Economic and environmental assessment of n-butanol production in an integrated first and second generation sugarcane biorefinery: Fermentative versus catalytic routes

Applied Energy 160 (2015) 120–131

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Economic and environmental assessment of n-butanol production in an integrated first and second generation sugarcane biorefinery: Fermentative versus catalytic routes L.G. Pereira a,⇑, M.O.S. Dias b, A.P. Mariano c, R. Maciel Filho a,c, A. Bonomi a,c a Brazilian Bioethanol Science and Technology Laboratory (CTBE), Brazilian Center of Research in Energy and Materials (CNPEM), Caixa Postal 6192, CEP 13083-970, Campinas, São Paulo, Brazil b Institute of Science and Technology, Federal University of São Paulo (ICT/UNIFESP), São José dos Campos, São Paulo, Brazil c School of Chemical Engineering, University of Campinas (UNICAMP), Campinas, São Paulo, Brazil

h i g h l i g h t s  Financial and environmental impacts of n-butanol production were investigated.  Analysis showed promising economic results for ABE fermentation scenarios.  Ethanol catalysis to butanol presented discouraging figures.  n-Butanol use as fuel demonstrated favorable GHG emissions results.

a r t i c l e

i n f o

Article history: Received 29 July 2015 Received in revised form 26 August 2015 Accepted 12 September 2015

Keywords: n-Butanol ABE fermentation Ethanol catalysis Life cycle assessment Risk analysis

a b s t r a c t n-Butanol produced from renewable resources has attracted increasing interest, mostly for its potential use as liquid biofuel for transportation. Process currently used in the industry (Acetone–Butanol–Ethanol fermentation – ABE) faces major technical challenges, which could be overcome by an alternative production through ethanol catalysis. In this study, both routes are evaluated by means of their financial viabilities and environmental performance assessed through the Virtual Sugarcane Biorefinery methodological framework. Comparative financial analysis of the routes integrated to a first and second generation sugarcane biorefinery shows that, despite the drawbacks, ABE process for fermentation of the pentoses liquor is more attractive than the catalysis of ethanol to n-butanol and co-products. n-Butanol use as fuel demonstrated favorable environmental results for climate change as figures showed over 50% reduction in greenhouse gas emission compared with gasoline. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Different biorefinery routes for the production of biofuels and chemicals are envisioned today. Among biofuels, ethanol has been extensively studied and used in large scale in countries such as Brazil and the US. Other compounds may be used as fuels as well; n-butanol has received a lot of attention and seems to be preferred over other alcohols, because of its superior fuel properties in comparison to ethanol [1] and due to its characteristics it is considered as a drop-in biofuel for Carnot cycle engines. Additionally, nbutanol is an important feedstock for the chemical industry, being ⇑ Corresponding author. Tel.: +55 19 3518 3197; fax: +55 19 3518 3104. E-mail addresses: (L.G. Pereira).

[email protected],

http://dx.doi.org/10.1016/j.apenergy.2015.09.063 0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.

[email protected]

used in the production of paint, solvents and plasticizers [2] with a projected global market of 5 million tonnes in 2018 [3]. China is the major consumer with 34% of the world demand, followed by Europe and North America with 25% and 24%, respectively [4]. Studies have pointed out the advantages of n-butanol as fuel in comparison with ethanol: n-butanol contains a longer hydrocarbon chain being more similar to gasoline (both are hydrophobic); may be mixed with gasoline and diesel at higher proportions [5,6]; Mendez et al. [7] concluded that blends of butanol with jet fuel present promising performance; according to Tao et al. [8], butanol fuel efficiency may equal that of gasoline; it may be used as oxygenate to allow more complete combustion, reducing carbon monoxide emissions [9]; it is capable of performing better for an

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

engine’s cold start and may also be used as an additive to ethanol for that function [10]. Most of the n-butanol currently produced in the world is derived from oil. The development of renewable chemicals to replace fossil-derived feedstock in the chemical industry is essential, considering the forecasted depletion of fossil resources and greenhouse gas emission target [11]. Production from renewable resources usually considers the ABE (acetone–butanol–ethanol) fermentation of sugars, such as sucrose extracted from sugarcane. Recent studies have evaluated the ABE process feasibility for other feedstocks such as eucalyptus hydrolysate [12], wastewater microalgae [13] and starchy food wastes [14]. The interest renewable butanol has attracted from the chemical industry is confirmed by the construction of pilot plants and the planning of industrial scale units for production of butanol from fermentation of sugars around the world [15]. Although this process has been extensively studied, it faces major technical challenges: low butanol titers derived from microorganism inhibition towards the product leading to high recovery costs and extremely high water usage, with consequent large stillage production, and low solvent productivity [16–18]. Mutant strains able to tolerate higher butanol concentrations have been developed, but their use in the Brazilian sugarcane industry is viewed with caution, since aseptic conditions must be created, which will certainly increase investment and operational costs [19]. An alternative process that could overcome some of these drawbacks for the production of n-butanol from sugarcane is the alcoholchemical route, in which ethanol is used as feedstock in catalyzed reactions. Several studies detail the development of catalysts for ethanol conversion into butanol at laboratory scale [1,2,20–25]. These studies, however, are often limited to the evaluation of the catalyst performance, focused on improving butanol selectivity, since catalytic reactions lead to simultaneous production of various chemicals besides n-butanol. Results available in the literature provide little information on the use of catalysts at industrial scale. Therefore, it is unknown whether their use in ethanol conversion processes provides advantages, especially in the Brazilian context, where ethanol is largely used as a fuel and has a consolidated market [26]. Second generation ethanol production can significantly increase ethanol production and has been studied over the past decades, but remains facing technical and economic challenges [27]; Macrelli et al. [28]. The concept of a biorefinery, which consists of maximized biomass conversion efficiency into various products (fuels, chemicals, materials, and energy) to improve its competitiveness against fossil-derived products [29], should be applied to sugarcane processing into second generation ethanol as well. Inclusion of butanol and chemicals production in first and second generation biorefineries can add flexibility to their product portfolio and possibly increase their revenues [30]. Process simulation can be used to evaluate biorefinery configurations taking into consideration the complexity of the process regarding technological routes, product portfolio, and biomass source, among others; it also allows the comparison of different process configurations and their impacts on the entire production process, which would be much harder to evaluate using only experimental data. Additionally, simulation can provide data required for estimation of economic [8,17,19,31,28,32–37] and environmental performance [33,38–43]. Previous recent studies from the authors of this paper have evaluated technical aspects and economics of n-butanol production from sugarcane sugars [17,32] and ethanol catalysis [19,30,44], as well as the environmental performance of the ABE process in a sugarcane biorefinery [42]. In this study two competing technological routes for the production of n-butanol (fermentative and catalytic route) were assessed as facilities added to a sugarcane biorefinery with

121

integrated production of first and second generation ethanol. Economic feasibility and environmental performance were determined with the aid of the Virtual Sugarcane Biorefinery tool. Uncertainties assigned to technical parameters of both technological routes as well as to the selling prices of n-butanol and co-products generated enabled the analysis of the economic risk. Life cycle assessment applied under two different approaches (economic and energy allocation) allowed the comparison of the environmental impacts of the production routes investigated as well the GHG emissions reduction associated with bio-based butanol use in the automotive fuel market context. 2. Methods 2.1. Virtual sugarcane biorefinery Developed by the Division of Integrated Evaluation of Biorefineries of the Brazilian Bioethanol Science and Technology Laboratory (CTBE) from the Brazilian Center for Research on Energy and Materials (CNPEM), the Virtual Sugarcane Biorefinery (VSB) is a comprehensive tool able to evaluate, from a sustainability standpoint, different biorefinery configurations [45,46]. The methodological framework integrates simulation platforms and assessment methods with the objective of identifying and evaluating technical parameters and sustainability impacts (economic, social and environmental) related to the introduction of new technologies (cellulosic ethanol and green chemistry products) in current Brazilian sugarcane biorefineries. The construction of this tool is directly focused on key scientific and technological aspects of future biorefineries, requiring the elaboration of mathematical models to be introduced in the simulation platforms. The term virtual refers to the fact that it can predict/calculate parameters of various biorefinery alternatives/ routes, without the need of performing tests at industrial scale. The assessment of the viability and level of success of a biorefinery alternative depends on the financial, socioeconomic and environmental impact indicators, calculated through different methods. The approach proposed has already been successfully applied to various case studies involving sugarcane biorefineries, such as the evaluation of annexed and autonomous biorefineries [46], integrated and stand-alone second generation ethanol facilities [31], ethanol and n-butanol production [17,19,30,32, 33,42,44] and anaerobic digestion of vinasse [47].

2.2. Process description and scenarios definition A summary description of the five scenarios investigated in this study is shown as follows:  Scenario 1G2G: production of anhydrous ethanol and surplus electricity in an integrated first and second generation sugarcane biorefinery;  Scenario ABEW: ABE fermentation with regular microorganism (Clostridium saccharoperbutylacetonicum DSM 2152) with a butanol yield of 0.20 g g 1 of sugars;  Scenario ABEM: ABE fermentation with genetically modified strain (Clostridium beijerinckii BA101) with improved butanol production of 0.34 g g 1 of sugars;  Scenario ALCD: catalysis of anhydrous ethanol for the production of n-butanol and co-products using [RuCl(g6-p-cymene) (bis(diphenylphosphanyl)-methane)]Cl catalyst;  Scenario ALCT: catalysis of anhydrous ethanol for the production on n-butanol and co-products using hydroxyapatite catalyst with Ca/P ratio of 1.67 in six reactors coupled in a seriesparallel arrangement.

122

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

This study considers a sugarcane ethanol plant with integrated first and second generation processes and power cogeneration. Simulations of the scenarios were carried out using the software Aspen Plus. The sugarcane biorefinery processes 500 tonnes of sugarcane (wet basis) per hour (2 million tonnes of sugarcane per year). As the current design of existing sugarcane mills, all the electricity and thermal energy required by the process is produced by combustion of bagasse and straw. This study considers an optimized autonomous sugarcane distillery (i.e., dedicated to ethanol production, which means no sugar production) with reduced steam consumption and efficient cogeneration system, which is fed with bagasse and straw produced in the field (considering 50% recovery). In second generation ethanol production, surplus bagasse is used as feedstock through pretreatment and hydrolysis (assuming all straw is used as feedstock). Thus, only a fraction of the bagasse is burnt for the production of steam and electricity, depending on the energy demand of the process. Surplus bagasse therefore undergoes pretreatment (acid catalyzed steam explosion, 150 °C, 10 min, 0.5 wt% H2SO4) with conversion of hemicellulose to pentoses and furfural of 65% and 10%, and of cellulose to glucose and hydroxymethylfurfural (HMF) of 5% and 1.5%, respectively; pretreated material is separated from pentoses liquor (sent to C5 fermentation) and hydrolyzed (15 wt% solids, 50 °C, 48 h, 10 FPU g 1 cellulose), with conversion of cellulose to glucose of 70% and hemicellulose to pentoses of 35%; unreacted solids are used as fuel in cogeneration while hexoses are fermented mixed with sugarcane juice. Fig. 1 shows a simplified scheme of the processes considered in this study for both technological routes (ABE fermentation and ethanol catalysis). The main values adopted for parameters of the integrated first and second generation ethanol production process simulation are listed in Table 1. Additional details can be found elsewhere [19,31,33,45]. 2.2.1. ABE fermentation Pentoses liquor obtained from the pretreatment of bagasse is sent to the butanol plant where it undergoes ABE fermentation.

Steam Electricity

Sugarcane

Cleaning Extraction

Straw Bagasse

Cogeneration

Juice treat/conc

Hydrolysis

Pre-treatment C6 liquor

Fermentation Distillation Rectification

Pentoses (II)

(II)

Dehydration

C5 Fermentation Ethanol Catalysis

(I)

(II)

Ethanol n-Butanol Acetone / Furfural Ethanol n-Butanol Co-products

(I) Ethanol

Fig. 1. Block flow diagram of a sugarcane ethanol plant with integrated first (1G) and second (2G) generation processes and power cogeneration. Butanol production is conducted either via ABE fermentation from C5 sugars (I) or ethanol catalysis (II).

Table 1 Main values for parameters in the simulation of the integrated first and second generation ethanol production process. Parameter

Value

Unit

Sugarcane straw recovered (dry basis) Sugarcane apparent sucrose (Pol) (wet basis) Sugarcane fibers content (wet basis) Sugarcane bagasse/straw cellulose content (dry basis) Sugarcane bagasse/straw hemicellulose content (dry basis) Sugarcane bagasse/straw lignin content (dry basis) Efficiency of juice extraction in the mills Fermentation efficiency (C6 sugars) Fermentation efficiency (C5 sugars) Ethanol content of the wine fed in the distillation columns Anhydrous ethanol puritya Fraction of bagasse for start-up of the plant 65 bar boiler efficiency – LHV basis 65 bar steam temperature Turbines isentropic efficiency Generator efficiency Steam pressure – process Steam pressure – molecular sieves Process electric energy consumption Molecular sieves steam consumption Steam explosion – temperature/acid concentration

70 14.0 12.7 47.0

kg t cane % % %

25.2

%

25.2 95.9 90 80 8.5

% % % % °GL

99.6 5 87 485 85 98 2.5 6 30 0.6 150/ 0.5 65

wt% % % °C % % bar bar kW h t cane 1 kg steam L 1b °C/wt%

8.5 15/48

% %/h

70

%

Steam explosion – hemicellulose to xylose conversion Steam explosion – cellulose degradation Enzymatic hydrolysis – solids content/reaction time Enzymatic hydrolysis – cellulose to glucose conversion a b

1

%

ANP [48]: minimum ethanol purity for anhydrous ethanol. L: liters of anhydrous ethanol.

The pentoses stream contains around 100 g sugars L 1 (20% glucose, 70% xylose and 10% oligomers). This liquor is diluted to 50–60 g L 1 sugars in order to avoid butanol inhibition to microorganisms [49]. The diluted feed stream is then continuously sterilized (100 °C) and sent to fermentation. The fermentation process using Clostridium cells consists of a continuous cell production stage (seed fermentors) and a second batchwise fermentation stage where acetone, n-butanol, ethanol (ABE) are produced [50]. As such, the pentoses liquor stream is split in two streams, and the smaller is fed to the seed fermentors after being further diluted to 20 g L 1 and sterilized at 130 °C [32]. The separation of the fermentation products (acetone, n-butanol and ethanol) takes place in a series of five continuous distillation columns, with the last two being responsible for the separation of butanol from water [49]. The water stream is recycled to the fermentation unit for juice dilution and the ethanol stream is sent to the ethanol distillation unit for water removal. A detailed description of the distillation unit can be found in Mariano and Maciel Filho [18]. Main performance parameters considered for ABE fermentation and associated triangular probability distributions assigned for the risk analysis are presented in Table 2. These parameters were taken from Mariano et al. [32] and are based on experimental data obtained from studies on fermentation of hydrolysates of different biomass feedstocks to ABE [51–57]. The fact that naturally-occurring Clostridium species can produce n-butanol from the fermentation of pentoses, which is an important fraction (around 25%) of the sugars available in straw and bagasse, represents an advantage over the existing fermentation to ethanol. Although engineered microorganisms able to ferment pentoses to ethanol have been developed, so far none

123

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131 Table 2 Point values and triangular probability distributions assigned to key process parameters of ABE fermentation (%).

a b

Parameter

ABEW

Fermentation Total conversion of pentoses C5 to ethanol C5 to n-butanol C5 to acetone C5 to butyric acid C5 to acetic acid C5 to cells

80.0 5.6 40.9 20.7 3.1 4.7 12.1

Recovery Ethanol n-Butanol Acetone Furfural

51.4 97.6 96.4 28.6

Purity Ethanola n-Butanol Acetone Furfural

84.4 99.0 81.2 92.4

Triangular distributionb

(5.0; 5.6; 6.2) (36.8; 40.9; 45.0) (18.6; 20.7; 22.8)

(30.0; 51.4; 80.0) (90.0; 97.6; 100.0) (80.0; 96.4; 100.0) (0.0; 28.6; 50.0)

ABEM 90.0 5.6 67.2 18.6 6.0 0.0 0.0 54.8 96.9 96.3 40.7

Triangular distributionb

(5.0;5.6;6.2) (60.5;67.2;73.9) (16.7; 18.0; 20.5)

(30.0; 54.8; 80.0) (90.0; 96.9; 100.0) (80.0; 96.3; 98.0) (0.0; 40.7; 50.0)

77.1 99.0 75.4 97.2

Ethanol is mixed with the anhydrous ethanol stream (from first and second generation ethanol production). Triangular distribution (minimum; most likely; maximum).

can outperform the yield and productivity achieved with Saccharomyces cerevisae fermenting hexoses [58].

Table 3 Point values and triangular probability distributions assigned to key process parameters of ethanol catalysis (%).

2.2.2. Ethanol catalysis In the scenarios where ethanol is catalytically converted to nbutanol and co-products, pentoses and hexoses obtained from bagasse are separately fermented to ethanol, and both first and second generation ethanol are used as feedstock. Ruthenium [20,59] and hydroxyapatite-based catalysts [24] were considered for the conversion of ethanol to n-butanol and co-products:  [RuCl(g6-p-cymene)(bis(diphenylphosphanyl)-methane)]Cl – 22.1% conversion in a single step batch reaction conducted for 4 hours at 150 °C. In the reactor, ethanol is converted to nbutanol (93.6% selectivity and 20.1% yield), 2-ethylbutanol (3.2% selectivity and 1.1% yield) and n-hexanol (5.1% selectivity and 1.7% yield) [59].  Hydroxyapatite (HAP-4 with Ca/P ratio of 1.67) – 20% ethanol conversion with 69.8% n-butanol selectivity at 298 °C, with a reaction time of 1.78 s. 2-ethylbutanol, n-hexanol and unsaturated C4 alcohols are the major co-products, with selectivity of 7.0%, 5.8% and 3.5%, respectively. At minor concentrations, 2-ethylhexanol and n-octanol are also produced [25]. Taking advantage of the low reaction time, the catalytic reactors were considered to be arranged in a series-parallel configuration, as described in a previous work [19]. In this configuration, anhydrous ethanol is split into six streams, and each of them is fed to a different reactor. Products obtained in the first reactor are mixed with fresh ethanol feed and both are fed in the second reactor, and so on until the sixth reactor, which has, therefore, the largest capacity. The same selectivities and conversion provided by the literature are assumed for each reactor. This configuration was not used in the case of the ruthenium-based catalyst because of the high reaction time, which would lead to an unreasonable investment cost. Main performance parameters considered for the ethanol catalysis plants and triangular probability distributions assigned for the risk analysis are presented in Table 3. For both catalysts, reaction products are cooled and sent to the purification system comprised of absorption and distillation columns and adsorption for products purification. Distillation

Parameter

Value

Triangular distributiona

ALCD reaction Ethanol – n-Butanol conversion Ethanol – n-Hexanol conversion Ethanol – 2-Ethylbutanol conversion

20.1 2.1 1.3

(17.5; 22.1; 24.3) (0.0; 2.1; 4.0) (0.0; 1.3; 2.0)

Recovery Ethanol n-Butanol n-Hexanol 2-Ethylbutanol

99.4 95.3 98.5 98.2

(98.0; (90.0; (80.0; (80.0;

20.0 87.4 54.7

(18.0; 20.0; 22.0) (78.6; 87.4; 96.1) (49.2; 54.7; 60.1)

2.8 1.6 99.9

(2.5; 2.8; 3.1) (1.4; 1.6; 1.7) (90.0; 99.9; 100.0)

97.1 95.3 98.5 87.9 96.4 99.0

(90.0; (90.0; (80.0; (80.0; (80.0; (80.0;

ALCT reaction Total ethanol conversion Ethanol – n-Butanol conversion (2:1) Ethanol + n-Butanol – 2-Ethylbutanol conversion (1:1) n-Butanol – 2-Ethylhexanol conversion (2:1) n-Butanol – n-Octanol conversion (2:1) Ethanol + n-Butanol – n-Hexanol conversion (1:1) Recovery Ethanol n-Butanol n-Hexanol 2-Ethylbutanol 2-Ethylhexanol n-Octanol a

99.4; 95.3; 98.5; 98.2;

97.1; 95.3; 98.5; 87.9; 96.4; 99.0;

100.0) 100.0) 100.0) 100.0)

100.0) 100.0) 100.0) 100.0) 100.0) 100.0)

Triangular distribution (minimum; most likely; maximum).

columns recover both unreacted ethanol (which is dehydrated using molecular sieves) and n-butanol. 2-ethylbutanol is separated from n-hexanol using selective adsorption, producing streams with >98 wt% purity. When the hydroxyapatite catalyst is used, small amounts of 2-ethylhexanol and n-octanol are also produced. These isomers are separated using selective adsorption, producing streams with >96 wt% purity. In addition to differences in reaction time and co-products, the hydroxyapatite catalysis takes place at 298 °C and the rutheniumbased catalysis is conducted at 150 °C. Thus, whereas the former operating condition requires the use of high pressure steam (40 bar), the 6-bar steam already available at the sugarcane facility (used for ethanol vaporization in the molecular sieves unit) suffices

124

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

to meet the thermal requirement of the latter. As a result, the configuration of the cogeneration system is slightly different for each catalyst evaluated, and so is the amount of sugarcane bagasse available for use as feedstock. 2.3. Financial analysis The financial viability of each scenario is assessed by means of internal rate of return (IRR) and net present value (NPV). In the case of the former, results are compared against a minimum attractive rate of return (MARR) of 12%, which is suitable for the Brazilian context. Following the principles of engineering economics, a cash flow is projected for each scenario, taking into account the initial investment and all expenses, and revenues for an expected project lifetime. The main expenses and revenues are calculated based on the mass and energy balances provided by the process simulation, as well as on costs and prices found in specialized literature. A discounted cash flow (DCF) analysis is used to calculate IRR and NPV of the scenarios considering the Brazilian context: project lifetime is 25 years, requiring two years for construction and startup of the plant; linear 10-year depreciation is considered. Income taxes account for 34% of the net income. 2.3.1. Investment Total investment cost (TIC) was estimated based on data provided by specialists, engineering and equipment companies, and the literature [17,19,30–33,46]. The calculation of the TIC needed for the different sections of the biorefinery was based on the following references:  Conventional autonomous distillery: TIC is based on quotes provided by Dedini, which is a major equipment manufacturer for the ethanol industry in Brazil.  Second generation ethanol plant: TIC is based on data from CGEE [60], considering estimates for an advanced hydrolysis technology (i.e., performance parameters expected in future), as detailed in a previous work [61].  ABE fermentation plant: TIC was estimated based on data from Mariano et al. [32], using as basis the amount of pentoses processed.  Ethanol catalysis plant: investment in each equipment were obtained from National Renewable Energy Laboratory (NREL) reports [62,63], and factored according to mass and energy balances obtained in the process simulation. Investment cost was correlated with capacity according to the capacity power law expression. Scaling factors not found in the NREL reports were considered equal to 0.6 as per Dias et al. [61]. All investment values were updated to June 2014 according to the Brazilian inflation rate (IPCA), except for the ethanol catalysis plant, whose investment was updated according to the Chemical Engineering Plant Cost Index (CEPCI). TIC for each scenario and investment breakdown are presented in Table 4. Investment in the first generation plant is slightly higher in the scenarios with ethanol catalysis because of a higher steam consumption, which increases investment in the cogeneration section

Table 4 Investment for the investigated scenarios (exchange rate Jan-Jun/2014 US$ 1.00 = R$ 2.30). Investment (MM US$)

1G2G

ABEW ABEM

ALCD

ALCT

First generation base plant Second generation plant n-Butanol plant Total

230 78 – 308

227 78 9 314

237 55 47 340

240 46 38 324

and decreases the investment in the second generation plant, since less bagasse is available for use as feedstock. 2.3.2. Feedstock, enzyme, and other costs As for the feedstock, a mill gate price of US$ 24.90 per wet tonne of sugarcane was assumed [64]. Sugarcane straw price on a dry basis was considered to be 92.8% of the sugarcane price as per Cardoso et al. [65]. Additionally, the cost of the cellulase enzyme for use in the second generation ethanol production was assumed to be US$ 0.10 per liter of ethanol produced [66] and costs related to other inputs were considered US$ 0.85 per tonne of processed sugarcane [45]. 2.3.3. Products wholesale prices Wholesale prices of anhydrous ethanol and renewable electricity were taken from Brazilian official databases and updated according to the IPCA index. The selling prices of chemicals were retrieved from various international references. The effects of the intrinsic uncertainties of n-butanol and co-products prices were considered in the economic model via Monte Carlo simulation. As such, triangular probability distributions were assigned according to variations reported in the referenced literature (Table 5). 2.4. Life cycle assessment There are two main objectives of the LCA in this evaluation. First, to assess the environmental performance of the technological scenarios investigated for the production of bio-based n-butanol in terms of local and global impact categories. Second, to compare the greenhouse gas emissions of the use of butanol as a vehicle fuel against gasoline. SimaPro v.7.3 LCA modeling software was used to develop and link unit processes. The Ecoinvent database was used for materials and processes that were not developed by the authors. Boundaries and functional units for the LCA study were established according to the objectives proposed. In the case of the comparison of butanol producing scenarios, boundaries were set from field to gate and the functional unit defined as one kg of nbutanol. For the greenhouse gas emissions of the use of butanol as a fuel, limits were considered from field to wheel, and the functional unit was one km traveled by a gasoline dedicated engine car.

Table 5 Products wholesale prices (exchange rate Jan-Jun/2014 US$ 1.00 = R$ 2.30). Product

Price

Unit

Anhydrous ethanola Electricityb n-Butanolc 2-Ethylbutanold 2-Ethylhexanole n-Hexanolf n-Octanolg Acetoneh Furfurali

0.74 58.22 0.94 1.23 1.38 1.38 1.75 1.40 1.40

US$ US$ US$ US$ US$ US$ US$ US$ US$

Triangular distributionj kg 1 MW h kg 1 kg 1 kg 1 kg 1 kg 1 kg 1 kg 1

1

– – (0.74; 0.94; 1.32) (1.15; 1.32; 1.45) (1.33; 1.38; 1.43) (1.33; 1.38; 1.43) (1.00; 1.75; 2.00) 1.00; 1.40; 1.50) (1.20; 1.40; 1.50)

a Price for anhydrous ethanol in the state of Sao Paulo considering the variation in the period 2003–2013 [67]. b Price based on renewable energy auctions (2005–2013) [68]. c Minimum price assumed to be the same as anhydrous ethanol’s price per mass; most likely price assumed to be the same as anhydrous ethanol’s price per energy; and maximum price considering n-butanol as chemical [69]. d Nominally assumed to be the same as n-butanol’s price as chemical [69]. e Orbichem [70]. f Nominally assumed to be the same as 2-ethylhexanol’s price. g CNCIC [71]. h Orbichem [72]. i Observed and projected prices (2009-2020) [73]. j Triangular distribution (minimum; most likely; maximum).

125

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

Life cycle inventory data for the agricultural stage of sugarcane production were generated by the CanaSoft model developed by CTBE. Details on the methodological approach of the model and the assumptions considered for the definition of the agricultural inventory can be found elsewhere [42,45]. Because the focus of this study is placed on the comparison of technological routes (industrial processes), main parameters for the agricultural stage were assumed as average values representative for sugarcane production in the state of Sao Paulo, which is the largest producer of sugarcane and ethanol in Brazil [74]. Mechanical harvest was assumed for all scenarios, since it presents higher efficiency than manual harvesting and is a current trend in the South-Central region of Brazil due to environmental regulations. The loading of sugarcane is performed using infield transport, and it is transported from field to the mill by trucks with combination of two semi-trailers connected by a two-axle dolly with loading capacity of 60 tonnes, called ‘‘rodotrens”, for an average distance of 25 km. Table 6 depicts selected key parameters for sugarcane production considered in this study. Life cycle inventory for the conversion processes were based on process modeling outputs from AspenPlus. Table 7 shows key parameters of the conversion stage per tonne of processed sugarcane. The main differences observed in the industrial inputs are related to sulfuric acid, ammonia, water, zeolites and cellulase enzyme:

as more sugars to be fermented. In ethanol catalysis scenarios more bagasse and straw are used to generate steam and power, instead of being diverted to pretreatment, in order to meet the additional demand of energy in the catalysis plant. (ii) Water consumption of the first generation process is assumed to be 1.5 cubic meters per tonne of processed sugarcane, based on data for the sugarcane industry in Brazil provided by ANA [76]. For the second generation process, water used in the pretreatment and hydrolysis operations (mostly for washing of pretreated material and dilution of cellulose prior to hydrolysis) is taken into account. Additional water consumption was assumed for the butanol production process: C5 liquor dilution and inoculum production for ABE fermentation and cooling water for alcoholchemical scenarios. (iii) Zeolites are used in the molecular sieves in the dehydration stage. Since the zeolite load depends on the volume of anhydrous ethanol produced, alcoholchemical scenarios use greater amounts of zeolites. This happens because, in addition to ethanol already produced by fermentation of C6 and C5 sugars, another stage of dehydration is needed in the ethanol catalysis plant. (iv) Enzymatic hydrolysis of the lignocellulosic material is characterized by the use of cellulase. Therefore the lower use of the enzyme is explained by the lower amount of available bagasse for second generation in the catalysis scenarios.

(i) Sulfuric acid is used for the pretreatment stage as well as for yeast treatment in the fermentation process. The higher amounts for scenarios 1G2G and ABE are explained by the fact that there is more lignocellulosic material (from bagasse) available to be pretreated in these cases, as well

Besides n-butanol and anhydrous ethanol, the biorefinery also produces acetone (co-product of the ABE fermentation), furfural (byproduct of pretreatment and recovered from the pentoses liquor), 2-ethylbutanol, 2-ethylhexanol, n-hexanol, n-octanol (co-product of the ethanol catalysis), and power. In this case, it

Table 6 Key parameters for sugarcane production. Parameter

Value

Unit

Observation

Productivity Vinasse application Nitrogen application

85 186 30 90 120 180 120 150

Mg ha 1 y 1 m3 ha 1 kg N ha 1 kg N ha 1 kg N ha 1 kg P2O5 ha 1 kg K2O ha 1 kg K2O ha 1

Considering a five-year cycle [75] Dose in the field during ratoon cultivation For plant cane For ratoon with vinasse application For ratoon with no vinasse application For plant cane For plant cane For ratoon with no vinasse application

146 47 24 8

kg diesel ha kg diesel ha kg diesel ha kg diesel ha

Phosphorus application Potassium application Diesel use Agricultural machinery Sugarcane and straw transport Vinasse transport Inputs transport a

1

Calculated by CanaSofta, according to fuel consumption for agricultural operations

1 1 1

Sugarcane production model in VSB platform.

Table 7 Key parameters for the industrial process (per tonne of processed sugarcane). Inputs

Unit

1G2G

ABEW

ABEM

ALCD

ALCT

Sulfuric acid Calcium oxide (quicklime) Phosphoric acid Inorganic chemicals Cellulase enzyme Zeolites Water Lubricating oil Steel equipment Stainless steel equipment Concrete Buildings

g g g g g g m3 g kg kg m3 m2

1115.71 817.44 193.51 8.26 393.44 36.10 1.95 13.04 2.57E 01 1.50E 02 3.44E 04 7.20E 05

1116.00 817.44 193.51 10.04 386.78 33.45 2.09 13.04 2.62E 01 1.53E 02 3.51E 04 7.34E 05

1112.31 817.44 193.51 9.27 384.57 33.43 2.14 13.04 2.62E 01 1.53E 02 3.51E 04 7.34E 05

815.43 817.44 193.51 9.13 220.08 57.35 1.68 13.04 2.84E 01 1.66E 02 3.80E 04 7.95E 05

712.84 817.44 193.51 8.67 163.22 45.95 1.81 13.04 2.70E 01 1.58E 02 3.62E 04 7.57E 05

126

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

bears noting that the allocation procedure in a multiproduct process is a critical issue in LCA. The ISO documents [77,78] recommend avoiding allocation whenever possible either through subdivision of certain processes or by expanding the system limits to include the additional functions related to them. If avoiding allocation is not possible, recommendation is the use of methods that reflect the physical relationship such as mass and energy content or the use of other relevant variables to allocate, such as the economic value of products, which is similar to the cost allocation methods in managerial accounting [79]. Cherubini et al. [80] and Luo et al. [81] have discussed the effect of different allocation in LCA studies with impact values varying up to 80% for the same product in face of the method chosen. Additionally, [82] point out that the choice of allocation method in policy directives has large influence on the outcomes: the uncertainty in the results (especially for global warming) hampers the optimal use of LCA in the policy context. Thus, two allocation criteria were adopted in this study: (i) Economic: the splitting is based on economic properties of the multifunctional process, such as the gross sales value or the expected economic gain. This method has been adopted by most of the studies carried out by the VSB team of CTBE for the comparison of different technological scenarios. Prices adopted for products generated in the sugarcane biorefinery are shown in Table 4. (ii) Energy (with co-products displacement/substitution): energy partitioning is applied to the main products generated (anhydrous ethanol and n-butanol), considering the lower heating value of 34.32 MJ kg 1 for n-butanol and 21.2 MJ L 1 for ethanol [5]. All co-products are considered as solvents and feedstocks for the chemical industry, therefore they are treated as avoided products using the product displacement method as discussed by Wang et al. [83]. Co-product displacement (also termed system boundary expansion or substitution) is based on the concept of displacing the existing product (usually oil-based) with the new product (bio-based). For example, acetone produced in the biorefinery process is considered as a renewable chemical: one kg of acetone produced was assumed to displace one kg of acetone from petroleum-based production. Scenarios investigated also generated a surplus of electricity, which is assumed to provide a credit. Applying the same co-product displacement method, the excess electricity displaces an equivalent amount of grid electricity, thus avoiding GHG emissions. A database was created for the average Brazilian electricity mix in 2013 in the SimaPro software for that purpose [84]. Co-products and electricity displacements for GHG emissions and fossil energy considered in this study are given in Table 8.

Table 8 GHG emissions and fossil energy consumption for production of oil-based coproducts and Brazilian electricity grid as modeled in SimaPro.

a

Product displaced

GHG emissions

Acetone Furfurala 2-Ethylbutanolb n-Hexanolc Brazilian electricity mixd

2.22 kg 5.73 kg 2.61 kg 3.09 kg 0.25 kg

CO2-eq kg 1 CO2-eq kg 1 CO2-eq kg 1 CO2-eq kg 1 CO2-eq kW h

Fossil energy

1

64.8 MJ kg 1 110.0 MJ kg 1 77.7 MJ kg 1 87.0 MJ kg 1 2.17 MJ kW h

1

Assumed as Tetrahydrofuran (THF) displacement. Assumed as 1-butanol displacement. c Assumed as Cyclohexanol displacement. d Hydropower (70.6%), natural gas (11.3%), biomass (7.6%), oil (4.4%), nuclear power (2.6%), coal (2.6%), wind (1.1%) [84]. b

For the comparison of the greenhouse gas emissions from the use of n-butanol as a vehicle fuel, the inventory for gasoline production was taken from the Ecoinvent database version 2.2 and adapted to some extent for the Brazilian market mix [85]. Distances for road transportation and distribution of fuels inside the country were estimated based on the average distance of main production units to the biggest consumption center: 340 km for butanol from sugarcane (Northeast of the state of Sao Paulo to the city of Sao Paulo) and 120 km for gasoline (from Paulinia, where the largest oil refinery of Brazil is located, to the city of Sao Paulo) as proposed by Pereira et al. [33]. Engine energy efficiency of 3.46 MJ km 1 was retrieved from statistics provided by the Brazilian Ministry of Environment [86]. The value represents an average efficiency for the Brazilian fleet in 2009 for vehicles with gasolinededicated engines. The quantity of n-butanol used per kilometer was estimated by means of its energy content of 34.32 MJ kg 1 [5]. In order to compare the environmental performance of the technological scenarios investigated, the LCA impact assessment methodology ReCiPe Midpoint (H) v1.05 was used. This methodology was preferred because, among other methods, it presents the broadest set of midpoint impact categories, including local and global impact categories.

3. Results and discussion 3.1. Process simulation The products output (given per tonne of processed sugarcane) for each biorefinery scenario is presented in Table 9. The largest production of anhydrous ethanol is obtained in the base case scenario (1G2G) since all fermentable sugars are used for that purpose. The larger production of ethanol in ABE fermentation scenarios (ABEW and ABEM) in comparison to alcoholchemical scenarios is explained by the fact that in these scenarios only the pentoses liquor is fermented to n-butanol, whereas sugarcane juice and glucose liquor are used for the production of ethanol. On the other hand, in scenarios with catalysis a portion of anhydrous ethanol is consumed to produce a greater amount of n-butanol and coproducts. In scenario ALCT, a higher conversion of ethanol to nbutanol is achieved and production of more co-products is observed. Therefore, less ethanol is obtained as a final product in comparison to the other scenarios. It is important to highlight the larger production of surplus electricity obtained in n-butanol producing scenarios in comparison to the base case; this fact is explained by the demand of steam in the butanol plants, meaning that more bagasse needs to be burned and more steam is expanded in the turbo generators in comparison to 1G2G. Additionally, in alcoholchemical scenarios (ALCD and ALCT) the steam demand is even higher than in ABE scenarios, because of the conversion of ethanol to various co-products and consequent need for steam for the separation process.

Table 9 Products output for biorefinery scenarios. Product

1G2G

ABEW

ABEM

ALCD

ALCT

Anhydrous ethanol (kg t cane 1) Surplus electricity (kW h t cane 1) n-Butanol (kg t cane 1) Acetone (kg t cane 1) Furfural (kg t cane 1) n-Hexanol (kg t cane 1) 2-Ethylbutanol (kg t cane 1) 2-Ethylhexanol (kg t cane 1) n-Octanol (kg t cane 1)

88 75 – – – – – – –

82 80 2.8 1.7 0.2 – – – –

82 80 4.4 1.7 0.3 – – – –

61 100 12.0 – – 0.2 0.5 – –

36 99 22.1 – – 2.5 2.7 0.8 0.4

127

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

3.2. Financial analysis The main financial metrics are presented in Table 10. The most attractive IRR and NPV values were found in the base case 1G2G and ABE fermentation scenarios, with their IRR exceeding the hurdle rate of 12%. Alcoholchemical scenarios presented discouraging results for both IRR and NPV. The slightly higher OPEX of 1G2G and ABE scenarios are due to the consumption of cellulase enzyme in the second generation ethanol processing. It results from the fact that more bagasse is available for hydrolysis in these cases than in the alcoholchemical scenarios. Both ABE scenarios presented the largest revenues per tonne of processed sugarcane. Analysis of the revenues breakdown is critical at this point, because one of the main hypothesis of the study is that investing in the production of n-butanol and co-products using ethanol as a feedstock would be financially attractive to an integrated first and second generation sugarcane biorefinery. Fig. 2 shows the revenues breakdown of the scenarios with the best financial results in each technology category (ABE fermentation and catalysis). Whereas in the alcoholchemical scenario (ALCT) 19% of the total revenues come from n-butanol and co-products,

Table 10 Financial metrics for investigated scenarios (exchange rate Jan-Jun/2014 US$ 1.00 = R$ 2.30). Parameter

1G2G

ABEW

ABEM

ALCD

ALCT

OPEX (MM US$ yr 1) Total investment (MM US$) Revenues (US$ t cane 1) Net Present Value (MM US$) Internal rate of return (% yr 1)

62.2 308 69.6 38.3 13.7

62.0 314 70.5 40.0 13.8

62.0 314 72.2 52.1 14.3

60.0 340 63.0 27.8 10.8

60.0 324 61.1 29.1 10.7

ALCD (US$ 63.0 t cane-1 )

in the ABE fermentation scenario this figure drops to 10% as a result of only using pentose sugars as feedstock. Although a greater diversification of revenues certainly implies in enhanced robustness against market fluctuations, better revenues (US$ 72.2 t cane 1) were associated with the design with emphasis in ethanol as a product and not a feedstock (ABEM). Taking into account variations of market prices for n-butanol and co-products (presented in Table 5) and process parameters uncertainties (presented in Tables 2 and 3), and considering a 90% confidence interval (i.e., 5th percentile and the 95th percentile), the probability of the IRR of the alcoholchemical scenarios ALCT and ALCD be greater than the hurdle rate (MARR) is 11.3% and 20.3% respectively (Fig. 3). For the same confidence interval (90%), the best maximum IRR value of 14.5% is associated with the ABEM scenario, and the lowest minimum value of 9.6% is projected for the ALCT scenario. Uncertainty is higher considering both the 90% and 50% confidence intervals for the IRR of alcoholchemical scenarios, which can be explained by the fact that the IRR is more influenced in these cases by the larger production of n-butanol and the price distribution assigned to it. The probabilities of the IRR of ABEW and ABEM to be greater than the 1G2G IRR (13.7%) are 42.2% and 100%, respectively, meaning that the conversion of the pentoses liquor to n-butanol and co-products with improved butanol yield showed promising financial figures. The biggest contribution to the IRR variability was related to n-butanol price. Probability distribution assigned considered minimum and mean wholesale prices as if n-butanol was sold as fuel and maximum price if it was sold as chemical. However, it is important to highlight that n-butanol’s market in Brazil as chemical is limited to projected 100 thousand tonnes in 2015. Consider-

ABEM (US$ 72.2 t cane-1 )

1%

6% 4% Anhydrous ethanol

18%

Electricity

6%

n-Butanol 9%

Other co-products 71% 84%

Fig. 2. Revenues breakdown for scenarios ALCT and ABEM.

MARR

1G2G IRR

95th percentile

ALCT 75th percentile

ALCD

ABEM

Median

ABEW 25th percentile

9.0%

10.0%

11.0%

12.0%

13.0%

14.0%

15.0%

Fig. 3. IRR values under market price and process parameters uncertainties.

5th percentile

128

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

Fig. 4. Environmental impacts scores per kilogram of n-butanol produced (economic allocation). Maximum values (corresponding to 100%) are shown in the graph for each category. Impact units: climate change in kg of CO2-eq; ozone depletion in kg of CFC-11-eq; acidification in kg of SO2-eq; photochemical oxidation in kg of NMVOC (nonmethane volatile organic compound); particulate formation in kg of PM10-eq (particulate matter up to 10 lm in size); fossil depletion in kg of oil-eq.

ing the processing capacity of the biorefinery assumed in this study, only 2 (for the ALCT process) or 11 (for the ABEM process) plants would be required to meet that demand. In comparison, U.S. and China holds an annual combined demand of around 2.5 million tonnes [3]. On the other hand, if n-butanol is considered as fuel, it would be embedded in a much larger market: a replacement of 5% of total vehicle fuels (ethanol and gasoline) in Brazil, for instance, would generate a demand of 2.7 million tonnes of n-butanol, which would require 62 (for the ALCT process) or 312 (for the ABEM process) sugarcane biorefineries. It is worthwhile to note that Brazil has 383 certified ethanol sugarcane biorefineries [87]. 3.3. Life cycle assessment Fig. 4 presents a comparison of environmental impacts of n-butanol produced via the bio-based routes of this study. Stacked bar depicts the contribution from each stage (agricultural and industrial) for biobutanol. Results show that agricultural stage impacts are basically the same for all scenarios within the impact categories. The small differences observed are due to the

application of vinasse in the field (fertirrigation), since the amount of the waste generated varies in each process. For climate change, ozone depletion, acidification, and fossil depletion categories, most of the impacts are associated to the agricultural stage, mainly due to the use of fertilizers and the emissions from diesel used in the field. For photochemical oxidation and particulate formation categories, most of the impact score is due to the industrial stage; bagasse burning is the main contributor for both impact categories, and ethanol emissions during distillation is also important for photochemical oxidation. Scores in terms of kilograms of n-butanol show that the ABE scenarios present the best environmental results among the biobased scenarios. The gross disparity between ABE and alcoholchemical scenarios in relation to the photochemical oxidation and particulate formation categories is explained by the fact that alcoholchemical scenarios demands a higher amount of steam in the catalysis plant for the separation of co-products, which consequently increase the need of bagasse burning in the boilers and the impact for both categories. Fig. 5 presents the environmental impacts scores considering an energy allocation approach combined with system boundary

Fig. 5. Environmental impacts scores per kilogram of n-butanol produced (energy allocation with system boundary expansion). Maximum values (corresponding to 100%) are shown in the graph for both total and net impacts (between parentheses) in each category. Impact units: climate change in kg of CO2-eq; ozone depletion in kg of CFC-11-eq; acidification in kg of SO2-eq; photochemical oxidation in kg of NMVOC (non-methane volatile organic compound); particulate formation in kg of PM10-eq (particulate matter up to 10 lm in size); fossil depletion in kg of oil-eq.

129

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

kg CO2 eq km-1

Gasolineemission (EPA)

0.320 0.270 0.220

Co-products credits

0.170

Electricity credits

0.120

Transport

0.070

Use Industrial

0.020

Agriculture -0.030

Net impacts

-0.080

ABEW ABEM ALCD

ALCT ABEW ABEM ALCD

Energy allocation with substitution

ALCT

Economic allocation

Fig. 6. Field-to-wheels GHG emissions per km traveled by a gasoline engine car propelled by n-butanol.

expansion. In this case, anhydrous ethanol and n-butanol are the main products (allocated by energy content) and credits are assigned to co-products generated. Net impacts are calculated by subtracting the credits from the impacts. Results for net values differ from those obtained with the economic allocation approach. Scenario ALCD presents the lowest impacts for climate change and ozone depletion with the electricity credit playing a major role in reducing the gross impact value. For the fossil depletion category, it is clear how important the contribution of electricity and co-products credits is to reduce the impacts for the alcoholchemical scenarios. These results also highlight the relevance of the allocation procedure chosen to the life cycle assessment method. Fig. 6 shows the field-to-wheels GHG emissions for 1 km run by a gasoline-dedicated engine car using n-butanol as fuel. The two types of allocation procedures considered in this study are shown and results are compared with the GHG emitted by conventional gasoline-propelled cars of 0.32 kg CO2-eq km 1 [88]. The emissions calculated for all scenarios are significantly lower than the average baseline of GHG emissions from gasoline. Impacts are even lower if calculated according to the energy content of anhydrous ethanol and n-butanol (with credits assigned to co-products) instead of the economic allocation. Although EPA’s directive recommends the first allocation procedure, n-butanol projected use assessed by both methods are in compliance with the GHG emissions reduction of 50%, and thus can be classified as an advanced biofuel. It is noteworthy that the direct CO2 emissions at the industrial stage in the biorefinery as well as the CO2 emissions from the n-butanol use as fuel in the car are assumed as biogenic (i.e., CO2 absorbed from the atmosphere and incorporated as biomass). Therefore, with its biomass origin, biogenic CO2 does not contribute to the increase of GHG emissions in the atmosphere and is not considered in the global warming potential methodology. Typically, the emission is not accounted for as a contributor to global warming because it is assumed to be removed from the atmosphere during the same time horizon as the GWP estimate. 4. Conclusions In this study we investigated the financial viability and the environmental impacts associated with the production of n-butanol and other chemicals by alcoholchemical and sugarchemical routes integrated to a first and second generation sugarcane biorefinery.

Financial deterministic analysis showed better results for ABE scenarios in terms of revenues per tonne of processed sugarcane and economic return, with discouraging results for the alcoholchemical route. The conversion of the pentoses liquor to n-butanol and co-products with improved butanol yield showed attractive figures. On the environmental end, ABE scenarios presented the best figures in general, with lower impacts in all categories when assuming an economic allocation with impacts per kilogram of n-butanol. However, if the energy allocation method is applied (with system boundary expansion), the credits from electricity and co-products play an important role in the calculation of life cycle metrics: alcoholchemical scenarios presented reduced net impacts reaching advantageous figures in comparison to ABE scenarios for ozone and fossil depletion categories. Furthermore, n-butanol use as fuel demonstrated favorable environmental results for climate change with more than 50% reduction in GHG emission as compared with gasoline, thus complying with the standard for advanced biofuels established by the US Environmental Protection Agency (EPA). Acknowledgments The authors would like to acknowledge the financial support provided by the Sao Paulo Research Foundation (FAPESP) (Grant number 2012/15192-0) and the National Council for Scientific and Technological Development (CNPq) (Grant number 476168/2013-9). References [1] Riittonen T, Toukoniitty E, Madnani DK, Leino A-R, Kordas K, Szabo M, et al. One-pot liquid-phase catalytic conversion of ethanol to 1-butanol over aluminium oxide – the effect of the active metal on the selectivity. J. Catal. 2012;2:68–84. [2] Carvalho DL, de Avillez RR, Rodrigues MT, Borges LEP, Appel LG. Mg and Al mixed oxides and the synthesis of n-butanol from ethanol. Appl. Catal. A: Gen. 2012;415–416:96–100. [3] MicroMarket Monitor, 2015. Asia-Pacific n-butanol market by applications (butyl acrylate, butyl acetate, glycol ethers, and others) & geography – global trends & forecasts to 2019. Accessed: 05.11.2015. [4] Nexant, 2011. Green Chemicals: Growing number of chemical firms enter biobutanol space. Accessed: 07.31.2014. [5] Dürre P. Biobutanol: an attractive biofuel. Biotechnol. J. 2007;2:1525–34. [6] Wu H, Nithyanandan K, Zhang J, Lin Y, Lee TH, Lee CF, et al. Impacts of Acetone–Butanol–Ethanol (ABE) ratio on spray and combustion characteristics of ABE–diesel blends. Appl. Energy 2015;149:367–78.

130

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131

[7] Mendez CJ, Parthasarathy RN, Gollahalli SR. Performance and emission characteristics of butanol/Jet A blends in a gas turbine engine. Appl. Energy 2014;118:135–40. [8] Tao L, He X, Tan ECD, Zhang M, Aden A. Comparative techno-economic analysis and reviews of n-butanol production from corn grain and corn stover. Biofuels Bioprod. Bioref. 2014;8:342–61. http://dx.doi.org/10.1002/bbb.1462. [9] Nevin RK. Ethanol in gasoline: environmental impacts and sustainability review article. Renew. Sust. Energy Rev. 2005;9:535–55. [10] Lee SY, Park JH, Jang SH, Nielsen LK, Kim J, Jung KS. Fermentative butanol production by clostridia. Biotechnol. Bioeng. 2008;101:209–28. [11] Brehmer B, Boom RM, Sanders J. Maximum fossil fuel feedstock replacement potential of petrochemicals via biorefineries. Chem. Eng. Res. Des. 2009;87 (9):1103–19. http://dx.doi.org/10.1016/j.cherd.2009.07.010. [12] Zheng J, Tashiro Y, Wang Q, Sakai K, Sonomoto K. Feasibility of acetone– butanol–ethanol fermentation from eucalyptus hydrolysate without nutrients supplementation. Appl. Energy 2015;140:113–9. http://dx.doi.org/10.1016/j. apenergy.2014.11.037. [13] Castro YA, Ellis JT, Miller CD, Sims RC. Optimization of wastewater microalgae saccharification using dilute acid hydrolysis for acetone, butanol, and ethanol fermentation. Appl. Energy 2015;140:14–9. http://dx.doi.org/10.1016/j. apenergy.2014.11.045. [14] Ujor V, Bharathidasan AK, Cornish K, Ezejia TC. Feasibility of producing butanol from industrial starchy food wastes. Appl. Energy 2014;136:590–8. [15] Harmsen PFH, Hackmann MM, Bos HL. Green building blocks for bio-based plastics. Biofuels, Bioprod. Bioref. 2014;8(3):306–24. http://dx.doi.org/ 10.1002/bbb.1468. [16] Green EM. Fermentative production of butanol – the industrial perspective. Curr. Opin. Biotech. 2011;22(3):337–43. http://dx.doi.org/10.1016/ j.copbio.2011.02.004. [17] Mariano AP, Dias MOS, Junqueira TL, Cunha MP, Bonomi A, Maciel Filho R. Butanol production in a first-generation Brazilian sugarcane biorefinery: technical aspects and economics of greenfield projects. Bioresour. Technol. 2013;135:316–23. [18] Mariano AP, Maciel Filho R. Improvements in biobutanol fermentation and their impacts on distillation energy consumption and wastewater generation. BioEnergy Res. 2012;5(2):504–14. http://dx.doi.org/10.1007/s12155-0119172-0. [19] Dias MOS, Pereira LG, Junqueira TL, Pavanello LG, Chagas MF, Cavalett O, et al. Butanol production in a sugarcane biorefinery using ethanol as feedstock – Part I: integration to a first generation sugarcane distillery. Chem. Eng. Res. Des. 2014;92(8):1441–51. http://dx.doi.org/10.1016/j.cherd.2014.04.030. [20] Dowson GRM, Haddow MF, Lee J, Wingad RL, Wass DF. Catalytic conversion of ethanol into an advanced biofuel: unprecedented selectivity for n-Butanol. Angew. Chem. Int. Ed. Engl. 2013;52(34):9005–8. http://dx.doi.org/10.1002/ anie.201303723. [21] Marcu I-C, Tichit D, Fajula F, Tanchoux N. Catalytic valorization ofbioethanol over Cu–Mg–Al mixed oxide catalysts. Catal. Today 2009;147(3–4):231–8. http://dx.doi.org/10.1016/j.cattod.2009.04.004. [22] Ndou AS, Plint N, Coville NJ. Dimerisation of ethanol to butanol over solidbase catalysts. Appl. Catal. A: General 2003;251(2):337–45. http://dx.doi.org/ 10.1016/S0926-860X(03)00363-6. [23] Tsuchida T, Sakuma S, Takeguchi T, Ueda W. Direct synthesis of n-butanol from ethanol over nonstoichiometric hydroxyapatite. Ind. Eng. Chem. Res. 2006;45:8634–42. [24] Tsuchida T, Kubo J, Yoshioka T, Sakuma S, Takeguchi T, Ueda W. Reaction of ethanol over hydroxyapatite affected by Ca/P ratio of catalyst. J. Catal. 2008;259:183–9. [25] Yang KW, Jiang XZ, Zhang WC. One-step synthesis of n-butanol from ethanol condensation over alumina-supported metal catalysts. Chin. Chem. Lett. 2004;15(12):1497–500. [26] Kostin AM, Guillén-Gosálbez G, Mele FD, Bagajewicz MJ, Jiménez L. Design and planning of infrastructures for bioethanol and sugar production under demand uncertainty. Chem. Eng. Res. Des. 2012;90:359–76. [27] Lee SM, Jellison T, Alper HS. Systematic and evolutionary engineering of a xylose isomerase-based pathway in Saccharomyces cerevisiae for efficient conversion yields. Biotechnol. Biofuels 2014;7(1):122. http://dx.doi.org/ 10.1186/s13068-014-0122-x. [28] Macrelli S, Mogensen J, Zacchi G. Techno-economic evaluation of 2nd generation bioethanol production from sugar cane bagasse and leaves integrated with the sugar-based ethanol process. Biotechnol. Biofuels 2012;5 (1):22. http://dx.doi.org/10.1186/1754-6834-5-22. [29] Liew WH, Hassim MH, Ng DKS. Review of evolution, technology and sustainability assessments of biofuel production. J. Clean. Prod. 2014;71:11–29. http://dx.doi.org/10.1016/j.jclepro.2014.01.006. [30] Pereira LG, Dias MOS, Junqueira TL, Chagas MF, Cavalett O, Maciel Filho R, et al. Production of butanol and other high valued chemicals using ethanol as feedstock integrated to a first and second generation sugarcane distillery. Chem. Eng. Trans. 2014;37:805–10. http://dx.doi.org/10.3303/CET1437135. [31] Dias MOS, Junqueira TL, Cavalett O, Cunha MP, Jesus CDF, Rossell CEV, et al. Integrated versus stand-alone second generation ethanol production from sugarcane bagasse and trash. Bioresour. Technol. 2012;103:152–61. [32] Mariano AP, Dias MOS, Junqueira TL, Cunha MP, Bonomi A, Maciel Filho R. Utilization of pentoses from sugarcane biomass: techno-economics of biogas vs. butanol production. Bioresour. Technol. 2013;142:390–9. [33] Pereira LG, Dias MOS, Junqueira TL, Pavanello LG, Chagas MF, Cavalett O, et al. Butanol production in a sugarcane biorefinery using ethanol as feedstock - Part

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48] [49] [50] [51] [52] [53]

[54]

[55]

[56]

[57] [58]

[59]

II: integration to a second generation sugarcane distillery. Chem. Eng. Res. Des. 2014;92(8):1452–62. http://dx.doi.org/10.1016/j.cherd.2014.04.032. Quintero JA, Cardona CA. Process simulation of fuel ethanol production from lignocellulosics using aspen plus. Ind. Eng. Chem. Res. 2011;50(10):6205–12. http://dx.doi.org/10.1021/ie101767x. Tao L, Aden A, Elander RT, Pallapolu VR, Lee YY, Garlock RJ, et al. Process and techno-economic analysis of leading pretreatment technologies for lignocellulosic ethanol production using switchgrass. Bioresour. Technol. 2011;102(24):11105–14. http://dx.doi.org/10.1016/j.biortech.2011.07.051. Trippe F, Fröhling M, Schultmann F, Stahl R, Henrich E. Techno-economic assessment of gasification as a process step within biomass-to-liquid (BtL) fuel and chemicals production. Fuel Process. Technol. 2011;92(11):2169–84. http://dx.doi.org/10.1016/j.fuproc.2011.06.026. van der Merwe AB, Cheng H, Görgens JF, Knoetze JH. Comparison of energy efficiency and economics of process designs for bio-butanol production from sugarcane molasses. Fuel 2013;105:451–8. Ojeda K, Ávila O, Suárez J, Kafarov V. Evaluation of technological alternatives for process integration of sugarcane bagasse for sustainable biofuels production – Part 1. Chem. Eng. Res. Des. 2011;89(3):270–9. http://dx.doi. org/10.1016/j.cherd.2010.07.007. Seabra JEA, Macedo IC. Comparative analysis for power generation and ethanol production from sugarcane residual biomass in Brazil. Energy Policy 2011;39 (1):421–8. http://dx.doi.org/10.1016/j.enpol.2010.10.019. Sánchez OJ, Cardona CA. Conceptual design of cost-effective and environmentally-friendly configurations for fuel ethanol production from sugarcane by knowledge-based process synthesis. Bioresour. Technol. 2012;104:305–14. http://dx.doi.org/10.1016/j.biortech.2011.08.125. Spatari S, Bagley DM, MacLean HL. Life cycle evaluation of emerging lignocellulosic ethanol conversion technologies. Bioresour. Technol. 2010;101: 654–67. Pereira LG, Chagas MF, Dias MOS, Cavalett O, Bonomi A. Life cycle assessment of butanol production in sugarcane biorefineries in Brazil. J. Clean Prod. 2015;96:557–68. http://dx.doi.org/10.1016/j.jclepro.2014.01.059. Walter A, Ensinas AV. Combined production of second-generation biofuels and electricity from sugarcane residues. Energy 2010;35(2):874–9. http://dx.doi. org/10.1016/j.energy.2009.07.032. Pereira LG, Dias MOS, MacLean HL, Bonomi A. Investigation of uncertainties associated with the production of n-butanol through ethanol catalysis in sugarcane biorefineries. Bioresour. Technol. 2015;190:242–50. http://dx.doi. org/10.1016/j.biortech.2015.04.095. Bonomi A, Mariano AP, Jesus CDF, Franco HCJ, Cunha MP, Dias MOS, Chagas MF, Cavalett O, Mantelatto PE, Maciel Filho R, Junqueira TL, Cardoso TF. The Virtual Sugarcane Biorefinery (VSB): 2011 Report. Campinas, São Paulo: Brazilian Bioethanol Science and Technology Laboratory (CTBE), Technological Assessment Program (PAT), Internal Report, 2012. Accessed 07.16.2014. Cavalett O, Junqueira TL, Dias MOS, Jesus CDF, Mantelatto PE, Cunha MP, et al. Environmental and economic assessment of sugarcane first generation biorefineries in Brazil. Clean Technol. Environ. 2012;14(3):399–410. http:// dx.doi.org/10.1007/s10098-011-0424-7. Moraes BS, Junqueira TL, Pavanello LG, Cavalett O, Mantelatto PE, Bonomi A, et al. Anaerobic digestion of vinasse from sugarcane biorefineries in Brazil from energy, environmental, and economic perspectives: Profit or expense? Appl. Energy 2014;113:825–35. http://dx.doi.org/10.1016/j. apenergy.2013.07.018. ANP, 2011. Brazilian Agency of Oil, Natural Gas and Biofuels. ANP Resolution #7, Feb 09, 2011. Roffler S, Blanch H, Wilke C. Extractive fermentation of acetone and butanol: process design and economic evaluation. Biotechnol. Prog. 1987;3:131–40. Afschar AS, Vaz Rossell CE, Schaller K. Bacterial conversion of molasses to acetone and butanol. Appl. Microbiol. Biotechnol. 1990;34:168–71. Yu EKC, Deschatelets L, Saddler JN. The bioconversion of wood hydrolyzates to butanol and butanediol. Biotechnol. Lett. 1984;6:327–32. Wayman M, Yu S. Acetone-butanol fermentation of xylose and sugar mixtures. Biotechnol. Lett. 1985;7:255–60. Marchal R, Ropars M, Vandecasteele JP. Conversion into acetone and butanol of lignocellulosic substrates pretreated by steam explosion. Biotechnol. Lett. 1986;8:365–70. Ezeji T, Qureshi N, Blaschek HP. Butanol production from agricultural residues: impact of degradation products on Clostridium beijerinckii growth and butanol fermentation. Biotechnol. Bioeng. 2007;97:1460–9. Ezeji T, Blaschek HP. Fermentation of dried distillers’ grains and soluble (DDGS) hydrolysates to solvents and value-added products by solventogenic clostridia. Bioresour. Technol. 2008;99:5232–42. Qureshi N, Saha BC, Hector RE, Dien B, Hughes S, Liu S, et al. Production of butanol (a biofuel) from agricultural residues: Part II – use of corn stover and switchgrass hydrolysates. Biomass Bioenergy 2010;34:566–71. Liu Z, Ying Y, Li F, Ma C, Xu P. Butanol production by Clostridium beijerinckii ATCC 55025 from wheat bran. J. Ind. Microbiol. Biotechnol. 2010;37:495–501. Chandel AK, Chandrasekhar G, Radhika K, Ravinder R, Ravindra P. Bioconversion of pentose sugars into ethanol: a review and future directions. Biotechnol. Mol. Biol. Rev. 2011;6:8–20. Dowson GRM, Haddow MF, Lee J, Wingad RL, Wass DF. Supporting information: catalytic conversion of ethanol into an advanced biofuel: unprecedented selectivity for n-Butanol. Angew. Chem. Int. Ed. Engl. 2013;52(34):9005–8. http://dx.doi.org/10.1002/anie.201303723. Accessed: 11.05.2014.

L.G. Pereira et al. / Applied Energy 160 (2015) 120–131 [60] CGEE, 2009. Center for Strategic Studies and Management in Science, Technology and Innovation. Bioethanol fuel: an opportunity for Brazil (in Portuguese). CGEE, Brasília, Brazil. [61] Dias MOS, Junqueira TL, Cavalett O, Cunha MP, Jesus CDF, Mantelatto PE, et al. Cogeneration in integrated first and second generation ethanol from sugarcane. Chem. Eng. Res. Des. 2013;91(8):1411–7. http://dx.doi.org/ 10.1016/j.cherd.2013.05.009. [62] Dutta A, Phillips SD. Thermochemical ethanol via direct gasification and mixed alcohol synthesis of lignocellulosic biomass. Report NREL/TP-510-45193, 2009. [63] Humbird D, Davis R, Tao L, Kinchin C, Hsu D, Aden A, Schoen P, Lukas J, Olthof B, Worley M, Sexton D, Dudgeon D. Process design and economics for biochemical conversion of lignocellulosic biomass to ethanol – dilute-acid pretreatment and enzymatic hydrolysis of corn stover. Technical Report NREL/ TP-5100-47764, Golden, Colorado, USA, NREL, 2011. 136 pp. Accessed: 03.06.2014. [64] UDOP, 2014.Union of Biofuel Producers.Technical report on Sugarcane prices 2014, Aracatuba, Sao Paulo, Brazil. Accessed 07.06.2014. [65] Cardoso T, Cavalett O, Chagas MF, Morais ER, Carvalho JLN, Franco HCJ, et al. Technical and economic assessment of trash recovery in the sugarcane bioenergy production system. Sci. Agric. 2013;70:353–60. [66] Hong Y, Nizami A-S, Bafrani MP, Saville BA, MacLean HL. Impact of cellulase production on environmental and financial metrics for lignocellulosic ethanol. Biofuels Biprod. Bioref. 2013;7:303–13. http://dx.doi.org/10.1002/bbb.1393. [67] CEPEA, 2014. Center of Advanced Studies in Applied Economics (ESALQ-USP). Anhydrous ethanol prices (monthly indicator) in the State of São Paulo. Accessed: 07.11.2014. [68] MME, 2014. Ministry of Mines and Energy. Renewable electric energy auctions (in Portuguese). Accessed: 06.28.2014. [69] Orbichem, 2013a. TecnonOrbichem. Prices on chemicals – n-Butanol. Accessed: 10.27.2014. [70] Orbichem, 2013b. TecnonOrbichem. Prices on chemicals – 2-Ethylhexanol. Accessed: 10.27.2014. [71] CNCIC, 2014. China National Chemical Information Center. Octanol market study. Accessed: 10.25.2014. [72] Orbichem, 2013c. TecnonOrbichem. Prices on chemicals – Acetone. Accessed: 10.27.2014. [73] Power, J., 2011. Furfural project review, 2011. Accessed: 11.17.2014. [74] UNICA, 2014.Data and Quotations Statistics 2014 (In Portuguese).Union of the Sugarcane Industry. Accessed: 02.02.15.

131

[75] IDEA, 2012. Institute of Agroindustrial Development. Indicators of the Sugar Agroindustry Performance: Seasons 2010/11 and 2011/12 (in Portuguese). IDEA Group, Ribeirao Preto, p. 98. [76] ANA, 2009. Conservation and Reuse of Water in the Sugar-energy Agroindustry (In Portuguese). Brazilian Water Agency. Accessed: 02.09.2015. [77] ISO 14040, 2006. Environmental management – Life cycle assessment – Principles and framework, International Standard Organization, 20 pp. [78] ISO 14044, 2006. Environmental management – Life cycle assessment – Requirements and guidelines, International Standard Organization, 46 pp. [79] Guinée JB, Heijungs R, Huppes G. Economic allocation - examples and derived decision tree. Int. J. Life Cycle Assess. 2004;9(1):23–33. [80] Cherubini F, Strømman AH, Ulgiati A. Influence of allocation methods on the environmental performance of biorefinery Products – a case study. Resour. Conserv. Recy. 2011;55:1070–7. [81] Luo L, van der Voet E, Huppes G. Life cycle assessment and life cycle costing of bioethanol from sugarcane in Brazil. Renew. Sust. Energy Rev. 2009;13(6– 7):1613–9. [82] Wardenaar T, van Ruijven T, Beltran AM, Vad K, Guinée J, Heijungs R. Differences between LCA for analysis and LCA for policy: a case study on the consequences of allocation choices in bio-energy policies. Int. J. Life Cycle Assess. 2012;17:1059–67. [83] Wang M, Huo H, Arora S. Methods of dealing with co-products of biofuels in life cycle analysis and consequent results within the US context. Energy Policy 2011;39(10):5726–36. [84] EPE, 2014. National Energy Balance, Year 2013. Final Report, 285p. Empresa de Pesquisa Energética. Accessed: 11.13.2014. [85] Chagas MF, Cavalett O, Silva CRU, Seabra JEA, Bonomi A. Adaptação de Inventários de Ciclo de Vida da cadeia produtiva do etanol de cana-de-açúcar no Brasil. In: III Congresso Brasileiro em Gestão do Ciclo de Vida de Produtos e Serviços: Novos desafios para um planeta sustentável, 2012, Maringá. Anais do III Congresso Brasileiro em Gestão do Ciclo de Vida de Produtos e Serviços, DentalPress, Maringá, Brazil, 2012. p. 1-5. [86] MMA, 2011. First National Inventory of Atmospheric Emissions by Vehicles. Final Report (in Portuguese). Brazilian Ministry of Environment. Accessed 28.07.13. [87] ANP, 2012. Brazilian Agency of Oil, Natural Gas and Biofuels. ANP Resolution #26, Aug 31, 2012. [88] EPA, 2005.US Environmental Protection Agency. Renewable Fuels: Regulations & Standards. Accessed: 01.26.2015.