tempera-
ture, which is a function of its operating pressure.
After performing the simulation, the results showed that only
two effects of evaporation are necessary in case FBF-200, while case
H-FBMF-240 needed three effects. Fig. 9 shows the GCC together
with the evaporation system for cases FBF-200 and H-FBMF-240.
In Fig. 9a (case FBF-200), it can be observed that the bar rep-
resenting the first effect is larger than the one that represents the
second. The upper side of the bar represents the heat supplied to
the effect (heat sink), while the bottom side (heat source)
 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480 475

Table 2
Streams used in the heat integration analysis for each case study.

Stream Name Supply Temperature Target Temperature Base Case FBF-200 H-FBMF-240

DH
Ti [ C] Tf [ C] MW MW

Sterilised juice 130 32 41.0 41.0

Wine at fermentation 32 28 12.2 12.2
Phlegmasse 103.9 35 2.7 2.7
Vinasse 109.3 35 33.6 16.8
Anhydrous ethanol 78.3 35 1.2 1.3
Condenser - column D 84.9 35 17.3 12.3
Condenser - column B 81.7 81.6 23.4 26.7
Condenser - extractive column 78.3 78.2 14.9 14.9
Imbibition water 25 50 4.1 4.1
Raw juice 34.2 105 44 44
Final wine 31.2 90 28.9 15.5
Reboiler - column A 109.3 109.4 38.7 23.2
Reboiler - column B 103.9 103.91 19.3 26.6
Reboiler - extractive column 134.5 134.6 7.1 7.1
Reboiler - recovery column 149.6 149.7 2.8 2.8

Where the case studies are:

 Base Case FBF-200: Fed-batch fermentation with a substrate concentration at feed of So ¼ 200 kg/m3 at 30  C.
 H-FBMF-240: Fed-batch membrane fermentation with a substrate concentration at feed of So ¼ 240 kg/m3 at 30  C.

Table 3
Parameters for the simulations [29].

Parameter Value

Processed cane [t/h] 500

Generated bagasse [t/h] 138.3
Generated bagasse [kg bagasse/t of cane] 276.6
Bagasse for filters [t/h] 2.5
Reserve bagasse [t/h] 6.9
Reserve bagasse [%] 5
Lower heating value of wet bagassea [MJ/kg] 7.83
Lower heating value of wet strawb [MJ/kg] 13.26
Generated strawc [kg straw/t of cane] 140
Straw used in cogenerationc [t/h] 35
Straw used in cogenerationd [t/h] 39
Bagasse humidity [% w.b.] 50
Straw humidity [% w.b.] 10
Percentage of straw used in cogenerationd [%] 23
Percentage of straw used in cogenerationc [%] 35
Boiler efficiency on PCI base [%] 86
Isentropic efficiency of steam turbine - high-pressure stages [%] 88
Isentropic efficiency of steam turbine - low-pressure stages [%] 80
Isentropic efficiency of pumps [%] 80
Mechanical efficiency of turbines [%] 97
Efficiency of the electric generator [%] 96
Process steam pressure - high level [bar] 6
Electricity consumption in the process [kWh/t of cane] 28
Electricity consumption in straw processing (shredder), [kWh/t of straw] 82
Electricity consumption in straw processing (fans and conveyors) [kWh/t of straw] 13.6
Electricity consumption in permeate compressing [kWh/kg of permeate]e 0.115
a
50% of moisture content.
b
10% of moisture content.
c
Dry basis.
d
Wet basis.
e
Case H-FBMF-240.

Table 4
Case studies.

Configuration of the cogeneration system Steam conditions at boiler outlet - Type of Fermentation

A (Steam cycle with condensing-extraction turbines) C1: Base Case FBF-200

C2: Base Case FBF-200
C1: Case H-FBMF-240
C2: Case H-FBMF-240
B (Supercritical steam cycle) C3: Base Case FBF-200
C3: Case H-FBMF-240
476 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480

Table 5
Results of the GCC analysis.

Parameter FBF-200 H-FBMF-240

Hot Pinch [ C] 108.1 123.1

Cold Pinch [ C] 98.1 113.1
Min hot utility [MW] 76.04 68.6
Min cold utility [MW] 76.15 78
Evaporation system
Pressure of 1st effect [bar] 2.25 2.392
Pressure of 2nd effect [bar] 1.862 1.985
Pressure of 3rd effect [bar] e 1.583
Vapour bleeding of 1st effect - V1 [MW] 23 e
Fig. 8. Grand Composite Curve for the cases FBF-200 and H-FBMF-240.
Vapour bleeding of 1st effect e V1 [t/h] 38.7 e
Vapour bleeding of 2nd effect e V2 [MW] e 25
Vapour bleeding of 2nd effect e V2 [t/h] e 40.9
Hot utility
Steam consumption at 6 bar [kg of steam/t of cane] 262.4 236.8

43% and 49% for cases FBF-200 and H-FBMF-240 respectively.

From the stream data (Table 2) for heat integration, a significant
reduction in energy consumption in column A (reduction of 40%)
can be observed, owing to the introduction of extractive fermen-
tation by membrane in case H-FBMF-240; however, there is an
increase of 37% in energy consumption in column B, because of the
permeate feed in this column. In the end, the net reduction in en-
Fig. 9. Grand Composite Curve with appropriate placement of evaporators. Two effects
for case FBFe200; and three effects for case H-FBMF-240.
ergy consumption in distillation was only of 14%.
Regarding the vinasse production, which is a residue of the
distillation step, it is worth mentioning its significant reduction
owing to the introduction of the membrane technology. In such a
represents the heat available from the vapour separated in the ef-
way, the vinasse production in Case FBF resulted in 13.1 kg/kg of
fect (vegetable vapour). Accordingly, a vapour bleeding can be
hydrated ethanol, while in Case H-FBMF-240, it was 6.37 kg/kg of
observed in the first effect, because one part of the vapour of the
hydrated ethanol (Table A1 from Annex). However, this also
first effect is used as heat input in the second effect, while the
affected the energy targets in case H-FBMF-240, since the reduction
remaining part is used to cover other heating demands in the
in vinasse production also reduced in 50% the heat availability of
process. On the other hand, in Fig. 9b, the vapour bleeding is
the vinasse stream in said case.
extracted in the second effect.
Analysing the curves in Fig. 8, the previous GCC in Case H-FBMF-
Moreover, it is worthwhile to mention that the increase of the
240 has a hot utility consumption which is 12.6% lower in com-
operating pressure of the evaporators was necessary in order to
parison to Case FBF-200. Moreover, taking into account that the
increase their respective operating temperatures. In this manner,
evaporation system in case H-FBMF-240 has to concentrate the
an appropriate placement of evaporators and the meeting of the
juice until 25.7 Brix, a larger amount of hot utility in the first effect
minimum allowed temperature difference assumed DTmin (be-
of its evaporation system is required. For this reason, it was not
tween the process and the evaporator streams) could be obtained.
possible to place two effects, as in case FBF-200, being necessary
Fig. 10 shows the final GCC with the final streams of the evaporation
the use of three effects.
system integrated into the overall process.
According to Table 5, Case H-FBMF-240 has a lower consump-
tion of hot utilities in comparison to the base Case FBF-200. As a 5.2. Comparison of surplus electricity
consequence, Case H-FBMF-240 has a steam consumption which is
10.8% lower in comparison to Case FBF-200. However, comparing After carrying out the mass and energy balances of the steam
the steam consumption of these integrated cases with a non- cycles of the cogeneration systems, the generation of surplus
integrated mill which consumes an average of 468 kg of steam/t electricity can be estimated and compared; which results can be
of cane [9], the reduction in steam consumption reaches values of

Fig. 10. Final GCC including the streams of the evaporation systems for the cases FBF-
200 and H-FBMF-240. Fig. 11. Surplus electricity for each case study.
 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480 477

observed in Fig. 11. It can be noticed that the electricity surplus

increases from 87.3 MW (175 kWh/t of cane) for case FBF-200 with
Conf. A C1, to 136.3 MW (273 kWh/t of cane) for case H-FBMF-240
with Conf. B C3.
Regarding only the case FBF-200, an increase in surplus elec-
tricity of 10% can be reached solely by changing the conditions of
the generated steam from C1 to C2. For the case H-FBMF-240, the
increase was approximately the same.
The results show that, for sugarcane cogeneration processes
with condensing-extraction turbines, the generation of surplus
electricity can be increased significantly using boilers operating
with better technology, such as those currently used to generate
steam at 120 bar and 520  C (C2). In Fig. 11, the increase in surplus
electricity, in supercritical conditions, is noticeable due to the in-
crease in the pressure and temperature of the generated steam. It
can be observed that the operating conditions of the boiler are a
very important variable in the generation of surplus power in the
cogeneration systems of sugar-alcohol plants.
In regard to the supercritical configuration, the surplus elec- Fig. 12. Hot utility and surplus electricity variation in relation to DTmin.
tricity for the case study FBF-200 has a value of 137.7 MW; while for
the case H-FBMF-240, the surplus electricity reached 136.3 MW. On
that account, supercritical cycles presented a significant increase of additional electricity consumption can be observed in the mem-
approximately 58% in comparison to their respective cases (FBF- brane fermentation case H-FBMF due to permeate compression,
200 and H-FBMF-240) at C1 conditions; and an increase of 43% in which reduced the availability of surplus electricity. Table A3 in
comparison to their respective cases at C2 conditions. It must be Annex presents a detailed electricity balance of the evaluated cases.
taken into account that for the supercritical configuration, all of the
available bagasse is used in the cogeneration system, as in the case
with the use of condensing-extraction turbines.
5.3. Sensitivity analysis
According to Fig. 11, it can be deduced that in order to increase
the efficiency of electricity generation in these cogeneration sys-
A sensitivity analysis was accomplished in order to evaluate the
tems, it is necessary the increase of the parameters of pressure and
behaviour of the surplus electricity that can be produced when
temperature of live steam, as well as the modification of the
some parameters are changed. Fig. 12 shows the variation of the hot
configuration of the steam cycle, introducing reheating and re-
utility and the surplus electricity in relation to the minimum
generators, such as in the supercritical configuration (Conf. B).
allowed temperature difference DTmin of process streams. More-
Supercritical steam cycles present the highest efficiencies of
over, the heat transfer area target, calculated with a heat transfer
electricity generation, and they are commonly dedicated to elec-
coefficient (h) of 0.2 kW/m2 C, is presented. The range for DTmin
tricity generation using coal as fuel. However, recent cogeneration
was selected taking into account the DTmin ¼ 4  C of evaporation
plants, such as Avedøre II, with a supercritical steam cycle that uses
streams and the selected temperature of hot utility, which were
biomass as fuel, are already in existence.
kept without changes.
In the Brazilian context, where the bioelectricity became
Regarding the cases with the process FBF-200, the results
another product of the sugar-and-ethanol-producing mills, the
showed a variation in the hot utility of 5% for DTmin ¼ 5  C, and
improvement of cogeneration systems is presented as an inter-
of þ2.3% for DTmin ¼ 12  C, in relation to the base case
esting alternative for increasing the surplus electricity that can be
(DTmin ¼ 10  C). However, the variation in surplus electricity
sold to the electrical grid.
resulted lower than 1% for all cases. The cases with the process H-
Additional results regarding the cogeneration system can be
FBMF-240 showed results similar to the cases FBF-200.
observed in Table 6.
Fig. 13 presents the variation in surplus electricity with respect
Regarding the cases FBF-200 and H-FBMF-240 at the same
to the percentage of sugarcane straw recovered from the field.
conditions in the cogeneration system, case H-FBMF-240 presented
Results for cases with the FBF-200 process were very similar to
a slightly lower surplus electricity (around 1% lower). Although
those with the H-FBMF-240 process. According to the calculations,
case H-FBMF-240 presented a lower steam consumption in com-
a potential of 340.8 kWh/t of cane (170.4 MW) of surplus electricity
parison to case FBF-200 (reduction of 9.8% according to Table 5), an
generation could be achieved with 80% of straw recovery.

Table 6
Results of the supercritical cogeneration system for each case study.

Configuration A C1 Configuration A C1 H- Configuration A C2 Configuration A C2 H- Configuration B Configuration B H-

FBF-200 FBMF-240 FBF-200 FBMF-240 FBF-200 FBMF-240

Steam generated at boiler [t/h] 446.4 446.4 440.6 440.6 486.3 486.3
Pressure of live steam [bar] 65 65 120 120 300 300
Temperature of live steam [ C] 480 480 520 520 600 600
Steam consumption [t/h] 131.4 118.4 131.2 118.4 131.2 118.4
Wsurplus EE [MW] 87.3 86.2 96.2 95.2 137.7 136.3
Specific Wsurplus EE (per tonne of cane) 175 172 192 190 275 273
[kWh/t of cane]
478 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480

The introduction of sugarcane straw into the cogeneration

process, for the different case studies, presents great advantages,
such as the significant increase in the quantity of raw material
available for the process. The sugarcane straw also presents a
lower degree of humidity and has a higher heating value in rela-
tion to sugarcane bagasse. The utilisation of this agricultural
residue proved to be a good alternative, generating greater
amounts of surplus electricity in the cogeneration system. Heat
integration through Pinch Analysis promoted a significant
reduction in process steam consumption; consequently, after
calculating the heat requirements, it can be concluded that case
H-FBMF-240 presents better results, with a steam consumption of
236.8 kg of steam (saturated at 6 bar) per tonne of cane, which is
Fig. 13. Surplus electricity variation in relation to the % of straw recovered from the
9.76% lower in comparison to case FBF-200, and 49% lower when
field.
compared to a conventional case with a steam consumption of
467 kg of steam/t of cane. However, regarding the surplus elec-
5.4. Discussion
tricity, case H-FBMF-240 presented slightly lower values in
comparison to case FBF-200, because of the additional electricity
The reduction in steam consumption in the distillation step
consumption in the permeate compression operation. Comparing
depends on the increase of the ethanol concentration of the wine
only the configurations of the cogeneration systems, the surplus
entering the distillation columns. It has been shown that perva-
electricity of the supercritical cogeneration system presented the
poration adapted to fermentation is an alternative for reducing the
highest value.
energy consumption in wine distillation, which leads to the
reduction in the steam consumption of the process. However, the
incorporation of this alternative technology has to be analysed Acknowledgments
taking into account the overall process, in order to evaluate the
impacts in each sub-system of said process, including the cogene- The authors wish to thank to CAPES and CNPq (Process PQ
ration system. Therefore, it has also been demonstrated that hybrid 306303/2014-0) for the researcher fellowship, and FAPESP for the
fermentation, in addition to increasing the ethanol production in Research Project Grant (Process 2011/51902-9).
the fermentation, may present an increase in the additional elec-
tricity consumption, both in the distillation and in the permeate ANNEX
recovery system, as well as presenting some technical complexities.
The aforementioned is not considered in the literature, where
hybrid fermentation is referred to as a highly convenient technique Table A1
to reduce energy consumption. Main streams and data for cases FBF-200 and H-FBMF-240.
The heat integration, the use of straw, and the improvement of
FBF- H-FBMF-
cogeneration systems promote the increase in surplus electricity 200 240
when steam cycles with extraction-condensing steam turbines are
Treated juice for concentration, (kg/s) 141.5 141.5
used. This increase in surplus electricity is highly advantageous in Brix 15 15
the Brazilian context since the bioelectricity became a new product Pol 12.9 12.9
of sugarcane-processing mills; therefore, the results presented in Concentration
this study are considerably useful to show the potential of surplus Mass flow of concentrated juice, (kg/s) 97.8 82.7
Substrate concentration of must, (kg/m3) 200 240
electricity that can be sold to the grid. Economic assessments,
Final Brix 21.7 25.67
however, are necessary in order to evaluate the feasibility of Pol 18.7 22.1
implementing these technologies. Total mass flow of separated vapour, (kg/s) 43.7 58.8
Fermentation
Substrate (recoverable sugars), (kg/s) 18.2 18.3
6. Conclusion Yield, % 83 84
Ethanol produced, (kg/s) 8.1 8.2
Mass flow of wine, (kg/s) 128.3 66.9
The analysis of the ethanol production process incorporating the Mass flow of ethanol in wine, (kg/s) 8.1 6.1
hybrid fermentation was carried out to evaluate the steam con- Ethanol concentration in wine, (wt. %) 6.3 9.04
sumption in the distillation; seeking to raise the most relevant Mass flow of permeate, (kg/s) 0 6.4
aspects regarding possible improvements in energy efficiency. The Mass flow of ethanol in permeate, (kg/s) 0 2.1
Ethanol concentration in permeate, (wt. %) 0 33.4
reduction in the steam consumption in the distillation depends on Distillation
the increase of the ethanol concentration entering the distillation. Mass flow of hydrated ethanol - HE, (kg/s) 8.4 8.7
It has been shown that pervaporation adapted to fermentation is an Mass flow of ethanol in HE, (kg/s) 7.9 8.1
alternative to reduce the energy consumption in wine distillation. Ethanol concentration in HE, (wt. %) 93.5 93.5
Mass flow of vinasse, (kg/s) 110.2 55.2
In this study, an optimisation procedure was followed to inte-
Mass flow of phlegmasse, (kg/s) 9.51 9.47
grate the multiple-effect evaporation system, reducing the number Specific vinasse production (kg of vinasse/kg of hydrated 13.1 6.37
of effects to two in the first evaluated case, and to three in the ethanol)
second. The heat integration method was also applied in this work, Specific phlegmasse production (kg of phegmasse/kg of 1.13 1.09
integrating the juice evaporation process with the fermentation hydrated ethanol)

and distillation processes.

 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480 479

Table A2
Streams for heat integration including evaporation system streams.

Stream Name Supply Temperature Target Temperature FBF-200 H-FBMF-240

DH DH
 
Ti [ C] Tf [ C] MW MW

Sterilised juice 130 32 41.0 41.0

Wine at fermentation 32 28 12.2 12.2
Phlegmasse 103.9 35 2.7 2.7
Vinasse 109.3 35 33.6 16.8
Anhydrous ethanol 78.3 35 1.2 1.3
Condenser - column D e section 1 (82%) 84.9 79 14.6 10.1
Condenser - column D e section 2 (18%) 79 35 3.1 2.2
Condenser - column B 81.7 81.6 23.4 26.7
Condenser - extractive column 78.3 78.2 14.9 14.9
Imbibition water 25 50 4.4 4.4
Raw juice 34.2 105 44 44
Pre-heating juice 98.1 124a/126b 14.3 15.5
Juice for sterilisation 118a/113b 130 4.3 5.8
Final wine 31.2 90 29.8 15.5
Reboiler - column A 109.3 109.4 38.7 23.2
Reboiler - column B 103.9 103.91 19.3 26.6
Reboiler - extractive column 134.5 134.6 7.1 7.1
Reboiler - recovery column 149.6 149.7 2.8 2.8
Condensate -vapour V1 124a/126b 35 9.9 8.5
Condensate - vapour V2 118a/120b 35 5.9 8.3
Condensate e vapour V3 113 35 e 4.3
Juice - first effect 124a/126b 124a/126b 58.2 48.9
Vapour - first effect V1 124a/126b 124a/126b 58.2 48.9
Juice - second effect 118a/120b 118a/120b 35.2 48.9
Vapour - second effect V2 118a/120b 118a/120b 37.8 51.6
Juice e third effect 113 113 e 26.6
Vapour e third effect V3 113 113 e 29.6
a
Case FBF-200; bCase H-FBMF-240.

Table A3
Electricity balance for evaluated cases.

Configuration A C1 Configuration A C1 Configuration A C2 FBF-200 Configuration A C2 Configuration B Configuration B

FBF-200 H-FBMF-240 H-FBMF-240 FBF-200 H-FBMF-240

Electricity produced in Rankine 106,636 108,198 116,806 118,361 162,874 164,089

cycle [kW]
Electricity consumed in Rankine 1637 1637 2831 2831 7421 7421
cycle [kW]
Electricity consumption in production 14,000 14,000 14,000 14,000 14,000 14,000
process [kW]
Straw processing (shredder) [kW] 3199 3199 3199 3199 3199 3199
Straw processing (fans and conveyors) 530.4 530.4 530.4 530.4 530.4 530.4
[kW]
Permeate compression [kW] 0 2649 0 2649 0 2649
Surplus electricity [kW] 87,269 86,182 96,245 95,151 137,723 136,297

References Trans 2013;35:23e8.

[9] Pina EA, Palacios-Bereche R, Chavez-Rodriguez MF, Ensinas AV, Modesto M,
Nebra SA. Reduction of process steam demand and water-usage through heat
[1] Fortes C, Trivelin PCO, Vitti AC. Long-term decomposition of sugarcane harvest
integration in sugar and ethanol production from sugarcane - evaluation of
residues in Sao Paulo state, Brazil. Biomass Bioenergy 2012;42:189e98.
different plant configurations. Energy 2017;138:1263e80.
[2] Gao C, Zhang M, Ding J, Pan F, Jiang Z, Li Y, Zhao J. Pervaporation dehydration
[10] Palacios-Bereche R, Ensinas AV, Modesto M, Nebra SA. Double-effect distil-
of ethanol by hyaluronic acid/sodium alginate two-active-layer composite
lation and thermal integration applied to the ethanol production process.
membranes. Carbohydr Polym 2014;99:158e65.
Energy 2015;82:512e23.
[3] Dong energy. The Avedøreværket CHP plant. Denmark. 2016. Available at:
[11] Palacios-Bereche R, Ensinas AV, Modesto M, Nebra SA. New alternatives for
https://stateofgreen.com/files/download/36 [accessed 19.10.16].
the fermentation process in the ethanol production from sugarcane: extrac-
[4] Moncada J, El-Halwagi MM, Cardona CA. Techno-economic analysis for a
tive and low temperature fermentation. Energy 2014;70:595e604.
sugarcane biorefinery: Colombian case. Bioresour Technol Biorefin 2013;135:
[12] Palacios-Bereche R, Ensinas AV, Modesto M, Nebra SA. Extraction process in
533e43.
the ethanol production from sugarcane: a comparison of milling and diffusion.
[5] Ensinas AV, Nebra SA, Lozano MA, Serra LM. Analysis of process steam de-
Chem Eng Trans 2014;39:1519e24.
mand reduction and electricity generation in sugar and ethanol production
[13] Walmsley TJ. A Total Site Heat Integration design method for integrated
from sugarcane. Energy Convers Manag 2007;48:78e87.
evaporation systems including vapour recompression. J Clean Prod 2016;136:
[6] Ensinas AV, Nebra SA, Lozano MA, Serra LM. Design of evaporation systems
111e8.
and heaters networks in sugar cane factories using a thermoeconomic opti-
[14] Ren X-Y, Jia X-X, Varbanov PS, Klemes JJ, Liu Z-Y. Targeting the cogeneration
mization procedure. Int J Thermodyn 2007;10:97e105.
potential for Total Site utility systems. J Clean Prod 2018;170:625e35.
[7] Morandin M, Toffolo A, Lazzaretto A, Marechal F, Ensinas AV, Nebra SA.
[15] Ong BHY, Walmsley TG, Atkins MJ, Walmsley MRW. Total site mass, heat and
Synthesis and parameter optimization of a combined sugar and ethanol
power integration and process graph. J Clean Prod 2017;167:32e43.
production process integrated with a CHP system. Energy 2011;36:75e90.
[16] Sharan P, Bandyopadhyay S. Energy integration of multiple-effect evaporator,
[8] Ensinas AV, Codina V, Marechal F, Silva MA. Thermo-economic optimization of
thermo-vapor compressor, and background process. J Clean Prod 2017:
integrated first and second generation sugarcane ethanol plant. Chem Eng
480 C. Lopez-Castrillon et al. / Energy 156 (2018) 468e480
1192e204.
[17] Tarighaleslami AH, Walmsley TG, Atkins MJ, Walmsley MRW, Neale JR. Total
Site Heat Integration: utility selection and optimisation using cost and exergy
derivative analysis. Energy 2017;141:949e63.
[18] Oluleye G, Smith R. A mixed integer linear programming model for integrating
thermodynamic cycles for waste heat exploitation in process sites. Appl En-
ergy 2016;178:434e53.
[19] Cortes EF, Fukushima NA, Palacios-Bereche R, Ensinas AV, Nebra SA. Vinasse
concentration and juice evaporation system integrated to the conventional
ethanol production process from sugarcane e heat integration and impacts in
cogeneration system. Renew Energy 2018;115:474e88.
[20] Leon JA, Palacios-Bereche R, Nebra SA. Batch pervaporative fermentation with
coupled membrane and its influence on energy consumption in permeate
recovery and distillation stage. Energy 2016;109:77e91.
[21] Leon JA. Analysis of the performance of pervaporation membranes in the
fermentation conventional process for the ethanol production (in Portu-
guese). Master’s thesis. Santo Andre , Brazil: Federal University of ABC; 2015.
[22] Hofmann D, Fritz L, Ulbrich J, Paul D. Molecular modelling of amorphous
membrane polymers. Polymer 1997;38:6145e55.
[23] Chen C, Tang X, Xiao Z, Zhou Y, Jiang Y, Fu S. Ethanol fermentation kinetics in a
continuous and closed-circulating fermentation system with a pervaporation
membrane bioreactor. Bioresour Technol 2012;114:707e10.
[24] Ensinas AV. Heat integration and thermoeconomic optimisation applied to the
industrial production process of sugar and ethanol from sugarcane (in Por-
tuguese). Doctoral Thesis. Campinas, Brazil: University of Campinas; 2008.
[25] Pellegrin LF, Oliveira JS, Burbano JC. Supercritical steam cycles and biomass
integrated gasification combined cycles for sugarcane mills. Energy 2010;35:
1172e80.
[26] Hassuani S, Leal MR, Macedo IDC. Biomass power generation: sugar cane
bagasse and trash. (Paths for Sustainability Series). Piracicaba: PNUD-CTC;
2005. p. 217.
[27] Atala D, Costa A, Maciel R, Maugeri F. Kinetics of ethanol fermentation with
high biomass concentration considering the effect of temperature. In:
Davison B, McMillan J, Finkelstein M, editors. Twenty-second symposium on
biotechnology for fuels and chemicals, ABAB symposium. Humana Press;
2001. p. 353e65.
[28] Klemes J, editor. Handbook of process integration (PI). USA: Woodhead Pub-
lishing Limited; 2013.
[29] Palacios-Bereche R, Pereira P, Nebra SA, Oliveira CEI, Rabi JA. Energetic eval-
uation of cogeneration systems in sugar cane plants in Brazil e case studies.
Proceedings of ECOS 2009-22nd international conference on efficiency, cost,
optimization, Foz do Iguaçu, Parana , Brazil; 449e458.