Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement

Energy Conversion and Management xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement Simone P. Souza a,b, Joaquim E.A. Seabra a,⇑ a

Faculdade de Engenharia Mecânica, UNICAMP. Rua Mendeleyev 200, Cidade Universitária ‘‘Zeferino Vaz’’, Campinas, SP Postal Code 13083-860, Brazil Brazilian Bioethanol Science and Technology Laboratory (CTBE) – CNPEM/ABTLuS – Rua Giuseppe Máximo Scolfaro 10.000, Polo II de Alta Tecnologia, P.O. Box 6170, Campinas, SP Postal Code 13083-970, Brazil b

a r t i c l e

i n f o

Article history: Available online xxxx Keywords: Biofuels Biorefinery LCA Sustainability Uncertainty analysis

a b s t r a c t The sugarcane industry in Brazil has been considered promising for the production of advanced fuels and bio-based products. However, the sugarcane crop requires high volumes of fossil fuel for cultivation and transport. The use of biodiesel as a diesel substitute could reduce the environmental burdens associated with this high consumption. This work performed a stochastic evaluation of the environmental and economic implications of the integrated production of sugarcane bioethanol and soybean biodiesel, in comparison with the traditional sugarcane-to-ethanol process. The analysis was focused on the states of Goiás, Mato Grosso and São Paulo, where this integration would be particularly attractive. The environmental aspects addressed were the fossil energy use and the GHG emissions in a cradle-to-gate approach. The economic analysis comprised the evaluation of the net present value of an incremental cash flow generated by the soybean production and by the adjacent plants of oil extraction and biodiesel. Results indicate that the integrated system is likely to improve the ethanol environmental performance, especially with regard to the fossil energy use. The integration is economically feasible but highly uncertain; however, it could be significantly improved through fiscal incentives to biodiesel producers, founded on the reduction of fossil energy use and on improvements in logistics. In addition, the proposed model may also assist in the design of other integrated systems applied to the sugarcane sector in Brazil. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Integrated complexes, also called biorefineries, have been proposed as an alternative to improve the interaction of bioenergy, chemicals and food production by applying sustainable processing of biomass [1]. Biorefineries are able to produce a range of products from different raw materials and, additionally, may reduce the commitment of land for bioenergy production and provide diversification and optimization of agricultural systems [2]. Initial studies concerning the integrated production of food and bioenergy have demonstrated that this interaction can also provide the rural development due to the high renewability and energy sustainability [3], and afford better energy and environmental performance [4]. Due to the diversity of products (sugar, bioethanol, bioelectricity, etc.) and the possible applications for the residues, the sucroenergetic sector in Brazil is already an important model of biorefinery. However, there is still potential for improvement and different systems can be applied to the sector, such as integrated ⇑ Corresponding author. Tel.: +55 19 3521 3284. E-mail address: [email protected] (J.E.A. Seabra).

production of bioenergy and food [4], ethanol, methane and heat [5], ethanol and biodiesel [2], among others. Biorefinery models can also add value to the sector [6]. Despite the potential to produce many bioproducts, the sugarcane sector is characterized by the high diesel consumption, which is the main non-renewable energy input and one of the main sources of greenhouse gas (GHG) emissions in the ethanol life cycle. Further, considering an average specific consumption of 4 L/t sugarcane [7], it can be estimated that the sugarcane sector alone was responsible for about 4% of the total diesel consumption in Brazil in 2011. Brazil is also known by its biodiesel program. Due to the soybean relevance in the Brazilian agribusiness as a result of a suitable development of agronomic, industrial and logistic aspects over decades, the soybean production chain met the Brazilian Biodiesel Program (PNPB) demand. Soybean has been leading as the main feedstock for biodiesel production, comprising about 80% of the biodiesel sources. Other feedstocks include beef tallow, cottonseed, waste frying oil, swine and chicken fat, palm oil, peanut and sunflower [8]. The high demand for diesel in sugarcane production and the relevance of soybean as feedstock for biodiesel production in Brazil

http://dx.doi.org/10.1016/j.enconman.2014.06.015 0196-8904/Ó 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

2

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

have encouraged efforts toward the integration of these bioenergy systems. Such system has already been tested in Brazil, including the joint production of biodiesel, ethanol, sugar and electricity [9]. In this study, we investigated an integrated system in which a soybean biodiesel plant is integrated into the sugarcane sector, assuming that the replacement of fossil diesel by the locally produced renewable fuel would improve the environmental and economic performance of the traditional sugarcane biorefinery. The environmental aspects addressed in this study were the fossil energy use and the GHG emissions. The traditional ethanol production was used as the reference case for comparisons. The environmental and economic implications of the integrated system were assessed though a stochastic approach based on the Monte Carlo method.

 





 2. Integrated system In the proposed model, sugarcane and soybean oil are processed in a combined ethanol–biodiesel plant, which uses only bagasse as fuel. The distillery provides the utilities for the biodiesel plant. We assumed that oil is provided by the soybean grown in the sugarcane reforming areas, including the direct oil from the grain and the additional oil acquired from the sale of soybean meal (defined as Façon exchange [9]) (Fig. 1). Based on the average conditions of the Brazilian south-central region (which mainly reflect the conditions of the São Paulo state), our previous study [10] showed that this integration would be able to considerably reduce the fossil energy use in the ethanol life cycle, but with minor implications for the GHG emissions. However, this integration would be particularly attractive under the conditions featured in the Midwest region (e.g., land availability, background in soybean cultivation, flat terrain and large scales). Therefore, in the present study we focused the analysis on the states of Goiás and Mato Grosso (Fig. 2), using representative data collected through site visits. Special attention was paid to the uncertainties of the model, which were evaluated through sensitivity analyses and the Monte Carlo (MC) simulation. Additionally, an economic analysis was performed to indicate the viability of this integration model. The integration described in this study comprises the following assumptions:    

Soybean is produced in the sugarcane renovation area. The biodiesel plant includes the oil extraction unit. The soy meal is traded for additional oil. The soybean oil converted into biodiesel is provided by the soybean cultivated in the sugarcane renovation area and by the

additional oil acquired from the soy meal exchange. No additional area was assumed. The soybean oil transesterification employs sugarcane ethanol (ethyl route). Diesel used for sugarcane production is partially replaced by biodiesel without engine modification [11]. No biodiesel surplus is produced. Vinasse (a liquid effluent from the distillation process) and filtercake (a coproduct from sugarcane juice filtration) are used for fertirrigation and organic fertilization, respectively. Filtercake is applied in 100% of the renovation area. Sugarcane bagasse is completely used for heat and power generation, meeting the energy requirements of the combined plant (ethanol and biodiesel) and producing electricity surplus. Fifty percent of the sugarcane area is irrigated (salvage irrigation).

Except for the states in the Northeast region, sugarcane is practically not irrigated in Brazil. When employed, irrigation is mostly used after the sugarcane planting to ensure sprouting (‘‘salvage irrigation’’). In the states of Mato Grosso and Goiás, sugarcane irrigation is still incipient, and the procedures and application rates vary within a wide range among the producers. 3. Methods 3.1. Life cycle assessment A comparative analysis was performed between a traditional ethanol production system and a sugarcane–soybean integrated system applying the life cycle assessment (LCA) technique according to ISO 14040:2006 [12] and ISO 14044:2006 [13]. The functional unit was defined as 1 MJ of ethanol. The fossil energy use and the GHG emissions were evaluated in a cradle-to-gate approach, comprising the sugarcane cultivation up to ethanol processing. Fossil energy use is expressed by the ratio between the fossil energy invested to produce the biofuel, and the bioenergy produced. The results for life cycle GHG emissions are expressed in CO2e using the GWP100 given by IPCC as characterization factors [14]. The foreground information (Appendix A, Tables A1 and A2) was based on data collected from sugarcane mills, sugarcane and soybean suppliers, and biodiesel plants. The data were collected in the states of São Paulo, Mato Grosso and Goiás (Fig. 2), employing direct field observation methods through questionnaires. The field data were compared to the literature information, which supported the statistical parameters used in the sensitivity and

Soy meal

Soy oil

System boundary Façon exchange Soybean Sugarcane field

Renovation area

Sugarcane cultivation

Oil extraction plant

Soybean cultivation

Ethanol Sugarcane

Machinery Trucks

Combined plant

Electricity Glycerin

Biodiesel

Fig. 1. Sugarcane–soybean integration design considered in this study.

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

3

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

Fig. 2. Spatial scope of this study.

uncertainty analyses. The diesel consumption was validated through a sugarcane mill’s database with more than 100,000 data for all sugarcane cultivation activities. As for nitrogen fertilizers, the contribution of each source was defined according to Seabra et al. [7]. The product system was modelled so that inputs and outputs at the boundary were elementary and product flows. The transportation of inputs (fertilizers, chemicals) to the processing units, however, was ignored because its contribution to the overall life cycle is minimal and the sources are variable. All energy flows were calculated in terms of primary fossil energy. For the GHG emissions, fluxes related to diesel use and production, biomass combustion, production of chemical inputs, production and use of fertilizers and limestone, pesticides, and residues that are returned to the soil (cane straw, filtercake, vinasse, soot and ash) were quantified. Direct and indirect N2O emissions from crop residues and synthetic nitrogen were estimated according to the IPCC Tier 1 method [15]. Direct N2O emissions as a result of biological nitrogen fixation were not considered due to the lack of evidence indicating significant emissions arising from this process [15]. The life cycle fossil energy use and GHG emissions of material and energy inputs were retrieved from Ecoinvent v.2.2. and from the CanaSoft Model, developed by the Brazilian Bioethanol Science and Technology Laboratory (CTBE) team, which were modelled for the Brazilian conditions. Regarding the treatment of co-products, the ISO 14044 standard suggests that when allocation cannot be avoided, as the case of this work, inputs and outputs of a system should be partitioned between its different functions or products in a way that reflects the underlying physical relationships between them [13]. For this reason, an energy-based allocation was applied to split the fossil energy use and the GHG emissions between ethanol and electricity surplus, as both ethanol and electricity are energy products, and energy was found to be the most appropriate physical relationship to be considered. As for biodiesel and glycerin, a mass-based

Table 1 Parameters of the economic analysis.a

a b c d e f g

Plant life (years) Depreciation – equipment (% per year) Depreciation – buildings (% per year) Income tax (%) Operational cost (US$/t biodiesel)b Maintenance (US$/t biodiesel)b Labor (US$/t biodiesel)b Discount rate (%)

25 10 4 34 360 18 2.5 12

Prices Soybean oil (US$/t) – ICMS includedc Soybean meal (US$/t)c Glycerin (US$/t)d Low sulfur diesel 10 ppm (US$/l)e Regular diesel (US$/l)e Soybean cost production (US$/t)f

1295 289 426 1.45 1.37 403

Taxes (ICMS) Glycerin (%)g Soybean meal (%)g Soybean oil (%)g

12 4.2 7

Values of 2011. [17]. [18]. [19]. [20]. [21]. [22].

method was deemed to be more appropriate. No environmental burdens were attributed to the soy meal, as well as no burdens were attributed to the additional oil acquired through the soy meal trade. 3.2. Economic analysis The profitability of the investment was evaluated using the Net Present Value (NPV) method. As this work involves a comparative

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

4

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

120

Mean = 116

Mean = 96

100

Frequency

80 60 40 20 0 50

60

70

80

90

100

Traditional

110

120

130

140

Integrated System

kJfossil/MJfuel Fig. 3. Monte Carlo results for fossil energy use – comparison between the traditional system and the sugarcane–soybean integrated system.

kJfossil/MJfuel 70

75

80

Cane transportation distance - km

85

90

95

100

105

26

Sugarcane diesel consumption - L/ha.y

234

Area for soybean - % of the total available area

115

120

238 51

100

Soybean yield - t/ha

2.1

3.2

Fuel consumption for oil extraction - MJ fossil/MJ fuel

0.09

Additional oil from Façon - t oil/t meal

1.2 0.19

0.26

Plant-cane N application rate - kg/ha.y

0

Soybean N application rate - kg/ha

0.3

Soybean transportation distance - km

76.7

Upside

110 103

5.25 7.3 123.3

Downside

Fig. 4. Sensitivity analysis for fossil energy use (base case) – integrated system. The centerline represents the baseline value.

45 40

Mean = 23.9

Mean = 23.4

Frequency

35 30 25 20 15 10 5 0 14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

g CO2e/MJfuel

Traditional

Integrated System

Fig. 5. Monte Carlo results for GHG emissions – comparison between the traditional system and the sugarcane–soybean integrated system.

analysis between mutually exclusive options (the traditional system and the alternative integrated system), an incremental cash flow was assembled considering only the incremental costs and revenues. Therefore, the capital expenditures include only the oil extraction plant and the biodiesel plant, while the operational expenditures include maintenance, labor and consumables (chemical inputs) related to the soybean and biodiesel production. In terms of revenues, only those related to the fossil diesel replacement and the sale of glycerin were considered. Once the traditional system does not include the soybean production, the costs to produce this grain was treated as expenditure

in the integrated system. The total value of soy meal is exchanged by oil. We included the taxes such as ICMS (Brazilian tax for goods and services) and PIS/COFINS (Brazilian tax for financing social security). The system boundary comprises the biodiesel produced from the soybean grown in the sugarcane renovation area and from the additional oil provided by the Façon exchange. To enable the gains of scale, it was assumed that a 100,000 toil/year biodiesel plant (88% utilization factor) would be adjacently installed to the ethanol distillery. This is a common size for the new biodiesel plants in Brazil [16]. However, the biodiesel comprised within

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

5

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

g CO2e/MJ fuel 10

13

Direct N2O emission from managed soil * Ind. N2O emission, managed soil - Leaching EF * Ind. N2O emission, managed soil - Leaching Fraction * Cane transportation distance - km Sugarcane diesel consumption - L/ha.y Ind. N2O emission, managed soil - Volatilization EF * Ind. N2O emission, managed soil - Volatilization Fraction * Area for soybean - % of the total available area Soybean yield - t/ha Plant-cane N application rate - kg/ha.y Fuel consumption for soy oil extraction - kg CO2e/MJ fuel Additional oil from Façon - t oil/t meal Soybean N application rate - kg/ha Soybean transportation distance - km

15

18

20

23

0.004

25

28

30

33

0.03 0.002 0.14 26 234 0 0.04 100 3.2 0 8.310 0.26 0.3 76.7

Upside

0.023 0.74 103 238 0.05 0.28 51 2.1 5.25 68.190 0.19 7.3 123.3

Downside

* See units in Table A3. Fig. 6. Sensitivity analysis for GHG emissions (base case) – sugarcane–soybean integrated system. The centerline represents the baseline value.

120

kJ/MJfuel

100 80 60 40 20 0 Traditional

Integrated System

Fertilizers

Pesticides

Diesel consumption

Biodiesel consumption

company [17]. Current prices were first converted into 2011 values using Brazilian price indexes (IPCA – Consumer Price Index; IGP-DI – General Price Index) and CEPCI Index (Chemical Engineering Plant Cost Index). The 2011 values were then converted into US dollar. The working capital was estimated assuming that the biodiesel plant must have cash (revenue subtracted by the costs) during 35 days, period in which the plant is not working. The objective of the integrated system is to replace the fossil diesel consumption, and the economic benefits were assessed considering the replacement of the Brazilian regular diesel. However, since January 2013 a new regulation on low sulfur diesel (10 ppm) is in place, which is currently focused on trucks and buses. But given the aims of the government in extending its use to agricultural machinery, we have also considered the displacement of low sulfur diesel as a sensitivity scenario.

Chemicals and Lubrifants

3.3. Uncertainty analysis Fig. 7. Life cycle fossil energy use of the traditional system and the sugarcane– soybean integrated system.

25

gCO2e/MJfuel

20 15 10 5 0 Traditional

Integrated System

Fertilizers

Pesticides

Diesel consumption

Biodiesel consumption

Chemicals and Lubricants

Residues

Fig. 8. Ethanol life cycle GHG emissions of the traditional system and the sugarcane–soybean integrated system.

the system boundary of the integrated system represent only 6.15% of the plant output, which was the share of the fixed costs attributed to the integrated system. The capital and operational costs of the biodiesel plant (Table 1) were estimated based on data reported by a Brazilian capital goods

Given the uncertainties associated with the integrated system, the environmental and economic performances were assessed through a stochastic approach employing the Monte Carlo simulation. In this method, a proper probability distribution (in the present case, based on the visited mills and farms, historical data, expert opinion and literature) is associated to each of the input variables subjected to uncertainties. Values for these variables are generated randomly and combined with other randomly-generated values. The results are presented as an average value associated with a probability distribution for all possible outputs [23]. When necessary, we also applied Student’s t-test to verify if two sets of data were significantly different from each other. Crystal BallÒ was used as auxiliary tool to define the best-fit distribution, the sensitivity analysis and the MC simulation. The probability distributions and assumptions adopted in the Monte Carlo simulation are presented in Table A3. For the additional oil provided by the Façon exchange, the range of values was estimated based on the ratio between the soy meal and oil prices verified in the Mato Grosso state [18]. With regard to nitrogen application, although plant-cane do not necessarily require N application after the crop rotation, during the site visits it was identified that some farms apply the same amount of N used in areas without rotation. In this case, it was assumed a 50/50% probability for each situation (N application and no-N application). For oil extraction, heat is provided by a weighted mean of fuel oil, firewood and natural gas, in accordance with the energy consumption by source in the food and beverage sector

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

6

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

40 35 Mean = 2,210

30

Frequency

25 1 Std Dev = + 8,990

-1 Std Dev = - 4,570

20 15 10 5 0 -15,780

-9,860

-3,940

1,980

7,900

13,830

19,750

1,000 USD Fig. 9. Net present value of the incremental cash flow.

1000 USD -20,000 Soybean oil price - US$/t Soybean yield - t/ha Regular diesel price - R$/l Soybean production cost - R$/ha Soybean meal price - US$/t Additional oil from Façon - t oil/t meal Area for soybean - % of the total available area Oil extraction plant - Equipments - 1000 US$ Glycerin price - R$/t Biodiesel plant - Equipments - 1000 US$ Oil extraction plant - Buildings - 1000 US$ Biodiesel plant - Buildings - 1000 US$

-10,000

0

10,000

1,940 2.07 2.30 1,750 230 0.19 50 2,650 610 1,240 520 130 Upside

20,000 720

3.22 2.85 1,340 370 0.27 100 1,650 990 770 325 80

Downside

Fig. 10. Sensitivity analysis for the NPV (base case) – integrated system.

in Brazil [24]. Due to uncertainties regarding the N2O emission from managed soil, we assumed probability distributions for direct and indirect (volatilization and leaching/runoff) emissions according to the IPCC ranges [15]. 4. Results 4.1. Environmental assessment The biodiesel produced in the sugarcane–soybean integrated system is able to replace 38% of the diesel consumed in the sugarcane ethanol life cycle. The mean values obtained in the Monte Carlo analysis indicate that the integrated system configuration is able to improve the ethanol life cycle performance by more than 17% when compared to the traditional system (Fig. 3). The sensitivity analysis shows that the most impacting parameters are the sugarcane transportation distance, the diesel consumption in the sugarcane cultivation and the available area for soybean, responsible for 57%, 20% and 7% of the variance in the MC analysis, respectively. For an individual variable test, the sensitivity analysis is presented in Fig. 4. As for the GHG emissions, the Monte Carlo analysis indicates that the sugarcane–soybean integrated system can reduce the ethanol life cycle emissions from 23.4 to 22.9 g CO2e/MJ when compared to the traditional system (Fig. 5). These estimates are relatively uncertain, but the Student’s t-test supports that the difference between the production systems are statistically significant at the 5% significance level. The sensitivity analysis

shows that the emissions of N2O from the managed soil, which include direct and indirect emissions, are responsible for 95% of the variance in the MC analysis. The sensitivity analysis for an individual variable test is showed in Fig. 6. The significance of the N2O emission from the managed soil in the GHG life cycle shows the importance of studies regarding the appropriate emission factors for different conditions, such as soil, climate and crops, the proper nitrogen application rate and management of residues. The response to the diesel replacement is higher for the fossil energy use than for the GHG emissions because diesel accounts for 61% of the fossil energy consumption over the traditional ethanol life cycle (Figs. 7 and 8), but only 23% of the life cycle emissions. Over 30% of the life cycle burden associated with GHG emissions comes from the residues, including straw (tops and leaves) decomposition, vinasse, filtercake, boiler ash, soot and bagasse burning. Nitrogen use, which includes its production and use (direct and indirect emissions from the soil), is responsible for 27% of the total emissions in the traditional system, followed by limestone, which represents 12% (Appendix A, Tables A4 and A5). 4.2. Economic assessment Considering the diesel displacement as the main revenue, the investment in the sugarcane–soybean integrated system is economically feasible. The inclusion of the oil extraction unit can be essential for the economic feasibility of biodiesel production in the current scenario of prices and taxes. According to the Monte

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

7

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

Carlo results, the NPV of the integrated system is 2 ± 7 MUS$ (mean ± std. dev.) (Fig. 9). Soybean oil price, soybean yield and diesel price are the most influential parameters for the NPV variability (Fig. 10). Considering a scenario where biodiesel replaces low sulfur diesel (10 ppm), the NPV according to MC simulation is 5 ± 7 MUS$. It must be noted that NPV is not affected by the biodiesel selling price since no surpluses were considered. Furthermore, sales of ethanol and electricity were not included as revenues because of the incremental cash flow approach. 5. Discussion Other works have studied similar approaches to this paper. A study reported that a switchgrass biorefinery system producing Table A1 Sugarcane and soybean production parameters.a.

a

Parameter

Value

Units

Sugarcane Number of cutsb Seedlings for mechanized plantingc Sugarcane yieldd Straw yielde Straw left in the fieldf

5 18 85 140 80%

t ha 1 t ha 1 year 1 kgdry t 1 cane

Agrochemicals input Nitrogen Phosphorus Potassium Insecticide Fungicide Herbicide Limestone Gypsum (CaSO4) Diesel consumption Vinasseg Boiler ashh Soot Filtercakei

84.5 27 73 0.2 <0.1 4 1,000 500 260 140 170 1020 5

kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 L ha 1 year 1 m3 ha 1 year 1 kgdry ha 1 year kgdry ha 1 year tdry ha 1 year 1

Soybean Seed consumptionj Soybean yieldk Yield loss by soil compactionl Effective soybean yield Oil yieldm Oil from Façonn Soy meal yieldo Soybean straw productiono

51 3 23% 2.3 5600 178 0.8 2810

kg ha 1 t ha 1 year

Agrochemicals input Nitrogeno Phosphoruso Potassiumo Insecticidej Fungicidej Herbicidej Diesel consumptionp

1.2 72 72 1 0.7 1.7 90

kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 kg ha 1 year 1 L ha 1 year 1

Table A2 Sugarcane and soybean processing parameters. Parameter

1

1 1

1

1

t ha year t year 1 kg t 1 grain kg kg 1 grain kgdry/ha

Data reported by visited mills and farms. Each sugarcane cycle has 5 cuts (semi-perennial), which is proceeded by a rotation crop (soybean in this study). The greater the number of cuts, the less the available area for soybean. c Planted area. d Harvested area. e [36]. f [7]. g Applied in 57% of the total area. h Applied in 3% of the area. i Applied in 18% of the area. j [37]. k [32]. l Soybean grown after sugarcane. m Including soybean produced in the renovation area and from Façon exchange. n Calculated. o Data provided by visited mills and farms. p Calculated using data provided by the plants. b

bioenergy, bioethanol and chemicals can save 79% of GHG emission and 80% of fossil energy, compared to the fossil reference system [25]. An assessment about the use of crop residues to produce bioenergy, bioethanol and chemicals reported the potential to save 53.7% of GHG emissions and 80% of fossil energy using corn stover, and 49% of GHG emissions and 80% of fossil energy with wheat straw, both compared to the fossil reference systems [26]. Integrating algae bio-based products to the sugarcane sector can reduce 50% the fossil energy use and 18% the GHG emissions in relation to the traditional ethanol production in Brazil [27]. The integration of palm oil biodiesel production to the sugarcane sector in Brazil can reduce 45% of the fossil energy use and 23% of the GHG emissions [2]. Another study reported that a corn–soybean integrated system, which produces ethanol and biodiesel, has a lower global warming impact and lower nonrenewable energy use when compared to continuous corn system [28]. Although all these studies present a similar scope, it is not possible to directly compare them to our results because of the differences in the system boundaries and assumptions. The production of soybean meal and the use of ethyl route are important aspects of the proposed integrated system. Around

Sugarcane Mill processing capacity Outputsa Ethanol Electricity surplusb Vinassec Filtercakec,d Consumablesc CaO Antifoam Dispersant Sulfuric acid Lubricantse Cyclohexane Sodium hydroxide Soybean Grain dryingf Electricity consumption Firewood consumption Oil extraction Electricity consumption Fuel consumption Biodiesel plant Biodiesel productiong Plant processing capacity Production/capacityg Oil–biodiesel conversionh Glycerin yieldg Electricity consumption Consumablesi Ethanol Sodium methylate Citric acid Hydrochloric acid Sodium hydroxide Sulfuric acid

Value

Units

4,000,000

t year

1

cane

1

85 104.6 11 32

L t cane kW h t 1 cane L L 1 ethanol kg t 1 cane

530 0.3 0.3 6.6 13 1.2 0.4

gt gL gL gL gt gL gL

10 6.5

kW h/t grain (wet basis) kg/t grain (wet basis)

69 2046

kW h/t grain (wet basis) M J/t grain (wet basis)

5,500 100,000 88% 97.5% 138 40.44

t year t year

154 33.4 0.65 9.5 1.5 0.2

kg/t biodiesel kg/t biodiesel kg/t biodiesel kg/t biodiesel kg/t biodiesel kg/t biodiesel

1

cane ethanol ethanol 1 ethanol 1 cane 1 ethanol 1 ethanol 1 1

1 1

of oil

kg/t biodiesel kW h/t biodiesel

a Vinasse and filtercake are industrial residues that are returned to the field as fertilizers. b Total electricity generation: 135.5 kW h/t cane [38]; electricity consumption: 30 kW h/t cane (ethanol plant) + 40.44 kW h/t biodiesel (biodiesel plant) (0.9 kW h/ t cane). c Data provided by visited mills. d 67% moisture content. e [39]. f [40]. g Data provided by visited plants. h [41]. i [9].

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

8

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

Table A3 Parameters of the Monte Carlo analysis. Parameter Environmental analysis Area for soybean cultivationa Additional oil from Façon Soybean transportation distance Soybean yield Soybean N application rate Plant-cane N application rate Oil extraction – Fuel consumptiond Emissions Fossil energy use Cane transportation distance Cane diesel consumption N2O emission from managed soils Direct Indirect Volatilization EFe Volatilization Fractionf Leaching EF Leaching Fraction Economic analysis Area for soybean cultivationa Additional oil from Façon Soybean yield Investment Oil extraction plant Buildings Equipment Biodiesel plant Buildings Equipment Diesel price Low sulfur diesel 10 ppm Regular diesel Soybean production cost Soybean meal price Soybean oil price Discount rate Glycerin price a b c d e f g h

Distribution

Range

Mean/SD

Unit

Uniform Normal Normal Normal Triangular P50/P50c

50–100 0.19–0.27 – 2.07–3.22 0–8 0/5.25

– 0.22/0.02 100/10 2.3/0.23 1.2b –

% t oil t 1 meal km t ha 1 kg ha 1 kg ha 1 year

Triangular Triangular Triangular Triangular

4.64–74 0.027–1.32 20–110 234–338

24b 0.36b 60b 262.5b

1 kg CO2 MJ fuel 1 MJ MJfuel km L ha 1 year 1

Triangular

0.003–0.03

0.01b

kg N2O–N (kg N applied)

Triangular Triangular Triangular Triangular

0.002–0.05 0.03–0.3 0.0005–0.025 0.1–0.8

0.01b 0.10b 0.0075b 0.3b

kg N2O–N (kg NH3–N + NOx–N volatilized) 1 kg NH3–N + NOx–N volatilized (kg N applied) kg N2O–N (kg N leaching) 1 kg N (kg N additions) 1

Uniform Normal Normal

50–100 0.19–0.27 2.07–3.22

– 0.22/0.02 2.3/0.23

% kg oil t t ha 1

Normal Normal

– –

423.15/42.3 2153.9/215.4

1000 USD 1000 USD

[17] [17]

Normal Normal

– –

104.1/10.41 1006.3/100.63

1000 USD 1000 USD

[g] [g]

Logistic Logistic Beta Lognormal Lognormal Triangular Normal

– – 1325.6–1756.7 – – 10–15 –

2.72/0.06 2.57/0.06 a = 1.38, b = 1.27 288.56/28.56 1210.5/260.5 12 800/80

R$/L R$/L R$/ha US$/t US$/t % R$ t 1

[20] [20] [21] [18] [18] [g] [19]

1

[g] [18] [h] [42] [h] [h]

1

meal

[24] [24] [h] [h] 1

[15]

1

[15] [15] [15] [15] [g] [18] [42]

Renovation area used for crop rotation. Likeliest value. 50% Probability for each situation: application and non-application. Lower values correspond to firewood. Emission factor for N2O from atmospheric deposition of N on soils. Fraction of synthetic N fertilizer that volatilizes as NH3 and NOx. Assumption. Field data.

80% of the biodiesel factories have been employing methanol in the transesterification process, which has progressively increased the Brazilian imports of methanol [29] – the country spent 250 million dollars importing methanol in 2012 [30]. The preference for the methyl route has been supported by its physical and chemical properties and by the lower production costs imposed to biodiesel production [31]. Ethanol, on the other hand, is renewable, less toxic, and widely available in the Brazilian domestic market. With regard to soybean meal, Brazil exported 13.6 million tonnes of soybean meal in 2012 and it has been widely used as animal feed [30]. The resulting glycerin from biodiesel production has multiple applications such as cosmetics, food, chemicals, fuel, and it can also generate heat and power. Thus, increasing its production must be seen as an opportunity, especially to develop new markets. Furthermore, it should be noted the current market, the logistics and the policies applied to fuels in Brazil. Some factories are already able to produce biodiesel at costs lower than the diesel price; however, the biodiesel transport cost can make it infeasible. Regardless of the region, the Brazilian federal law n. 11097/2005 requires a blend of 5% biodiesel and 95% diesel by volume (B5), implying long-distance transportation between the factories and the final consumer, which is performed by trucks fueled with

diesel. Mato Grosso and Goiás are among the largest soybean producers in Brazil, leading the biodiesel production along with the Rio Grande do Sul state. The distance between these states and the largest biodiesel consumers and refineries are around 1000 km, while diesel has to travel in the opposite direction from the refineries located near the coast to the states of Mato Grosso and Goiás. Provided that all these exchanges are heavily based on road transportation, incentives for local consumption of biodiesel could help in the mitigation of the logistical issues in the country. In addition to the environmental benefits and the opportunities to improve logistics, incentives to the integrated production of biofuels would also reduce diesel imports. In 2012, Brazil imported 8.1 billion liters of diesel, equivalent to 6.6 billion dollars [30]. For comparison purposes, according to our life cycle inventory, the diesel required for farming and transportation of 1 tonne of sugarcane is around 4 liters, which leads to an overall consumption of 2 billion liters by the sugarcane sector in 2012, with an upward trend due to the harvesting mechanization. This is equivalent to 4% of the national consumption considering the Brazilian Energy Balance [24] and the estimate of sugarcane production [32]. Although the current conditions in Brazil favors soybean, many other feedstocks can be used in the integrated production of

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

9

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx Table A4 GHG emissions of the assessed systems [gCO2e (MJf)

1

]. Traditional system

Integrated system

0.1

0.1

2.3 2.8 0.2 0.2 0.1 2.4 <0.1

2.3 2.8 0.2 0.2 0.1 2.4 <0.1

0.2 <0.1 <0.1

0.2 <0.1 <0.1

4.4

2.7 (Diesel) 1 (Biodiesel)

Residue emission Straw decomposition Vinasse Filtercake Boiler Ashes Soot Bagasse burning

1.75 1.55 0.7 <0.1 0.7 1.6

1.75 1.55 0.7 <0.1 0.7 1.6

Industrial phase Chemicals and lubricants TOTAL

0.2 19.2

0.2 18.6

Traditional system

Integrated system

Agricultural phase Seeds

0.6

0.5

Fertilizer Nitrogen Phosphorus Potassium Limestone (CaCO3) Gypsum (CaSO4)

27 2.45 2.4 1.3 <0.1

27 2.45 2.4 1.3 <0.1

3.4 0.3 <0.1

3.4 0.3 <0.1

67.85

41.85 7.95

5.9 111

5.9 93

Agricultural phase Seeds Fertilizer Nitrogen (production) Nitrogen (direct and indirect emission) Phosphorus Potassium Limestone (production) Limestone (use) Gypsum (CaSO4) Pesticides Herbicides Insecticides Fungicides Fuel consumption Cropping practices, harvesting, infield wagon, cane transportation, vinasse disposal, irrigation

Table A5 Fossil energy inputs of the assessed systems [kJ (MJf)

1

].

Pesticides Herbicides Insecticides Fungicides Fuel consumption Cropping practices, harvesting, infield wagon, cane transportation, vinasse disposal, irrigation Industrial phase Chemicals and lubricants TOTAL

biodiesel within the sugarcane industry, such as palm [2] and algae [27]. It is important to exploit the regional feedstock availability [33] in order to reduce the dependence on soybean and keep the price stable. Projections indicate that the diesel consumption will increase 65% until 2030 [34], which would require around 8 million ha of soybean considering the current blend of biodiesel (although the government is considering to increase the biodiesel blend to 7% in 2014). But given the conditions discussed above, policy models focused on the local consumption of biodiesel could be a better alternative to increase the use of biofuels in Brazil while reducing the environmental burdens related to ethanol production. Besides the economic feasibility of the proposed integrated system, the decision to expand the core business and to apply new technologies to the sugarcane sector depends on the supply chains and on the market demand [33].

6. Conclusion The high demand for diesel in sugarcane production and the relevance of soybean as feedstock for biodiesel production in Brazil encourage efforts toward the integration of these bioenergy systems. In this study, we investigated the environmental and economic implications of the diesel displacement through the integrated production of soybean biodiesel and sugarcane ethanol. The results indicate that the integration is likely to reduce the fossil energy consumption in the ethanol life cycle, with minor contributions to the GHG emissions reduction. The integration is also economically feasible, assuming as main income the savings from diesel displacement. Regardless the high uncertainties, the economic performance could be enhanced through governmental incentives founded on the reduction of fossil energy use and on

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015

10

S.P. Souza, J.E.A. Seabra / Energy Conversion and Management xxx (2014) xxx–xxx

improvements in the logistics. Additionally, in a scenario of higher prices due to the low sulfur diesel, the use of biodiesel becomes more attractive. Provided that the sugarcane sector is responsible for around 4% of the diesel consumption in Brazil, any action to reduce the fossil diesel use in this sector can be relevant countrywide. The soybean– sugarcane integrated system evaluated in this study is among these options. Such integrated system can only be applied in areas suitable for sugarcane and soybean cultivation, which include mainly the states of São Paulo, Mato Grosso and Goiás. Conversely, another important limitation is that the current spare capacity of the biodiesel sector is over 50%, which has discouraged new investments in Brazil [35]. Still, there is a range of integrated system opportunities for the sugarcane sector that must be properly considered and studied in order to indicate the most suitable designs for each region given the local opportunities. Acknowledgements Authors thankfully acknowledge Fapesp and CNPq for the financial support. Authors are grateful to the blind reviewers for their valuable comments on the manuscript. Appendix A (see Tables A1–A5). References [1] Cherubini F. The biorefinery concept: using biomass instead of oil for producing energy and chemicals. Energy Convers Manage 2010;51:1412–21. [2] Souza SP, de Ávila MT, Pacca S. Life cycle assessment of sugarcane ethanol and palm oil biodiesel joint production. Biomass Bioenergy 2012;44:70–9. [3] Ometto AR, Ramos PAR, Lombardi G. The benefits of a Brazilian agro-industrial symbiosis system and the strategies to make it happen. J Clean Prod 2007;15:1253–8. [4] Lombardi G, Ramos PAR, Ometto AR, Corsini R, Ones OP, Cárdenas LZ. A comparative study of GERIPA ethanol with other fuels. Rev Ing E Invest 2009;29:77–80. [5] Rabelo SC, Carrere H, Maciel Filho R, Costa AC. Production of bioethanol, methane and heat from sugarcane bagasse in a biorefinery concept. Bioresour Technol 2011;102:7887–95. [6] Grisi EF, Yusta JM, Khodr HM. A short-term scheduling for the optimal operation of biorefineries. Energy Convers Manage 2011;52:447–56. [7] Seabra JEA, Macedo IC, Chum HL, Faroni CE, Sarto CA. Life cycle assessment of Brazilian sugarcane products: GHG emissions and energy use. Biofuels Bioprod Biorefining 2011;5:519–32. [8] ANP. Biocombustíveis, Biodiesel: boletim mensal do biodiesel, Jan–Dez 2012 [Biofuels, Biodiesel: Monthly bulletin of biodiesel, Jan–Dec 2012]. Brasília: Agência Nacional de Petróleo, Gás Natural e Biocombustíveis [Brazilian National Agency of Petroleum, Natural Gas and Biofuels]; 2013. [9] Oliverio JL, Barreira ST, Rangel SCP. Integrated biodiesel production in barralcool sugar and alcohol mill. Int Sugar J 2007;109:12. [10] Souza SP, Seabra JEA. Environmental benefits of the integrated production of ethanol and biodiesel. Appl Energy 2013;102:5–12. [11] Fazal MA, Haseeb ASMA, Masjuki HH. Biodiesel feasibility study: an evaluation of material compatibility; performance; emission and engine durability. Renew Sustain Energy Rev 2011;15:1314–24. [12] ISO 14040. Environmental Management – Life Cycle Assessment: Principles and Framework; 2006. [13] ISO 14044. Environmental Management – Life Cycle Assessment: Requirements and Guidelines; 2006. [14] IPCC. Climate change 2007: the physical science basis: contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. New York: Cambridge University Press; 2007.

[15] IPCC. Prepared by the National Greenhouse Gas Inventories Programme, Eggleston H.S., Buendia L., Miwa K., Ngara T. and Tanabe K., editors. 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Japan: IGES; 2006. [16] Soares P. Interview regarding the actual outlook of the biodiesel sector in Brazil; 2014. [17] Implementation of biodiesel plants. BNDES Workshop – Invest. Biodiesel, Rio de Janeiro: BNDES; 2006. [18] IMEA. Preços da soja e da torta [Meal and Soybean prices]. Cuiabá, MT: Instituto Mato Grossense de Economia Agropecuária [Mato Grossense Institute of Agricultural Economics]; 2012. [19] UFV. Indicadores de Preços Médios - Glicerina Loira [Average Prices Index Glycerin]. Aboissa/Centro de Referência da Cadeia de Biocombustíveis para a Agricultura Familiar, UFV; 2014. [20] ANP. Preço dos combustíveis [Fuel prices]. Sistema de Levantamento de Preços [Price Survey System]. Brasília: Brazilian National Agency of Petroleum, Natural Gas and Biofuels; 2013. [21] IMEA. Custo de Produção Soja – safra 2011/12 [Soybean Cost Production – 2011/12 crop season]. Mato Grosso: Instituto Mato Grossense de Economia Agropecuária [Mato Grossense Institute of Agricultural Economics]; 2012. [22] SEFAZ. Regulamento do ICMS [ICMS Regulation]. Título III - da obrigação principal [Title III – about the main obligation]. Capítulo II: do cálculo dos impostos [Chapter II: about the tax calculation]. Seção II: das alíquotas [Section II: about the aliquots] n.d. [23] ADB. Handbook for integrating risk analysis in the economic analysis of projects. Manila: Asian Development Bank; 2002. [24] EPE. Brazilian Energy Balance 2013, Year 2012. Rio de Janeiro: Brazilian Energy Research Office; 2013. [25] Cherubini F, Jungmeier G. LCA of a biorefinery concept producing bioethanol, bioenergy, and chemicals from switchgrass. Int J Life Cycle Assess 2009;15:53–66. [26] Cherubini F, Ulgiati S. Crop residues as raw materials for biorefinery systems – a LCA case study. Appl Energy 2010;87:47–57. [27] Souza SP, Gopal A, Seabra JEA. Life cycle environmental benefits of a cane– algae biorefinery design. In: 3rd Int. Conf. Algal Biomass Biofuels Bioprod., vol. 1, Toronto, Canada; 2013. [28] Kim S, Dale BE. Life cycle assessment of various cropping systems utilized for producing biofuels: bioethanol and biodiesel. Biomass Bioenergy 2005;29:426–39. [29] Rathmann R, Szklo A, Schaeffer R. Targets and results of the Brazilian biodiesel incentive program – has it reached the promised land? Appl Energy 2012;97:91–100. [30] MDIC. Brazilian importation. Brasília: Ministry of Development, Industry, and Trade; 2012. [31] Leung DYC, Wu X, Leung MKH. A review on biodiesel production using catalyzed transesterification. Appl Energy 2010;87:1083–95. [32] CONAB. Brazilian crop assessment: sugarcane, crop 2013/2014, Second Estimate. Brasília: National Food Supply Company; 2013. [33] Kokossis AC, Yang A. On the use of systems technologies and a systematic approach for the synthesis and the design of future biorefineries. Comput Chem Eng 2010;34:1397–405. [34] MME. Plano Nacional de Energia 2030 [2030 Energy National Plan]. Brasília: Ministry of Mines and Energy – MME, Brazilian Energy Research Company – EPE; 2007. [35] MME. Boletim mensal dos combustíveis renováveis [Monthly report on renewable fuels]. Brasília: Ministério de Minas e Energia [Ministry of Mines and Energy]; 2013. [36] Hassuani SJ, Leal MRLV, Macedo IC. Biomass power generation: sugar cane bagasse and trash. Piracicaba, SP, Brazil: United Nations Development Programme (UNDP) and Sugarcane Technology Center (CTC); 2005. [37] Capaz R. Estudo do desempenho energético da produção de biocombustíveis: aspectos metodológicos e estudos de caso [energy performance of biofuel production: methodological aspects and case studies]. Thesis. Federal University of Itajuba; 2009. [38] Dias MOS, Cunha MP, Jesus CDF, Rocha GJM, Pradella JGC, Rossell CEV, et al. Second generation ethanol in Brazil: can it compete with electricity production? Bioresour Technol 2011;102:8964–71. [39] 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 Policy 2012;14:399–410. [40] Mourad AL, Walter A. The energy balance of soybean biodiesel in Brazil: a case study. Biofuels Bioprod Biorefining 2011;5:185–97. [41] Ferrari RA, Oliveira V da S, Scabio A. Biodiesel from soybean: characterization and consumption in an energy generator. Quím Nova 2005; 28: 19–23. [42] CONAB. Brazilian crop assessment: Grains 2011/2012, First Estimate. Brasília: National Food Supply Company; 2011.

Please cite this article in press as: Souza SP, Seabra JEA. Integrated production of sugarcane ethanol and soybean biodiesel: Environmental and economic implications of fossil diesel displacement. Energy Convers Manage (2014), http://dx.doi.org/10.1016/j.enconman.2014.06.015