- Email: [email protected]
Optimal production of Furfural and DMF from algae and switchgrass
Krist V. Gernaey, Jakob K. Huusom and Rafiqul Gani (Eds.), 12th International Symposium on Process Systems Engineering and 25th European Symposium on Computer Aided Process Engineering. 31 May – 4 June 2015, Copenhagen, Denmark © 2015 Elsevier B.V. All rights reserved.
Optimal production of Furfural and DMF from algae and switchgrass Mariano Martína, Ignacio E. Grossmannb a
Departamento de Ingeniería Química. Univ Salamanca.Plz. Caídos 1-5, 37008, Salamanca, Spain b Department of chemical Engineering. Carnegie Mellon University. Pittsburgh, PA, 15213, USA.
Abstract In this paper we present conceptual designs of optimal integrated processes for the production of DMF and furfural from biomass, switchgrass and algae. The processes consist of four stages, (1) biomass pretreatment into intermediates such as oil and glucose from algae and glucose and xylose from switchgrass, (2) dehydration of the sugars, (3) HMF and furfural purification and (4) synthesis of DMF out of HMF and FAEE synthesis from oil in the case of using algae. Simultaneous optimization and heat integration is performed for the processes using each raw material. For switchgrass, the use of AFEX pretreatment is recommended for a production of 9 MMgal of furfural and 14 MMgal/yr of DMF. The production cost is $3/kg of biofuel ($570MM of investment cost). When using algae, its composition should be 60% oil, 30% starch and 10% protein to obtain 98 MMgal/yr of biofuels, 16% of DMF, at the cost of $1.98/gal, $0.61/kg of biofuel, requiring $693 MM of investment. Keywords: Algae, Switchgrass, DMF, Furfural, Mathematical optimization
1. Introduction Biomass is a rich and diverse raw material for a number of products that have sugars or oil as building blocks. In particular, sugars are the starting block for most biochemical process including bioethanol production. However, we can use sugars to produce added value products and/or intermediates. Furans are a family of chemicals that serve as fuels but can also be used as a building block. Algae biomass comprises lipids, proteins and starch, while lignocellulosic raw materials contain celluloses and hemicelluloses which are a source for pentoses and hexoses. Thus, it is expected we can produce dimethyl furfural (DMF) out of the algae starch, and from switchgrass we should be able to obtain not only DMF but also furfural. Lately, a number of papers have evaluated the production of furan and HMF and DMF from sugars, xylose, fructure or glucose, but all of them start with the sugars without evaluating the production of those sugars from non edible biomass, which is a highly energy and capital intensive stage (Kazi et al 2011, de Ávila & Guiradello 2012). In this work we present conceptual optimal designs for processes from the biomass to furans.
2. Process description The process is divided into four stages: (1) biomass pretreatment into intermediates such as oil and glucose from algae and glucose and xylose from switchgrass, (2)
1044
M. Martín, I. E. Grossmann
dehydration of the sugars, (3) HMF and furfural purification and (4) synthesis of DMF out of HMF and FAEE synthesis from oil in the case of using algae. 2.1. Sugar production from algae The algae are grown in ponds, harvested using a capillarity based belt. The dry algae biomass is assumed to be composed of oil, up to a maximum of 60 %w/w, starch and protein with a minimum of 10%w/w. On the one hand, the oil, extracted using a hybrid method combining mechanical action and the use of solvents, will be transesterified with ethanol using enzymes. The yield of the reaction is predicted using a surface response model as a function of the excess of ethanol, the load of catalysis and the operating temperature. The excess of ethanol is recovered in a distillation column and the polar and organic phases are separated recovering the glycerol and biodiesel which is finally purified by distillation. On the other hand, the biomass containing the starch and protein is liquefied, 85ºC, and saccharified, 65 ºC, to breakdown the structure into maltose and glucose. (Martín & Grossmann, 2013). Figure 1 shows the flowsheet for the use of algae for the production of DMF and biodiesel.
Figure 1.- Flowsheet for the production of DMF and biodiesel from algae
2.2. Sugars production from switchgrass Two pretreatments are considered for breaking up of the biomass into cellulose, hemicelluloses and lignin, dilute acid and ammonia fiber explosion (AFEX). Once the physical structure of the switchgrass has been broken down, the cellulose and lignocellulose are next hydrolyzed at 50ºC obtaining pentoses and hexoses (Martín & Grossmann, 2012). Figure 2 shows the superstructure for the production of furfural and DMF from switchgrass.
Figure 2.-Superstructure for the production of furfural and DMF from switchgrass
2.3. Sugar transformation into furfural and HMF The stream containing sugar stream is mixed with butanol, HCl, CrCl3 and NaCl together with recycled streams coming from different separation stages such as the
Optimal Production of Furfural and DMF from Algae and Switchgrass
1045
filter, where the solid catalyst (CrCl3) is recovered, the first distillation column, where mainly water and butanol are recycled, and the distillate of the second distillation column consisting mainly of the same species. The CrCl3 – HCl is the catalyst system and we use a two phase liquid reactor (organic, butanol, - aqueous) to extract the product from the mixture. The NaCl increases the recovery ratio of the products. In the reactor, the sugars, glucose and xylose, are dehydrated, eqs. (1)-(2). In case of using algae only reaction (1) takes place. The operating temperature and pressure at the reactor are fixed to be 180ºC and 11 bar since it is a common temperature for both sugars to be dehydrated. With this operating conditions the conversion from glucose to HMF becomes 0.67 (Roman –Leshkov et al., 2007) and from xylose to furfural to 0.56 (Marcotulio et al., 2010). Glucose dehydration: C6 H12O6 C6 H 6O3 3H 2O Xylose dehydration: C5 H10O5 C5 H 4O2 3H 2O
(1) (2)
Figure 3 shows the detailed flowsheet for the sugar dehydration process and the subsequent purification process. In case when there is only glucose in the feed, one column is not needed.
Figure 3.-Sugar dehydration and purification stages
2.4. Products purification. The purification stage consists of neutralization, filtration, phase separation and a sequence of distillation columns. The HCl is neutralized using NaOH so that the resulting salt is already in the system. Next, the solid catalyst, CrCl3, is recovered in a filter. After the filter, a phase separation is performed. Based on experimental results from the literature, an operating temperature of 37 ºC is selected (de Álvila et al., 2012) The partition coefficient of the HMF and furfural is taken to be 3 in volume ratio towards the organic phase (RomanLeshvok, 2009; Roman-Leskov et al., 2007). The polar phase contains butanol, water and NaCl and part of the sugars. The organic phase coming from the phase separator is distilled. The column operates at low pressure, 0.5 atm, to avoid species decomposition. From the distillate we recover 99,9% of butanol and the water in the feed together with the rest of the sugars. From the bottoms we obtain HMF, furfural and the rest of butanol. A second column recovers
1046
M. Martín, I. E. Grossmann
butanol as distillate, and HMF and furfural at the bottoms. Finally, the third column separates furfural to be sold from the HMF that is further used for the production of DMF. In case only hexoses are available, only two columns are needed. 2.5. DMF production from HMF The production of DMF from HMF is carried out in a catalytic reactor at 120 ºC and 16 bar using CuRu as catalyst. HMF and butanol are fed so that the concentration of the first one is 5% of the liquid mixture. The reactor operates with an excess of hydrogen 12 times the one given by the stoichiometry. The conversion is reported to be 100% (Kazi et al. 2011). The excess of hydrogen is recovered using a flash, and the DMF, butanol and water are sent to a distillation column. The reaction taking place is given in eq. (3). Figure 4 shows the detailed flowsheet for this stage. C6 H 6O3 4 H 2
C6 H 8O 2 H 2O
(3)
Figure 4.- Production of DMF from HMF
3. Solution procedure. We formulate the problem for the use of switchgrass as raw material as an MINLP where the objective function is a simplified production cost involving the income due to the furfural and the DMF assumed to have a cost of $2/kg and the energy consumed in the process. Due to the fact that there is only one integer variable we decompose the problem into two NLP with 3000 equations and 3600 variables each to solve simultaneously optimize the operating conditions of the pretreatment and heat integrate the entire flowsheet using Duran & Grossmann’s (1986) model. In the case of the use of algae, the problem is formulated as an NLP with 4100 equations and 5400 variables where the main decision variables are the algae composition, the operating pressures of the distillation columns and the operating conditions of the transesterification reactor. Again simultaneous optimization and heat integration is performed. We use multiple initialization points to solve the problems with CONOPT 3.0 as the preferred solver. Next, we develop the optimal heat exchanger network using SYNHEAT (Yee & Grossmann 1990). Finally, the cost analysis is performed following Sinnot´s method (Sinnot 1999) .
Optimal Production of Furfural and DMF from Algae and Switchgrass
1047
4. Results 4.1. Lignocellulosic based DMF and furfural If we use lignocellulosic switchgrass as raw material we have as main products DMF and furfural and as byproduct lignin. We consider a price for the switchgrass of $75/t. AFEX turns out to provide the better performance for a production cost of $11.3/gal ($3/kg) versus $12.15/gal ($3.20/kg) for the acid based pretreatment with an investment cost of $570 MM (vs. 764MM). We obtain 23MMgal of products, 9 MMgal of furfural and 14 MMgal/yr of DMF, representing 15% by weight of the initial biomass for the best of the two processes. Due to the need of energy for the process, the lignin produced is fully used to provide part of it 4.2. Algae based DMF In this case we have as variable the algae composition to produce as main products FAEE and DMF while as byproducts we can obtain glycerol and protein. The algae composition that yields the optimal production facility corresponds to 60% oil, 30% starch and 10% protein, as in previous studies. The reason is that the production of biodiesel is cheaper in terms of energy consumption and investment cost while the use of starch as raw material is much more expensive. The operating conditions at the transesterification reactor are given in Table 1. They match the ones of other integrated processes for the production of biodiesel and starch based products such as ethanol (Martín & Grossmann, 2013). Table 1.- Operating conditions at the transesterification reactor
Temperature(ºC) Pressure(bar) Molar ratio (EtOH/oil) Time (h) Cat/lipase(%) Water added
Algae based FAEE + DMF 30 4 (fixed) 4.2 8.0 13.0 0.0
The total production capacity of biofuels is 98 MMGal /yr where 81MMGal/yr correspond biodiesel (FAEE) and 17 MMGal/yr of DMF. The investment cost for this facility is high, $693 MM. In terms of production cost for the main products, namely FAEE and DMF, is $1.98/gal, $0.61/kg of biofuel.
5. Conclusions When switchgrass is used, the preferred pretreatment is AFEX followed by hydrolysis so that the xylose and glucose are available for dehydration. If algae are used, we can not only produce DMF due to the starch within the algae, but also biodiesel that we assume is FAEE for which we need ethanol. The resulting optimal processes reveal interesting production prices, around $3/kg if switchgrass is used and $1.7/gal in the case of producing biodiesel and DMF from algae. However, the energy consumption as well as the water consumption are higher
1048
M. Martín, I. E. Grossmann
due to the purification and separation stages for the intermediates. It is technologically feasible to produce furfural and DMF out of algae and switchgrass but at a cost. The energy consumption of the process is highly dependent on the composition of the mixture in the dehydration reactor. The experimental optimization of the dehydration process is suggested to reduced the amount of butanol so that the separation of the products from the separating agent is less expensive and /or the use of a different agent that has a lower vaporization heat. This can somehow be obtained if the concentration of sugars in the aqueous phase is increased. The concentration of the sugars in the fed can also be adjusted with evaporation previous to the dehydration as long as there is no such need for aqueous phase during the synthesis.
References F. de Ávila Rodrigues, R. Guirardello, 2012, Techno-Economic Evaluation of Large Scale 2.5Dimethylfuran Production from Fructose. Chap. 17 In Advances in Chemical Engineering Intech M.A. Duran, I.E. Grossmann, 1986, Simultaneous optimization and heat integration of chemical processes. AIChE, Journal, 32, 123-138 F.K. Kazi, A.D. Patel, J.C. Serrano-Ruiz, J.A. Dumesic, R.P. Anexa, 2011, Techno-economic analysis of dimethylfuran (DMF) and hydroxymethylfurfural (HMF) production from pure fructose in catalytic processes Chemical Engineering Journal 169, 329–338 G. Marcotullio, W. De Jong, 2010, Chloride ions enhance furfural formation from D-xylose in dilute aqueous acidic solutions Green Chem., 12, 1739–1746 M. Martín, I.E. Grossmann, 2012, Energy optimization of lignocellulosic bioethanol production via Hydrolysis of Switchgrass. AIChE Journal, 58, 5, 1538-1549 M. Martín, I.E. Grossmann, 2013. Optimal engineered algae composition for the integrated simultaneous production of bioethanol and biodiesel AIChE Journal, 59, 8, 2872–2883 Y. Roman-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, 2007, Production of dimethylfuran for liquid fuels from biomass-derived carbohydrates. Nature, 447, 982-986 R.K. Sinnot, 1999, Coulson and Richardson, Chemical Engineering 3ªEd. Singapore: Butterworth Heinemann, 1999 T.F. Yee, I.E. Grossmann, 1990, Simultaneous optimization models for heat integration. II. Heat exchanger networks synthesis. Computers and Chemical Engineering, 28, 1165-1184
Optimal production of Furfural and DMF from algae and switchgrass Mariano Martína, Ignacio E. Grossmannb a
Departamento de Ingeniería Química. Univ Salamanca.Plz. Caídos 1-5, 37008, Salamanca, Spain b Department of chemical Engineering. Carnegie Mellon University. Pittsburgh, PA, 15213, USA.
Abstract In this paper we present conceptual designs of optimal integrated processes for the production of DMF and furfural from biomass, switchgrass and algae. The processes consist of four stages, (1) biomass pretreatment into intermediates such as oil and glucose from algae and glucose and xylose from switchgrass, (2) dehydration of the sugars, (3) HMF and furfural purification and (4) synthesis of DMF out of HMF and FAEE synthesis from oil in the case of using algae. Simultaneous optimization and heat integration is performed for the processes using each raw material. For switchgrass, the use of AFEX pretreatment is recommended for a production of 9 MMgal of furfural and 14 MMgal/yr of DMF. The production cost is $3/kg of biofuel ($570MM of investment cost). When using algae, its composition should be 60% oil, 30% starch and 10% protein to obtain 98 MMgal/yr of biofuels, 16% of DMF, at the cost of $1.98/gal, $0.61/kg of biofuel, requiring $693 MM of investment. Keywords: Algae, Switchgrass, DMF, Furfural, Mathematical optimization
1. Introduction Biomass is a rich and diverse raw material for a number of products that have sugars or oil as building blocks. In particular, sugars are the starting block for most biochemical process including bioethanol production. However, we can use sugars to produce added value products and/or intermediates. Furans are a family of chemicals that serve as fuels but can also be used as a building block. Algae biomass comprises lipids, proteins and starch, while lignocellulosic raw materials contain celluloses and hemicelluloses which are a source for pentoses and hexoses. Thus, it is expected we can produce dimethyl furfural (DMF) out of the algae starch, and from switchgrass we should be able to obtain not only DMF but also furfural. Lately, a number of papers have evaluated the production of furan and HMF and DMF from sugars, xylose, fructure or glucose, but all of them start with the sugars without evaluating the production of those sugars from non edible biomass, which is a highly energy and capital intensive stage (Kazi et al 2011, de Ávila & Guiradello 2012). In this work we present conceptual optimal designs for processes from the biomass to furans.
2. Process description The process is divided into four stages: (1) biomass pretreatment into intermediates such as oil and glucose from algae and glucose and xylose from switchgrass, (2)
1044
M. Martín, I. E. Grossmann
dehydration of the sugars, (3) HMF and furfural purification and (4) synthesis of DMF out of HMF and FAEE synthesis from oil in the case of using algae. 2.1. Sugar production from algae The algae are grown in ponds, harvested using a capillarity based belt. The dry algae biomass is assumed to be composed of oil, up to a maximum of 60 %w/w, starch and protein with a minimum of 10%w/w. On the one hand, the oil, extracted using a hybrid method combining mechanical action and the use of solvents, will be transesterified with ethanol using enzymes. The yield of the reaction is predicted using a surface response model as a function of the excess of ethanol, the load of catalysis and the operating temperature. The excess of ethanol is recovered in a distillation column and the polar and organic phases are separated recovering the glycerol and biodiesel which is finally purified by distillation. On the other hand, the biomass containing the starch and protein is liquefied, 85ºC, and saccharified, 65 ºC, to breakdown the structure into maltose and glucose. (Martín & Grossmann, 2013). Figure 1 shows the flowsheet for the use of algae for the production of DMF and biodiesel.
Figure 1.- Flowsheet for the production of DMF and biodiesel from algae
2.2. Sugars production from switchgrass Two pretreatments are considered for breaking up of the biomass into cellulose, hemicelluloses and lignin, dilute acid and ammonia fiber explosion (AFEX). Once the physical structure of the switchgrass has been broken down, the cellulose and lignocellulose are next hydrolyzed at 50ºC obtaining pentoses and hexoses (Martín & Grossmann, 2012). Figure 2 shows the superstructure for the production of furfural and DMF from switchgrass.
Figure 2.-Superstructure for the production of furfural and DMF from switchgrass
2.3. Sugar transformation into furfural and HMF The stream containing sugar stream is mixed with butanol, HCl, CrCl3 and NaCl together with recycled streams coming from different separation stages such as the
Optimal Production of Furfural and DMF from Algae and Switchgrass
1045
filter, where the solid catalyst (CrCl3) is recovered, the first distillation column, where mainly water and butanol are recycled, and the distillate of the second distillation column consisting mainly of the same species. The CrCl3 – HCl is the catalyst system and we use a two phase liquid reactor (organic, butanol, - aqueous) to extract the product from the mixture. The NaCl increases the recovery ratio of the products. In the reactor, the sugars, glucose and xylose, are dehydrated, eqs. (1)-(2). In case of using algae only reaction (1) takes place. The operating temperature and pressure at the reactor are fixed to be 180ºC and 11 bar since it is a common temperature for both sugars to be dehydrated. With this operating conditions the conversion from glucose to HMF becomes 0.67 (Roman –Leshkov et al., 2007) and from xylose to furfural to 0.56 (Marcotulio et al., 2010). Glucose dehydration: C6 H12O6 C6 H 6O3 3H 2O Xylose dehydration: C5 H10O5 C5 H 4O2 3H 2O
(1) (2)
Figure 3 shows the detailed flowsheet for the sugar dehydration process and the subsequent purification process. In case when there is only glucose in the feed, one column is not needed.
Figure 3.-Sugar dehydration and purification stages
2.4. Products purification. The purification stage consists of neutralization, filtration, phase separation and a sequence of distillation columns. The HCl is neutralized using NaOH so that the resulting salt is already in the system. Next, the solid catalyst, CrCl3, is recovered in a filter. After the filter, a phase separation is performed. Based on experimental results from the literature, an operating temperature of 37 ºC is selected (de Álvila et al., 2012) The partition coefficient of the HMF and furfural is taken to be 3 in volume ratio towards the organic phase (RomanLeshvok, 2009; Roman-Leskov et al., 2007). The polar phase contains butanol, water and NaCl and part of the sugars. The organic phase coming from the phase separator is distilled. The column operates at low pressure, 0.5 atm, to avoid species decomposition. From the distillate we recover 99,9% of butanol and the water in the feed together with the rest of the sugars. From the bottoms we obtain HMF, furfural and the rest of butanol. A second column recovers
1046
M. Martín, I. E. Grossmann
butanol as distillate, and HMF and furfural at the bottoms. Finally, the third column separates furfural to be sold from the HMF that is further used for the production of DMF. In case only hexoses are available, only two columns are needed. 2.5. DMF production from HMF The production of DMF from HMF is carried out in a catalytic reactor at 120 ºC and 16 bar using CuRu as catalyst. HMF and butanol are fed so that the concentration of the first one is 5% of the liquid mixture. The reactor operates with an excess of hydrogen 12 times the one given by the stoichiometry. The conversion is reported to be 100% (Kazi et al. 2011). The excess of hydrogen is recovered using a flash, and the DMF, butanol and water are sent to a distillation column. The reaction taking place is given in eq. (3). Figure 4 shows the detailed flowsheet for this stage. C6 H 6O3 4 H 2
C6 H 8O 2 H 2O
(3)
Figure 4.- Production of DMF from HMF
3. Solution procedure. We formulate the problem for the use of switchgrass as raw material as an MINLP where the objective function is a simplified production cost involving the income due to the furfural and the DMF assumed to have a cost of $2/kg and the energy consumed in the process. Due to the fact that there is only one integer variable we decompose the problem into two NLP with 3000 equations and 3600 variables each to solve simultaneously optimize the operating conditions of the pretreatment and heat integrate the entire flowsheet using Duran & Grossmann’s (1986) model. In the case of the use of algae, the problem is formulated as an NLP with 4100 equations and 5400 variables where the main decision variables are the algae composition, the operating pressures of the distillation columns and the operating conditions of the transesterification reactor. Again simultaneous optimization and heat integration is performed. We use multiple initialization points to solve the problems with CONOPT 3.0 as the preferred solver. Next, we develop the optimal heat exchanger network using SYNHEAT (Yee & Grossmann 1990). Finally, the cost analysis is performed following Sinnot´s method (Sinnot 1999) .
Optimal Production of Furfural and DMF from Algae and Switchgrass
1047
4. Results 4.1. Lignocellulosic based DMF and furfural If we use lignocellulosic switchgrass as raw material we have as main products DMF and furfural and as byproduct lignin. We consider a price for the switchgrass of $75/t. AFEX turns out to provide the better performance for a production cost of $11.3/gal ($3/kg) versus $12.15/gal ($3.20/kg) for the acid based pretreatment with an investment cost of $570 MM (vs. 764MM). We obtain 23MMgal of products, 9 MMgal of furfural and 14 MMgal/yr of DMF, representing 15% by weight of the initial biomass for the best of the two processes. Due to the need of energy for the process, the lignin produced is fully used to provide part of it 4.2. Algae based DMF In this case we have as variable the algae composition to produce as main products FAEE and DMF while as byproducts we can obtain glycerol and protein. The algae composition that yields the optimal production facility corresponds to 60% oil, 30% starch and 10% protein, as in previous studies. The reason is that the production of biodiesel is cheaper in terms of energy consumption and investment cost while the use of starch as raw material is much more expensive. The operating conditions at the transesterification reactor are given in Table 1. They match the ones of other integrated processes for the production of biodiesel and starch based products such as ethanol (Martín & Grossmann, 2013). Table 1.- Operating conditions at the transesterification reactor
Temperature(ºC) Pressure(bar) Molar ratio (EtOH/oil) Time (h) Cat/lipase(%) Water added
Algae based FAEE + DMF 30 4 (fixed) 4.2 8.0 13.0 0.0
The total production capacity of biofuels is 98 MMGal /yr where 81MMGal/yr correspond biodiesel (FAEE) and 17 MMGal/yr of DMF. The investment cost for this facility is high, $693 MM. In terms of production cost for the main products, namely FAEE and DMF, is $1.98/gal, $0.61/kg of biofuel.
5. Conclusions When switchgrass is used, the preferred pretreatment is AFEX followed by hydrolysis so that the xylose and glucose are available for dehydration. If algae are used, we can not only produce DMF due to the starch within the algae, but also biodiesel that we assume is FAEE for which we need ethanol. The resulting optimal processes reveal interesting production prices, around $3/kg if switchgrass is used and $1.7/gal in the case of producing biodiesel and DMF from algae. However, the energy consumption as well as the water consumption are higher
1048
M. Martín, I. E. Grossmann
due to the purification and separation stages for the intermediates. It is technologically feasible to produce furfural and DMF out of algae and switchgrass but at a cost. The energy consumption of the process is highly dependent on the composition of the mixture in the dehydration reactor. The experimental optimization of the dehydration process is suggested to reduced the amount of butanol so that the separation of the products from the separating agent is less expensive and /or the use of a different agent that has a lower vaporization heat. This can somehow be obtained if the concentration of sugars in the aqueous phase is increased. The concentration of the sugars in the fed can also be adjusted with evaporation previous to the dehydration as long as there is no such need for aqueous phase during the synthesis.
References F. de Ávila Rodrigues, R. Guirardello, 2012, Techno-Economic Evaluation of Large Scale 2.5Dimethylfuran Production from Fructose. Chap. 17 In Advances in Chemical Engineering Intech M.A. Duran, I.E. Grossmann, 1986, Simultaneous optimization and heat integration of chemical processes. AIChE, Journal, 32, 123-138 F.K. Kazi, A.D. Patel, J.C. Serrano-Ruiz, J.A. Dumesic, R.P. Anexa, 2011, Techno-economic analysis of dimethylfuran (DMF) and hydroxymethylfurfural (HMF) production from pure fructose in catalytic processes Chemical Engineering Journal 169, 329–338 G. Marcotullio, W. De Jong, 2010, Chloride ions enhance furfural formation from D-xylose in dilute aqueous acidic solutions Green Chem., 12, 1739–1746 M. Martín, I.E. Grossmann, 2012, Energy optimization of lignocellulosic bioethanol production via Hydrolysis of Switchgrass. AIChE Journal, 58, 5, 1538-1549 M. Martín, I.E. Grossmann, 2013. Optimal engineered algae composition for the integrated simultaneous production of bioethanol and biodiesel AIChE Journal, 59, 8, 2872–2883 Y. Roman-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, 2007, Production of dimethylfuran for liquid fuels from biomass-derived carbohydrates. Nature, 447, 982-986 R.K. Sinnot, 1999, Coulson and Richardson, Chemical Engineering 3ªEd. Singapore: Butterworth Heinemann, 1999 T.F. Yee, I.E. Grossmann, 1990, Simultaneous optimization models for heat integration. II. Heat exchanger networks synthesis. Computers and Chemical Engineering, 28, 1165-1184