Fuel 208 (2017) 329–336

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Life cycle assessment of biodiesel produced by the methylic-alkaline and

ethylic-enzymatic routes
Cristina Santos de Mello Marcia, Gomes D’Amato Villardi Hugo ⇑, Ferreira Young André,
Luiz Pellegrini Pessoa Fernando, Medeiros Salgado Andrea
Escola de Química, Universidade Federal do Rio de Janeiro, Av. Athos da Silveira Ramos, 149, Ilha do Fundão, Rio de Janeiro 21941-909, Brazil

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

Article history: A comparative environmental impact analysis of the soybean biodiesel production by two different tech-
Received 8 April 2017 nologies was done in this study. The production routes evaluated were the alkali-catalyzed (catalyst:
Received in revised form 2 July 2017 sodium hydroxide) methylic transesterification and the enzyme-catalyzed (catalyst: lipase) ethylic trans-
Accepted 5 July 2017
esterification. In an early work, simulations of the biodiesel production processes with the software
Aspen HYSYS, from AspenTech Inc., were carried out. Now, a life cycle assessment (LCA) of the entire bio-
diesel production chain is done. The inventories related to each production subsystem were developed
Keywords:
based on the mass and energy balances obtained from the simulations and on literature information.
Biodiesel
Life cycle assessment
The results clearly indicated the best environmental performance of ethanol over methanol and of the
Process simulation enzymatic technology over the traditional alkaline technology, but also demonstrated some bottlenecks
Alkaline catalysis that should be attacked in a seek for more sustainable solutions.
Enzymatic catalysis Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction have considered soy production in the life cycle assessments of

Biodiesel is a biodegradable fuel consisting of a mixture of fatty
acids alkyl esters, which can be produced from raw materials such
as vegetable oils, animal fats and residual oils [18]. Homogeneous
alkaline transesterification is the most common biodiesel produc-
tion process of industrial scale [34,35,37]. Reaction occurs between
the oil and short chain alcohols, such as methanol and ethanol with
an alkaline catalyst, such as NaOH [9,11,10,19].
However, in a search for more economical and sustainable solu-
tions, there is an evident tendency in the scientific literature the
study processes that use heterogeneous catalysts [8,16]. Specially,
and already with industrial application, stands out the enzymatic
catalysis, where lipases are used as catalysts in transesterification
[21,36]. This process seems to have certain economic [36] and
environmental [13] advantages over other emerging technologies,
such as the supercritical route [17]. The greatest obstacle to the
definitive implementation of the enzymatic technology at indus-
trial level is probably the high catalyst cost [36].
In Brazil, about 75% of the feedstock’s used for biodiesel produc-
tion is soybean. Soybean is also the most used feedstock for biodie-
sel production in the world and, because of that, many authors
⇑ Corresponding author.
E-mail address: hugovillardi@eq.ufrj.br (H. Gomes D’Amato Villardi).
the biodiesel production chain [2,3,25].
Although the most used alcohol for biodiesel production is
methanol, due to its non-renewable origin, production with etha-
nol appears as a particularly interesting alternative in an environ-
mental point of view. It is even more interesting to Brazil, which
has one of the biggest and well established ethanol industries of
the world [12]. The Brazilian ethanol industry has also great inter-
dependence with the industries of United States and Europe, the
two major methylic biodiesel producers of the world [7]. An effi-
cient integration of both industries is an interesting, if not essen-
tial, path to the definitive establishment of a green fuels culture.
Life Cycle Assessment (LCA) is an environmental management
tool used to analyze and evaluate the environmental impacts of a
product, along its entire life cycle, i.e., since its formation until
its complete disappearance in the nature [29,32]. This type of anal-
ysis is complex and generally limited by data availability and gen-
eralizations [14,20]. In this context, it is necessary to understand
LCA as an evolution tool that measures potential environmental
impacts, not real impacts [4]. Thus, the objective of LCA is not to
define solutions, but rather to provide the necessary basis for
decision-making towards more sustainable scenarios [28,29,32].
The life cycle assessment carried out in this study was con-
ducted in accordance with ISO 14.040 and had the objective to
compare the environmental performance of soybean biodiesel
obtained by two different routes: the methylic-alkaline

http://dx.doi.org/10.1016/j.fuel.2017.07.014
0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
330 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336
(traditional) production route, and the ethylic-enzymatic route.
This work is a direct continuation of our paper ‘‘Biodiesel produc-
tion by the methylic-alkaline and ethylic-enzymatic routes: dis-
cussion of some environmental aspects” [5].
In the former work, biodiesel synthesis was simulated in the
software Aspen HYSYS, from AspenTech Inc., and a subsequent
analysis was carried out in terms of the 12 Principles of Green
Chemistry [1] and calculation of some sustainability metrics [31].
2. Methodology
In this work, biodiesel was considered as an energy source for
compression ignition (diesel cycle) engines. In order to quantify
the energy to be used, a reference flux was determined to indicate
the quantity of product needed to meet this objective based on the
calorific value of soybean biodiesel (39.9 MJ/kg), since it does not
show significant variation as a function of the chosen alcohol.
2.1. Definition of the production systems
Biodiesel production was divided in four main subsystems: soy-
bean oil production (SOP), anhydrous ethanol production (ETP) or
methanol production (MEP), caustic soda (sodium hydroxide) pro-
duction (CSP) and biodiesel synthesis (BIS). A scheme of the pro-
duction system and its main subsystems can be viewed in Fig. 1.
In this work, we consider a LCA approach of the type ‘‘cradle to
gate”, where impacts associated with biodiesel production are
listed from resource extraction to the end of the production chain,
with inclusion of the product distribution stage.
The life cycle assessment does not include the environmental
loads related to lipases, since it is believed that the impacts of bio-
logical catalysts are insignificant and because it is possible to reuse
it for several cycles [13,24]. Plant and equipment construction
were also not considered in the analysis, since the useful life of
these assets is considerably large, resulting in low relevant envi-
ronmental impacts when compared to daily environmental
impacts of the production process.
In addition to these main subsystems, there are intermediate
subsystems, which provide the necessary support structure for
production. These are: the production of diesel, which is used to
transport materials and to generate thermal energy; the produc-
tion of fuel oil, also used in the production of thermal energy; elec-
Fig. 1. Main production subsystems.
tricity generation; production of fertilizers, pesticides and
limestone, necessary for the agricultural stages; and the transport
subsystem, which includes the analysis of the environmental load
attributed to the transport stages between the production
subsystems.
For big countries like Brazil, production decentralization is a
very interesting strategy, since the reduction of the distances
between producers and consumers would minimize transport
costs [15]. Therefore, in this work we considered that the biodiesel
production units could be installed in two different places: the
States of Goiás (GO) and Rio Grande do Sul (RS). These places
require different distribution systems with different environmen-
tal performances.
2.2. Environmental impact evaluation
The environmental impact factors evaluated were non-
renewable energy usage (NEU), liquid effluents generation (LEG),
solid waste generation (SWG), the greenhouse effect potential
(GEP), the potential of ozone layer depletion (OLD), the potential
of ozone photochemical formation (OPF) and the acidification
potential (ACP). It was also evaluated the CO2 balance and the pet-
roleum, natural gas, coal and water consumption in each produc-
tion system. Once all the factors have been calculated for each
system, an overall impact indicator was calculated, as a way to
quantitatively base the choice for the more sustainable technology
and place. The impact indicator was calculated as a sum of the
products between each factor and its normalized relevance factor
[33,23]. The relevance factor is calculated as the ratio between
the biggest value of the impact factor among the production sys-
tems and the annual value of the factor in Brazil.
The criterion adopted by Vianna [33] and Marzullo [23] was
used in the calculation of the LEG factor. The necessary amount
of water added to the effluents was calculated to promote dilution
of these effluents until critical limits determined by the ‘‘Regula-
tion on requirements for the discharge of wastewater into surface
waters” (Abwasserverordnung – Abw) of 1997 adopted in the BASF
method [27]. These are 75 mg/L for COD (chemical oxygen
demand) and 15 mg/L for BOD (biochemical oxygen demand).
The energy-mass ratios indicated in Table 1 were used to calcu-
late the required amount of non-renewable energy (NEU). These
can be directly obtained from the inventories of each subsystem,
 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336 331

Table 1 which is equivalent to 8,433.33 kg of CO2 per hectare of planted

Energy-mass ratios for non-renewable energy sources.
Resource MJ/kg
Oil 45.8
Natural gas 38.0
Coal 19.1
available as Supporting Information. The calculation of the NEU
factor was based on the amounts of coal, oil and natural gas used
for energy production. For coal, all mass entry was considered a
source of energy, except in the production of agricultural defen-
sives in the sugarcane production stage, where part of the coal
would be directed as a non-energetic input in the production of
herbicides and insecticides. Thus, with the exception of the ethanol
production subsystem, all the coal was used for energy purposes.
Table 2 summarizes the formulas and definitions of each factor
that cannot be obtained directly from the simulations. Table 3
shows the values of the environmental impact factors associated
with each of the atmospheric emissions evaluated in this work.
In all analyzed cases, it was assumed a production of 1000 kg of
biodiesel.
soy [23]. As it is necessary 0.347 hectare to produce 1 ton of soy,
Reference there is a fixation of 2,926.37 kg of CO2 for each ton of produced
[28] soy, or 1.83 kg of CO2 per kg of dry matter. Table 4 shows the total
[26] CO2 balance involved in the biodiesel production in Goiás and Rio
[28] Grande do Sul, considering the two production routes under anal-
ysis in this work.
It is observed that in all analyzed routes the total CO2 balance is
negative because the quantities absorbed in the agricultural stages
are greater than the sum of the quantities emitted in all subsys-
tems. Another observation that can be made is that the ethylic
route, regardless of the location of the biodiesel plant, presents a
more negative balance than the methylic route. This is due to the
agricultural origin of the alcohol chosen for this route, while the
methylic technology still has significant emissions attributed to
the methanol production subsystem. It is worth mentioning that
the significant CO2 emission in the methanol production subsys-
tem refers to natural gas consumption, both for thermal energy
generation and methanol synthesis.
In both methylic and ethylic routes, it is verified that the emis-
sions due to the transport and biodiesel synthesis subsystems have
little weight in comparison to the others.

3. Results 3.2. Fossil fuel consumption

3.1. CO2 balance The material demand analysis can adopt as reference any inflow
or outflow in the product inventories. Among the analyzed mate-
Biodiesel synthesis and energy production subsystems have a rial resources, there are oil, natural gas and coal, all of them non-
positive contribution to atmospheric carbon levels in the transport. renewable energy sources. These three fossil fuels are responsible
The agricultural stages (sugarcane and soy production) contribute for attend more than 50% of the Brazilian energy matrix and an
in the other direction, with chemical sequestration. Thus, the sug- even greater fraction of the world energy consumption [32]. These
arcane and soy production subsystems allow the capture of carbon resources contribute significantly to the emission of local pollu-
in the form of CO2, which will be considered as an input flow since tants, such as hydrocarbons and CO2, which act in the formation
it is a natural resource [30]. of tropospheric ozone or promote the greenhouse effect.
Sugarcane harvesting in the State of São Paulo (SP) (where great For calculation of natural gas consumption, it was considered
part of the ethanol production facilities are supposed to be located) the inclusion of raw natural gas, which had its quantity in m3 con-
in 2012/2013 was about 74.94 tons per hectare. It is estimated that verted to mass, and of natural gas used as energy source, which
about 16.20 tons per hectare of carbon are sequestered annually by had its quantity in MJ converted to mass. For these conversions,
the sugarcane. So, it can be inferred that 793 kg of CO2 is captured it was adopted the values of 0.74 kg/m3for density [6] and 38 MJ/
per ton of cane produced [22]. According to Sugawara [30], this kg for the calorific value [26].
value would be 653 kg of CO2 per ton of cane. So, we decided to Table 5 presents the fossil fuel consumption profile in each pro-
use the mean value of 723 kg to calculate the amount of CO2 duction system. According to this data, the most consumed fossil
absorbed in the inventories. For the production of 1000 kg of bio- fuel in themethylic-alkaline production system is natural gas,
diesel, 172.03 kg of ethanol are necessary, which means that which occurs due to the large amount of this input required for
1,711.52 kg of CO2 are swallowed in this stage. thermal energy production (96%) and as raw material for methanol
In the case of the calculation of carbon absorption inthe soy pro- production (4%). There is no significant difference in the amount of
duction, for each 4.6 tons of dry matter per hectare, there is a petroleum and coal consumed among the production systems,
reduction equivalent to 2.3 tons of carbon in the atmosphere, what makes the methylic technology more environmentally
expensive in this sense. Differences regarding localities influence
Table 2
only the transport subsystem and the transport stages inserted in
Impact factors: formulas and definitions. the other subsystems.
Fig. 2 shows an approximation of the distances covered by etha-
Factor Equation (units) Definition
P
nol transportation to the biodiesel plants and biodiesel distribution
GEP GEPi
ei ðkg eq CO2 Þ It measures the capability of infrared to the consumer sites.In the transport subsystem, distances cov-
i
radiation absorption of a certain substance
ered by biodiesel from plant to the distribution sites and caustic
in comparison to CO2.
P soda from its production plant to the biodiesel plants show small
OLD OLDi
ei ðkg eq CFCÞ It measures the potential of a certain
i
substance to favor the ozone layer depletion differences when comparing production systems located in Goiás
to CFCs. and Rio Grande do Sul. So, for the methylic route, there is no big
P
OPF OPF i
ei ðkg eq C 2 H4 Þ It measures the oxidation potential of a
economic or environmental difference between one or another
i
certain substance in the troposphere in
comparison to ethene.
place to install the production facilities. However, when analyzing
P the distances covered by ethanol from its production plants until
ACP ACP i
ei ðkg eq SO2 Þ It measures the propensity of a certain
i
substance to emit H+in comparison to sulfur the biodiesel production plants, differences become significant.
dioxide. Therefore, locality is important if biodiesel is supposed to be pro-
ei are the atmospheric emissions showed in Table 3 and GEPi, OLDi, OPFi, ACPi are duced with ethanol, because of the integration between the biodie-
the factors associated with each emission. sel and sugarcane industries.
332 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336

Table 3
Environmental impact factors associated with some atmospheric emissions.

Atmospheric Emissions Greenhouse Effect Potential (GEPi) Ozone Layer Depletion (OLDi) Ozone Photoche. Formation (OPFi) Acidification Potential (ACPi)
CO2 1.00 – – –
SO2 – – – 1.00
NOx – – – 0.70
CH4 21.00 – 0.01 –
Hydrocarbons – – 0.42 –
Halogenated 4000.00 1.00 – –
NH3 – – – 1.88
N2O 310.00 – – –
HCl – – – 0.88
HF – – – 1.20

Table 4
CO2 balance in the biodiesel production systems.

Subsystems Ethylic route (GO) Methylic route (GO) Ethylic route (RS) Methylic route (RS)
3 3 3
Soybean oil production 2.76  10 2.72  10 2.74  10 2.70  103
Ethanol production 1.68  103 – 1.68  103 –
Methanol production – 1.61  103 – 1.61  103
Soda production – 3.61  100 – 3.61  100
Transport 4.76  101 5.91  101 7.39  101 5.82  101
Biodiesel synthesis 6.12  101 8.57  101 6.12  101 8.57  101
Total balance 4.33  103 9.60  102 4.28  103 9.41  102

Table 5
Fossil fuel consumption profile by production system.
Rote Oil Natural gas
Methylic (GO) 23.2% 76.0%
Methylic (RS) 24.1% 75.1%
Ethylic (GO) 89.9% 7.0%
Ethylic (RS) 90.6% 6.6%
Specifically in relation to petroleum derivatives, there is a
slightly higher consumption when the biodiesel facilities are
located in the State of Rio Grande do Sul. It occurs because in all
the analyzed routes the main consumption of oil occurs in the soy-
bean oil production subsystem, which contemplates the agricul-
tural production of soy. This subsystem presents about 2.50
times greater consumption of diesel (and, consecutively, petro-
leum) in the transportation stages of fertilizer and limestones
when the plant is installed in Rio Grande do Sul.
In addition, the consumption of petroleum derivatives in the
transport subsystem and in the biodiesel synthesis subsystem,
between 20 and 30 kg per 1000 kg of biodiesel, must be high-
lighted. In the last one, this demand is related to the production
of electricity and thermal energy. Petroleum derivatives are also
present in the production of pesticides, fertilizers, limestone, elec-
tricity and diesel for the agricultural machinery, besides soybean
oil processing.
Coal consumption occurs in small quantities and refers to the
inputs involved in electricity generation and in the production of
diesel from petroleum. The highlight is the consumption of elec-
tricity in the fertilizers production stage. It is also consumed in
small quantities in the methanol and caustic soda production
subsystems.
It can be verified that in the two studied routes and in the two
States, the soybean oil production subsystem presented the great-
est consumption of the analyzed fossil materials, except for the
natural gas consumed in the methylic route. Although the subsys-
tems that involve agricultural stages, both for sugarcane produc-
tion and for soybean production, also consume natural gas in
reasonable quantities (approximately 70 kg per 1000 kg of biodie-
sel), which is used in the production of energy and as raw material
for fertilizer production, the demand for this input by the methanol
production subsystem is much bigger, approximately 500 kg per
Coal 1000 kg of biodiesel [26].
0.8%
0.8%
3.1% 3.3. Water consumption
2.8%
Another important aspect is the consumption of water, since it
is an essential resource for the maintenance of the economy and all
sectors of society. Given its importance, water management has
become an emerging area of great importance for industry. There-
fore, it is imperative to control water consumption in every stage of
the biodiesel production chain.
Table 6 shows the water consumption in each production sys-
tem and subsystems. Water consumption was separated from the
other subsystems in order to enable the verification of its contribu-
tion in details.
The subsystem that most contributes to the total consumption
of water in the biodiesel production is the soybean oil production
subsystem. Additionally, it is verified that, although there is signif-
icant water consumption during soybean oil extraction, in the agri-
cultural stage the demand for water resources is highly
preponderant, due to irrigation.
It is observed that the biodiesel synthesis subsystem presents
higher consumption of water in the alkaline route, since in this
case there is a greater demand for washing in order to reduce
the sodium concentration and attend biodiesel specifications. It is
also possible to observe the influence of the caustic soda produc-
tion subsystem in the total water consumption.
Comparing the alcohol production subsystems, ethanol produc-
tion requires a greater amount of water than the methanol produc-
tion. Although the last one requires a reasonable consumption of
water for electricity generation, ethanol production demands it
in significant quantities both for the ethanol production process
itself and, mainly, for sugarcane cultivation.
The ethylic route presents a higher demand for water than the
methylic route, as a function of the largest amount of soybean oil
required for biodiesel production (1000 kg). This higher consump-
tion of oil implies a greater production of soybean and, conse-
quently, the greater impact on water usage.
 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336 333

Fig. 2. Approximated distances covered by ethanol transportation to the biodiesel plants and biodiesel distribution to the consumer sites.

Table 6
Water consumption per production system and subsystems (kg/1000 kg biodiesel).

Met. Route (GO) Met. Route (RS) Et. route (GO) Et. route (RS)
Soy Production 2.29  106 2.29  106 2.32  106 2.32  106
Soybean Oil Production 9.00  103 9.00  103 9.00  103 9.00  103
Ethanol/Methanol 4.32  103 4.32  103 1.05  104 1.05  104
Soda Production 6.60  103 6.60  103 – –
Transport 3.01  102 2.96  102 2.42  102 3.76  102
Biodiesel Synthesis 2.02  103 2.02  103 5.81  102 5.81  102
Total 2.31  106 2.31  106 2.34  106 2.34  106

3.4. Analysis of the environmental impact factors
Table 7 shows the environmental impacts caused by each pro-
duction system based on the seven categories evaluated in this
work.Fig. 3 shows the relative factors for each system in the envi-
ronmental impact categories.
It is possible to observe that the methylic route presents values
greater or almost equal to those verified for the ethylic route in all
the environmental impact categories analyzed, except for the
potential of ozone photochemical formation (OPF).
As no significant variation was observed between the impacts
caused by the two proposed locations (GO and RS), either in the

Table 7
Environmental impact caused by each biodiesel production system as a function of technology and locality.

Impact categories Met. route (GO) Met. route (RS) Et. route (GO) Et. route (RS)
Non-renewable energy usage (MJ) 2.32  104 2.36  104 7.31  103 8.22  103
Liquid effluents generation (kg) 3.82  103 3.85  103 2.16  103 2.25  103
Solid waste generation (kg) 7.62  102 7.62  102 7.81  102 7.81  102
Greenhouse effect potential (kg CO2 eq.) 4.91  102 4.71  102 4.27  103 4.22  103
Potential of ozone layer depletion (kg CFC eq.) 1.79  104 1.79  104 0.00  100 0.00  100
Potential of ozone photochemical formation (kg C2H4 eq.) 2.62  100 2.63  100 5.55  100 5.59  100
Acidification potential (kg SO2 eq.) 6.18  100 6.36  100 4.74  100 5.19  100
334 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336

Fig. 3. Relative factors for each biodiesel production system and locality in the evaluated environmental impact categories.

ethylic or methylic route, by simplification, only one of these sites
was selected to carry out a more detailed comparison of the envi-
ronmental impacts among the two different technologies. A contri-
bution analysis of each subsystem composing the evaluated
production routes is presented in Tables 8 and 9.
Regarding the use of non-renewable energy, it is observed that
the major contribution in the ethylic route comes from the soybean
oil production subsystem. This observation was already mentioned
in the material consumption analysis, when the great consumption
of fossil fuels in the soy production stage was verified. When ana-
lyzing the relative value for the methylic route on this impact cat-
egory, it is observed that, in addition to the soybean oil production
subsystem, the methanol production subsystem has a much more
significant character, due to natural gas consumption as thermal
energy source.
The generation of liquid effluents and solid waste has also sig-
nificant contribution from the soybean oil production subsystem,
where fertilizers account for a great part. Effluents in the form of
sodium chloride ions originate from potassium fertilizers produc-
tion, while dissolved solids, for example, are generated in the pro-
duction of phosphate fertilizers.
In the alkaline route, however, there is also a significant contri-
bution to the liquid effluent generation from the biodiesel synthe-
sis subsystem. It happens because, in this route, it is necessary to
consume a large amount of water in the washing step in order to
the final product to reach the specified limits for sodium foreseen
by the regulatory agency [34,35]. Water here is released as a liquid
effluent, diluting the contaminants generated during the produc-
tion process.
The generation of solid residues, both in the methylic and
ethylic routes, is almost completely due to the soybean oil produc-
tion subsystem. These residues come mainly from soybean produc-
tion, as already explained. The slight difference presented in this
factor between the two routes occurs as a function of the reference
flow. The ethylic-enzymatic route requires 1,054.37 kg of soybean,
whereas the methylic-alkaline route requires 1,040.29 kg of it.
Analysis of the potential to cause greenhouse effect shows that,
due to the agricultural stages of both routes, the amount of seques-
tered carbon is greater than the emitted amount, what is an excel-
lent result. The ethylic route presents a negative value related to
carbon sequestration during two agricultural stages: soybean pro-
duction (included in SOP) and sugarcane production (included in
ETP). On the other hand, the methylic route presents a significant
positive contribution to the CO2 release from the MEP subsystem.
This difference makes the ethylic route ‘‘greener” than the
methylic route in this sense.
In the MEP subsystem, CO2 emissions are present in all stages.
The more relevant are those from natural gas processing, for ther-
mal energy generation and methanol synthesis itself. The contribu-
tion to the potential of ozone layer depletion was mainly due to the
presence of the halogenated hydrocarbon dichloromethane in the
process.
The potential of ozone photochemical formation of the ethylic
route is greater than the potential of the methylic route. In the first
one, the main contribution to this factor is the agricultural stage of
the ethanol production subsystem, where incomplete burning of
the straw results in hydrocarbon emissions. In the methylic route,
the impact of this category is mainly due to the methanol produc-
tion subsystem, where hydrocarbons are released during natural
gas processing. Both technologies present significant contribution
from the soybean oil production subsystem.
Emissions related to soil acidification (NH3, NOx, SOx, HCl, HF,
SOx, SO2, H2S) are present in the soybean oil production subsystem
and are generated in almost all steps. However, NH3, NOx and SOx
stand out as the major emissions. In the oil extraction stage, fuel oil
is burned, releasing great amounts of SOx. The agricultural stage is
the main source of NH3 and NOx, in the production of fertilizers
and in their application in the ground. Important emissions are
also observed by diesel burning in the agricultural machinery
and in the transport phases. In the ethylic route, these emission
sources are present in the sugarcane cultivation, but the natural
gas consumption for methanol production in the methylic route

Table 8
Contribution of the methylic-alkaline route (GO) subsystems by environmental impact category.

Impact categories/Subsystems SOP MEP CSP TRA BIS Total

Non-renewable energy usage (MJ) 4.61  103 1.57  104 6.51  101 1.18  103 1.74  103 2.32  104
Liquid effluents generation (kg) 1.75  103 2.75  102 3.27  100 1.12  102 1.67  103 3.82  103
Solid waste generation (kg) 7.61  102 3.98  101 6.09  101 3.30  102 4.96  102 7.62  102
Greenhouse effect potential (kg CO2 eq.) 2.70  103 2.05  103 4.62  100 6.28  101 8.76  101 4.91  102
Potential of ozone layer depletion (kg CFC eq.) 0.00 0.00 1.79  104 0.00 0.00 1.79  104
Potential of ozone photochemical formation (kg C2H4 eq.) 7.60  101 1.80  100 9.85  104 4.11  102 1.55  102 2.62  100
Acidification potential (kg SO2 eq.) 3.44  100 2.05  100 5.32  102 5.65  101 8.00  102 6.18  100
 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336 335

Table 9
Contribution of the ethylic-enzymatic route (GO) subsystems by environmental impact category.

Impact categories/Subsystems SOP ETP CSP TRA BIS Total

3 2 2 3 3
Non-renewable energy usage (MJ) 4.68  10 4.35  10 9.49  10 1.24  10 7.31  10 4.68  103
Liquid effluents generation (kg) 1.78  103 5.06  101 8.99  101 2.41  102 2.16  103 1.78  103
Solid waste generation (kg) 7.71  102 9.50  100 2.65  102 3.58  102 7.81  102 7.71  102
Greenhouse effect potential (kg CO2 eq.) 2.73  103 1.65  103 5.05  101 6.26  101 4.27  103 2.73  103
Potential of ozone layer depletion (kg CFC eq.) 0.00 0.00 0.00 0.00 0.00 0.00
Potential of ozone photochemical formation (kg C2H4 eq.) 7.71  101 4.74  100 3.31  102 1.11  102 5.55  100 7.71  101
Acidification potential (kg SO2 eq.) 3.48  100 7.48  10-1 4.55  10-1 5.72  102 4.74  100 3.48  100

is also an important source. In fact, the proposed methylic-alkaline
route is considerably more harmful in this impact category.
Finally, the impact factors generated can be used in the calcula-
tion of an overall environmental impact indicator, which summa-
rizes the discussions made in this paper in a quantitative way. It
is shown in Table 10.
The methylic-alkaline route, traditional biodiesel production
technology, is more harmful to environment than the ethylic-
enzymatic technology. In fact, it obtained worse results in almost
all the impact factors analyzed in this paper, consequently obtain-
ing the worst overall indicators (0.74–0.75 against 0.46–0.48 of the
ethylic route). Is not only worse, but the difference is large, and it is
due to methanol. As we show here, the renewable origin of ethanol
is responsible to a large increase in the CO2 absorption, decreasing
considerably the greenhouse effect potential of the production sys-
tem. Without this factor, the performance of both technologies is
almost equal.
The choice for ethanol generates a 100% green fuel and must be
encouraged, but it is not a perfect solution. In our quantitative
analysis, its potential of ozone photochemical formation compen-
sates great part of the environmental impacts caused by the
methylic-alkaline technology. Methanol usage is responsible for
an increase in non-renewable energy usage and in the emission
of acid gases. The biggest quantity of water required for washing
out the alkaline catalyst NaOH is the major responsible to the big-
gest generation of liquid effluents in the biodiesel synthesis. The
production of this catalyst is responsible to increase the potential
of ozone layer depletion.
Finally, it can be verified that the differences regarding the two
different locations did not have significant influence on the envi-
ronmental performances of the production systems. Therefore,
social or economic reasons should guide the decision for one or
the other site.
3.5. Sensitivity to the CSP subsystem
Because the catalyst of the enzymatic route, due to the possibil-
ity of reuse, was not considered in LCA, the sensitivity of the alka-
line route to the CSP subsystem was analyzed. The proposed
analysis was based on the exclusion of the caustic soda production
subsystem from the alkaline production system in the State of
Goiás. All environmental aspects related to the production of caus-
tic soda were excluded, as well as the contribution of diesel con-
sumption and generation of emissions related to the transport of
the catalyst to the biodiesel synthesis plant.
As a consequence, there is a reduction of 0.93% in oil consump-
tion, 0.02% in natural gas consumption, 13.62% in coal consumption
and 0.29% in the consumption of water. The major part of the coal
reported in the inventory of the NaOH production is used for elec-
tricity generation. Regarding water, although the production of
soda shows a significant water consumption, in the order of
6.6 kg/1000 kg of biodiesel, this consumption is little relevant when
compared to the soybean oil production subsystem. Thus, only a
small variation inthe water intake occurs when comparing the alka-
line production with and without the influence of the catalyst data.
Table 11 shows the differences in the seven environmental
impact categories. The OLD factor is the only one significantly
affected by the CSP subsystem, because of the large amount of
dichloromethane released as atmospheric emission. The reduction
of 1.22% in the GEP factor is related to gas emissions such as CO2,
CH4, N2O and dichloromethane, released during soda production
and transportation. Although CO2 is released in a greater quantity,
dichloromethane, even released in a small order (104), has a
4000-fold effect, which makes it the second largest contributor to
this environmental impact factor. For the acidification potential,
there is a reduction of 1.06%. It is also related to the consumption
of coal, since it has a significant sulfur and nitrogen content, which
can be released in its oxides forms. Due to the insignificant contri-
bution of the caustic soda production subsystem, no significant
changes in the other impact categories and in the overall indicator
were verified.
Based on these results, we conclude that the CSP subsystem has
no relevant effect on the quantitative results of the LCA analysis.
However, a qualitative analysis indicates how this type of catalyst
can be harmful to the environment in comparison to biological cat-
alysts, such as lipases.
The results indicated a better environmental performance for
the enzymatic route in almost all the analyzed issues, what proved
that this route is an excellent option for future industrial applica-
tions, regarding safety and sustainability. All the analysis pro-
ceeded here are based in the mass and energy balances of the
biodiesel synthesis stage simulations.

Table 10
Overall environmental impact indicator.

Impact Categories Met. Route (GO) Met. Route (RS) Et. Route (GO) Et. Route (RS) Relevance factor Normalized rel. factor
Non-renewable energy usage (MJ) 0.98 1.00 0.31 0.35 3.18  109 27.32%
Liquid effluents generation (kg) 0.99 1.00 0.56 0.58 2.31  1011 0.20%
Solid waste generation (kg) 0.98 0.98 1.00 1.00 3.32  109 28.53%
Potential to cause greenhouse effect (kg CO2 eq.) 1.04 1.00 9.07 8.96 3.77  1010 3.24%
Potential of ozone layer depletion (kg CFC eq.) 1.00 1.00 0.00 0.00 1.29  1011 0.11%
Potential of ozone photochemical formation 0.47 0.47 0.99 1.00 3.88  109 33.30%
(kg C2H4 eq.)
Acidification potential (kg SO2 eq.) 0.97 1.00 0.75 0.82 8.51  1010 7.31%
Overall impact indicator 0.74 0.75 0.46 0.48 1.16  108 –
336 M. Cristina Santos de Mello et al. / Fuel 208 (2017) 329–336

Table 11 [7] de Fagundes P de M, Padula AD, Padilha ACM. Interdependent international

Environmental impacts of the methylic-alkaline route (GO), with and without the CSP relations and the expansion of ethanol production and consumption: the
subsystem. Brazilian perspective. J Cleaner Prod 2016;133:616–30.
[8] Farooq M, Ramli A, Subbarao D. Biodiesel production from waste cooking oil
Impact category With soda Without (%) using bifunctional heterogeneous solid catalysts. J Cleaner Prod
soda 2013;59:131–40.
[9] Foglia TA, Nelson LA, Dunn RO, Marmer WN. Low-temperature properties of
Non-renewable energy usage (MJ) 2.32  104 2.32  104 0.19% alkyl esters of tallow and grease. J Am Oil Chem Soc 1997;74:951–5.
Liquid effluents generation (kg) 3.82  103 3.81  103 0.15% [10] Freedman B, Pryde EH, Mounts TL. Variables affecting the yields of fatty esters
Solid waste generation (kg) 7.62  102 7.62  102 0.08% from transesterified vegetable oils. J Am Oil Chem Soc 1984;61:1638–43.
Greenhouse effect potential 4.91  102 4.97  102 1.22% [11] Freedman B, Butterfield RO, Pryde EH. Transesterification kinetics of soybean
(kg CO2 eq.) oil. J Am Oil Chem Soc 1986;63:1375–80.
Potential of ozone layer depletion 1.79  104 0.00 100.00% [12] Hall J, Matos S, Severino L, Beltrão N. Brazilian biofuels and social exclusion:
(kg CFC eq.) established and concentrated ethanol versus emerging and dispersed
0 0 biodiesel. J Cleaner Prod 2009;17:S77–85.
Potential of ozone photochemical 2.62  10 2.62  10 0.07%
formation (kg C2H4 eq.) [13] Harding KG, Dennis JS, von Blottnitz H, Harrison STL. A life-cycle comparison
Acidification potential (kg SO2 eq.) 6.18  10 0
6.12  10 0
1.06% between inorganic and biological catalysis for the production of biodiesel. J
Cleaner Prod 2008;16:1368–78.
[14] Holma A, Koponen K, Antikainen R, Lardon L, Leskinen P, Roux P. Current limits
A significant bottleneck in the traditional technology that must of life cycle assessment framework in evaluating environmental sustainability
– case of two evolving biofuel technologies. J Cleaner Prod 2013;54:215–28.
be highlighted is the large amount of non-renewable energy [15] Iglesias L, Laca A, Herrero M, Díaz M. A life cycle assessment comparison
resources consumed during the production chain, mainly due to between centralized and decentralized biodiesel production from raw
the choice for methanol as raw material. This feature is completely sunflower oil and waste cooking oils. J Cleaner Prod 2012;37:162–71.
[16] Islam A, Taufiq-Yap YH, Chan ES, Moniruzzaman M, Islam S, Nabi MN.
off the sustainability scope that is supposed to be reached with Advances in solid-catalytic and non-catalytic technologies for biodiesel
investments in renewable fuels. Ethanol appears as an excellent production. Energy Convers Manage 2014;88:1200–18.
option, but care must be taken with the harvesting methods. The [17] Kiwjaroun C, Tubtimdee C, Piumsomboon P. LCA studies comparing biodiesel
synthesized by conventional and supercritical methanol methods. J Cleaner
incomplete burning of sugarcane straw can release a large amount
Prod 2009;17:143–53.
of hydrocarbons, contributing to ozone photochemical formation [18] Knothe G, Gerpen JVan, Krahl J. The biodiesel handbook. Champaign,
and compromising part of the benefits related to its renewable Illinois: AOCS Press; 2005.
[19] Lee I, Johnson LA, Hammond EG. Use of branched-chain esters to reduce the
origin.
crystallization temperature of biodiesel. J Am Oil Chem Soc 1995;72:1155–60.
[20] Macombe C, Leskinen P, Feschet P, Antikainen R. Social life cycle assessment of
4. Conclusions biodiesel production at three levels: a literature review and development
needs. J Cleaner Prod 2013;52:205–16.
[21] Magalhães SP, Pessoa FLP, Alves TLM. Production of biodiesel via enzymatic
The results of the LCA regarding material consumption and the palm oil ethanolysis: kinetic study. Comput Aided Chem Eng 2015;37:539–44.
evaluation of seven impact categories clearly indicated the envi- [22] MAPA – Ministry of Agriculture, Livestock and Food Supply. Evolution of
sugarcane productivity in Brazil by regions and producing States 2013 [in
ronmental benefits of the ethanol usage instead of methanol and Portuguese].
of the enzymatic technology over the traditional alkaline [23] Marzullo RCM. Eco-efficiency analysis of the vegetable oils originated from
technology. soybean and palm, aiming at biodiesel production. University of São Paulo;
2007 [in Portuguese].
Regarding the two different localities evaluated, i.e. Goiás and
[24] Ravindra P, Saralan S, Abdulla R. LCA studies for alcaline and enzyme catalyzed
Rio Grande do Sul, LCA results are affected through the transport biodiesel production from palm oil. Adv Biol Chem 2012;2:341–52.
subsystem, but the differences are not appreciable. Therefore, in [25] Reijnders L, Huijbregts MAJ. Biogenic greenhouse gas emissions linked to the
an environmental point of view, there is no such a difference in life cycles of biodiesel derived from European rapeseed and Brazilian
soybeans. J Cleaner Prod 2008;16:1943–8.
the choice for one place or the other. So, the choice should be based [26] Ribeiro BD. Contribution to the Brazilian database for support to life cycle
in social or economic aspects, according to the Brazil’s Biodiesel assessments: nitrogenated fertilizers. University of São Paulo; 2009 [in
Production and Use Program. Portuguese].
[27] Saling P, Kicherer A, Dittrich-Krämer B, Wittlinger R, Zombik W, Schmidt I,
et al. Eco-efficiency analysis by basf: the method. Int J Life Cycle Assess
Acknowledgement 2002;7:203–18.
[28] Sallaberry RR. Applyance of life cycle assessment to assess the environmental
performance of sunflower biodiesel and diesel oil. Federal University of Rio
The authors thank the financial support from CNPq and FAPERJ. Grande do Sul; 2009 [in Portuguese].
[29] Sato DH. An environmental impact study in product development. University
of São Paulo; 2010 [in Portuguese].
Appendix A. Supplementary data
[30] Sugawara ET. Comparison of the environmental performances of soybean
ethylic B5 and diesel oil by means of a life cycle assessment (LCA). University
Supplementary data associated with this article can be found, in of São Paulo; 2012 [in Portuguese].
the online version, at http://dx.doi.org/10.1016/j.fuel.2017.07.014. [31] Tallis B The Sustainability Metrics: Sustainable Development Progress Metrics
Recommended for use in the Process Industry. Rugby, UK; 2002.
[32] Viana MM. Inventory for the life cycle of sunflower oil ethylic
References biodiesel. University of São Paulo; 2008 [in Portuguese].
[33] Vianna FC. Eco-efficiency analysis: evaluation of the biodiesel and petrodiesel
[1] Anastas PT, Warner JC. Green chemistry: theory and practice. New York, economic-environmental performances. University of São Paulo; 2006 [in
USA: Oxford University Press; 1998. Portuguese].
[2] Castanheira ÉG, Grisoli R, Coelho S, Anderi Da Silva G, Freire F. Life-cycle [34] Young AF, Pessoa FLP, Queiroz EM. Comparison between biodiesel production
assessment of soybean-based biodiesel in Europe: comparing grain, oil and from soybean oil and palm oil with ethanol: design and economic evaluation.
biodiesel import from Brazil. J Cleaner Prod 2015;102:188–201. Chem Eng Trans 2015;43:325–30.
[3] Cavalett O, Ortega E. Integrated environmental assessment of biodiesel [35] Young AF, Pessoa FLP, Queiroz EM. Design and economic evaluation of
production from soybean in Brazil. J Cleaner Prod 2010;18:55–70. alternatives to effluents treatment on biodiesel production from soybean oil
[4] Costa RE. Life cycle inventory of biodiesel obtained from palm oil for and palm oil. Comput Aided Chem Eng 2015;37:1067–72.
conditions of Brazil and Colombia. Federal University of Itajubá; 2007 [in [36] Young AF, Pessoa FLP, Queiroz EM. Biodiesel production technologies:
Portuguese]. production in supercritical and enzymatic environments. Novas Edições
[5] de Mello M, Young A, Villardi H, Pessoa F, Salgado A. Biodiesel production by Acadêmicas; 2015 [in Portuguese].
the methylic-alkaline and ethylic-enzymatic routes: discussion of some [37] Zhang Y, Dubé MA, McLean DD, Kates M. Biodiesel production from waste
environmental aspects. J Cleaner Prod 2017;144:347–57. cooking oil: 1. Process design and technological assessment. Bioresour Technol
[6] EPE – Energy Research Enterprise. Notional Energy Balance 2013 [in 2003;89:1–16.
Portuguese].