b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Available online at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

A comparison of the European renewable energy directive

default emission values with actual values from operating
biodiesel facilities for sunflower, rape and soya oil seeds
in Italy
C. Buratti*, M. Barbanera, F. Fantozzi
CRB e Biomass Research Centre, Via G. Duranti e Strada S. Lucia Canetola s.n., 06125 Perugia, Italy

article info

abstract

Article history:

The European Union (EU) set a binding greenhouse gas (GHG) emission reduction target for

Received 20 June 2012

transportation biofuels and other bioliquids. In this study, the GHG emissions of biodiesel

Received in revised form

chain from sunflower, rapeseed and soybean were calculated in compliance with the

3 October 2012

European Union Renewable Energy Directive 2009/28/EC (RED).

Accepted 6 October 2012

Available online 12 November 2012

Input data used for the agricultural step were referred to the Umbrian region for
sunflower and rapeseed and to the Veneto region for soybean, while data obtained from
the main Italian biodiesel plants were employed for the processing step. Results showed

Keywords:

that GHG emissions were higher than default values reported in the RED for sunflower and

GHG emissions

rapeseed and lower for soybean. Only sunflower biodiesel does not reach the minimum

RED EU Directive

value of GHG saving (35%). The main differences with data used in the RED concern

Biodiesel

cultivation step, while the processing step has overall the same values of GHG emissions.

Sunflower

Finally, three case studies were examined in order to identify possible improvements to

Rapeseed and soya seeds

make the analyzed supply chains more sustainable.

2012 Elsevier Ltd. All rights reserved.

1.

Introduction

The perspective of oil depletion, the concerns of energy

security and global warming are the main drivers of biofuels
promotion by public authorities in industrialized countries.
While full substitution of gasoline and diesel is technically
possible, sustainability conditions are the bottleneck that
should limit the global market share of biofuels to 10e15% in
the long term [1].
The instability of oil prices and the low competitiveness of
sustainable biofuels when oil prices decrease under a certain
threshold claim for stable incentives in the early development
of biofuels markets. Due to controversies about the energy

and greenhouse gas (GHG) balances of biofuels, however,

public authorities in several countries imposed minimum
targets for biofuels to be eligible for incentives, i.e. sustainable
biofuels.
In June 2009 the European Parliament adopted the Directive 2009/28/EC [2] as part of the new Renewable Energy
Directive (RED). It amended and repealed the Directive 2003/
30/EC [3] on the promotion of biofuels, that explicitly sets
reference values for the share of biofuels of all fuels. In
particular Directive 2003/30/EC called for a biofuel market
share of 5.75% to be implemented by each Member State by
December 2010. The 2009 Directive sets instead a mandatory
overall target of a 20% share of energy from renewable sources

* Corresponding author. CRB e Biomass Research Centre, Dept. of Industrial Engineering, Via G. Duranti e Strada S. Lucia Canetola s.n.,
06125 Perugia, Italy. Fax: 39 (0)755853697.
E-mail addresses: cburatti@unipg.it (C. Buratti), barbanera@crbnet.it (M. Barbanera), fanto@unipg.it (F. Fantozzi).
0961-9534/$ e see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biombioe.2012.10.008

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

of the EUs gross final consumption in 2020, even if not every

Member State has to exactly meet this criterion. Twenty
percent will supposedly be the average over all Member
States. Depending on the current share of renewable energy
sources, each Member State has to achieve an individual
target.
The new Directive does not set targets for specific sectors,
except for the transport one. The share of energy from
renewable sources in all forms of transport has to be at least
10% of the final consumption of energy in transport by 2020.
This does not explicitly refer to the share of biofuels and could
also be achieved by electricity in cars or in rail transport.
However, it is likely that this target will act as a driver for the
use of biofuels, as other technologies are unlikely to reach
large market shares by 2020. The new 2009 Directive establishes also sustainability criteria for biofuels that do not have
a counterpart in the 2003 Directive. Biofuels not fulfilling these
newly formulated sustainability criteria may not be taken into
account for:
a calculating the shares of energy from renewable sources;
b measuring compliance with the targets set in the
Directives;
c the eligibility for financial support for biofuels.
Among the most important sustainability criteria, the GHG
emission savings from the use of biofuels or bioliquids
compared to fossil fuels have to be at least 35%. From 2017
onwards these savings have to be at least 50% and from 2018
onwards 60% for biofuels produced in installations that start
their production in 2017 or later.
The RED indicates the reference GHG emission values
(typical and default values e Annex V, part D) for the entire
production chain (cultivation, processing and transport).
Typical value is an estimate of the representative GHG emission for a particular biofuel production pathway, while default
value is calculated by multiplying typical value for processing
emissions by 1.4. Economic operators can use default values
or own actual data, applying the GHG saving calculation
methodology, described in the following paragraphs, to
demonstrate that their products satisfy GHG thresholds. Only
for the cultivation phase, the Directive delegates to the
Member States the evaluation of the GHG emissions for each
area classified as level 2 in the nomenclature of territorial
units for statistics (NUTS).
The aim of the present paper was to investigate if the
biodiesel production from rapeseed, soybean and sunflower
may be a viable alternative in Italy to the achievement of
European targets. In particular, it was investigated if these
chains are able to reach the minimum value of GHG saving,
considering the current Italian context. For this purpose,
agronomic data were referred to two NUTS 2 areas: Umbria for
sunflower and rapeseed and Veneto for soybean. This choice
was carried out considering both data availability and crop
diffusion in the Italian territory. With regard to the biodiesel
production process, data supplied by the main Italian biodiesel plants were examined.
According to the RED, each chain was divided into three
steps (cultivation, processing and transport); in this way, it
was possible to determine if GHG emissions from cultivation

27

step are lower than the default value defined in the RED and if
those relating to the processing step can be considered
consistent for the Italian production. Especially the first
comparison is important because Italy decided to apply the
actual values of typical GHG emissions from cultivation to all
the Italian NUTS 2 areas and all crops [4], renouncing the
possibility to adopt any default values for the cultivation
phase. The paper investigates if GHG emissions of the cultivation phase lower than those proposed in the Directive may
occur for crops and NUTS 2 areas analyzed, allowing
economic operators to simplify the calculation even for
national supply chains.

2.

System description

Biodiesel production system evaluated in this work consisted

of three main processes: cultivation, processing and transport
(Fig. 1).
For each crop the cultivation technique typical of the
relative NUTS 2 area was assumed.
As regards sunflower, the soil preparation practices are
constituted by a ploughing at the depth of 0.5e0.6 m after
wheat harvesting, followed by a harrowing at the depth of
0.05e0.15 m in the sowing time. Sowing is generally performed from mid March to mid April, employing 6 kg ha1 of
seeds [5]. Rapeseed is grown in rotation with wheat, whose
straw has to be removed from the field because it can lead to
an excessive macroporosity at the top layer of the soil.
Afterward it is necessary to refine the soil with a medium
depth ploughing (0.25 m), routing and harrowing operations.
Sowing is planned in the mid September with 7 kg ha1 of
seeds [6].
Veneto region was chosen for soybean cultivation because
it accounts for 48% of the national soybean production [7].
Italian soybeans are planted from the end of April through the
end of May and harvest runs from the first of October through
the end of November, then oil seeds are dried and transported
to the biodiesel production plant.
Biodiesel production consists of several processing steps
including: oil extraction, oil refining and transesterification.
Solvent extraction with hexane is considered in this study.
Solvent is removed from the seed cake remaining from oil
extraction while the oil is clarified and centrifuged. The
crude oil obtained by solvent extraction contains impurities.
Some of them, such as seed fragments and meal fines, are
oil insoluble and thus can be readily removed by filtration.
Others, including free fatty acids, hydrocarbons, ketones,
tocopherols, glycolipids, phytosterols, phospholipids,
proteins, pigments and resins, are soluble or form stable
colloidal suspensions in the oil. Most of these have to be
removed by chemical or physical refining processes. In this
study both types of refining were considered. The main
difference is related to the degumming stage, that allows to
remove phospholipids. In the physical refining, degumming
is carried out only with the addition of water to the oil. After
a certain reaction period, the hydrated phospholipids are
separated by means of centrifuges. In the chemical refining,
phospholipids become flocculants by adding hot water
acidulated with phosphoric acid. The next step of the

28

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Fig. 1 e System boundary of the biodiesel supply chains.

chemical refining is represented by neutralization, in order

to avoid that organic acidity of the oil increases its instability; the neutralization is carried out by hot aqueous soda,
while soap is separated by centrifugation. The neutralized
oil is washed with hot water to remove residual soap.
Finally, a vacuum dryer is used after washing to remove the
residual moisture. Then, in both cases, the pigments
responsible for dark color of the oil are removed processing
the oil by Fullers earth turned to 90e100  C under agitation.
The final step of refining is the deodorization. Its purpose is
to eliminate fatty acids and odoriferous substances (aldehydes and ketones) by steam and vacuum at high temperature [8]. Finally, refined oil is transesterified with the help
of a catalyst, sodium hydroxide (NaOH), and methanol to
produce biodiesel. The outputs from the process include
biodiesel and glycerol. Processing step comprises also the
purification process of crude glycerol. In fact glycerol
contains unused catalysts and soaps, that are neutralized
with an acid. Then water and alcohol are removed to
produce 80e88% pure glycerol and then it is distilled to 99%
or higher purity.
In the following paragraphs the methodologies to calculate
GHG emissions of the chains (RED methodology), N2O emissions from nitrogen fertilizers application (IPCC methodology)
and the amount of nitrogen fertilizer input for sunflower and
rapeseed (Umbrian Integrated Production Guidelines) are
described.

3.

Methodology

3.1.

RED methodology

The formula for calculating the actual values for the total
emissions from the use of biofuel is given in the part C of Annex
V of the RED [2]. It takes into account all the different phases of
biofuel production. GHG emissions are expressed in terms of
grams of equivalent CO2 per MJ of fuel. The GHGs taken into
account are carbon dioxide (CO2), methane (CH4) and nitrous
oxide (N2O), whose effects on global warming are reported as
the amount (mass unit) of CO2 equivalent (CO2eq) through the
following coefficients: 1 for CO2, 23 for CH4 and 296 for N2O.
Relative reduction in the GHG emissions achievable by
replacing fossil fuel comparator by certain biofuel is defined as:
Emission Saving EF  EB =EF

(1)

where
EB total emissions from the use of the biofuel;
EF total emissions from the fossil fuel comparator
(83.8 g MJ1).
In this work any land use change was assumed
because cultivation sites were considered already croplands.
Emission saving from soil carbon accumulation via improved

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

agricultural management (called esca in the RED) does not take

place, because standard agricultural techniques were assumed.
Carbon capture, sequestration or replacement were not
considered and therefore the terms eccs or eccr in the RED were
assumed to be zero. Finally, in the reference scenario the
emission saving from cogeneration (called eee in the RED) was
assumed to be zero.
Biodiesel production chain produces more than one output
(meal and glycerol), therefore decisions are required about
how to allocate the environmental burdens of each one. The
RED (paragraphs 17 and 18 of part C, Annex V) states that the
allocation of emissions between the products inside the
system boundary shall be carried out in proportion to the
energy content of the products (determined by lower heating
value in the case of co-products other than electricity). Allocation procedure is required for all the emissions that take
place before the process step, where the co-product is obtained. Consequently, all the emissions, except those from
biodiesel distribution, shall be divided between the main
product and the co-products.
The same emission factors used for the calculation of the
typical values reported in the RED were applied in order to
adopt the same assumptions.

3.2.

IPCC methodology

GHG emissions from nitrogen fertilizers application represent

one of the main sources of the overall GHG emissions in the
analyzed chains [9,10]. In Italy IPCC methodology [11] is recommended by the Decree [12] that implements RED for their
calculation. Tier 1 approach was used; it allows to determine
direct and indirect N2O emissions and CO2 from urea distribution. Direct N2O emissions are caused by the use of nitrogen
fertilizers and the degradation of crop residues. The emission
factor was assumed equal to 10 g kg1 of N applied. In
particular, the amount of above and below-ground residues
and their nitrogen content were calculated with the methodology reported in chapter 11 (Eq. (11.6)-Tier 1). For their
calculation the following parameters were assumed:
e RAG ratio of above-ground residues dry matter to dry
matter yield (sunflower: 2.23 [13], rapeseed: 2.23 [13],
soybean: 1.39 [11]);
e NAG N content of above-ground residues dry matter
(sunflower: 9.77 g kg1 of N [13], rapeseed: 5.6 g kg1 of N
[13], soybean: 8 g kg1 of N [11]);
e RBG ratio of below-ground residues dry matter to dry
matter yield (sunflower: 0.1 [13], rapeseed: 0.08 [13],
soybean: 0.19 [11]);
e NBG N content of below-ground residues dry matter
(sunflower: 0.48 g kg1 of N [13], rapeseed: 5.6 g kg1 of N
[13], soybean: 8 g kg1 of N [11]).
Indirect N2O emissions were also taken into account; IPCC
Guidelines detailed the calculation of the two indirect N2O
emissions sources that occur during the agricultural step: N2O
due to leaching and runoff (emission factor: 7.5 g kg1 of N
leached and runoff) and N2O due to NH3 and NOx volatilization
(emission factor: 10 g kg1 of NH3 and NOx volatilized). The
following parameters were considered:

29

e FracGASF synthetic N fertilizer fraction that volatilizes as

NH3 and NOx e the default is 100 g kg1 of N applied [6];
e FracGASM organic N fertilizer fraction that volatilizes as
NH3 and NOx - the default is 200 g kg1 of N applied or
deposited [11]
e FracLEACH N applied fraction lost through leaching and
runoff - the default is 300 g kg1 of N applied [11];
The evaluation of the CO2 emissions from urea application
was carried out adopting the Eq. (11.13) of the IPCC Guidelines
(emission factor: 200 g kg1 of urea).

3.3.

Umbrian Integrated Production Guidelines

For sunflower and rapeseed crops, the yearly mass of nitrogen

fertilizer inputs (NF) were determined applying the Umbrian
Integrated Production Guidelines [14], which define the
optimal dose on the basis of site-specific parameters (Eq. (2)):
NF A  B C D  E  F  G

(2)

where:
A yearly N requirement by crop (sunflower: 107.3 kg ha1,
rapeseed: 170.9 kg ha1), calculated as crop N absorption per
unit of product (sunflower: 4.24% [14], rapeseed: 6.01% [14])
multiplied by the awaited yield (sunflower: 2530 kg ha1 [7],
rapeseed: 2820 kg ha1 [7]);
B yearly N availability from soil (sunflower: 13.6 kg ha1,
rapeseed: 12.06 kg ha1), calculated as sum between promptly
available nitrogen (b1) and nitrogen from organic matter
mineralization (b2). In the description of the C parameter, it is
demonstrated that b1 is equal to zero because promptly
available nitrogen is lost by leaching, while b2 is determined
by the Eq. (3):
b2 NM  T

(3)

where:
NM yearly mineralized N availability (18 kg ha1 [14]);
T fraction of a year during which crop is cultivated
(sunflower: 0.75 [14], rapeseed: 0.67 [14]);
C yearly N leaching loss, calculated from the precipitation
mass from 1 October to 31 January [14]; only nitrogen promptly
available is considered leachable and it is quantified in this
way:
e precipitation mass < 150 mm: no loss;
e 150 < precipitation mass < 250: mass fraction percentage
loss of nitrogen promptly available (150 < precipitation
mass < 250)150;
e precipitation mass > 250 mm: all nitrogen promptly
available is lost.
The precipitation mass was 350.4 mm [15], obtained from
data, registered by the meteorological station located in Marsciano (Perugia) in the period 1 October 2009e31 January 2010;
therefore b1 was assumed equal to zero;
D: N loss due to immobilization, volatilization and denitrification processes, calculated according to Eq. (4)

30

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

D b1 b2  fc

(4)

where fc is a correction factor which depends on soil texture

and soil drainage; fc was assumed equal to 0.3 [14] for
medium-texture clay soil with normal drainage.
E N available from previous crop residues; if residues are
characterized by a high C/N ratio, a temporary nitrogen
reduction occurs. It was assumed wheat as previous crop and
straw manuring, therefore E was equal to e 30 kg ha1 [14].
F N available from organic fertilization during the previous
year; it was assumed zero in both cases.
G yearly N supplied by precipitations:
G HP  NHP

(5)

where:
HP precipitation mass from sowing to harvesting (sunflower:
271.1 mm AprileAugust [15], rapeseed: 680 mm SeptembereJune [15]);
NHP nitrogen content of precipitation (2 mg l1 [16]).

Table 1 e Agricultural operations for sunflower, rapeseed

and soybean cultivation: diesel consumptions.
Diesel (L ha1)

Agricultural operation

Sunflower
Slight ploughing
Routing
Deep ploughing
Surface harrowing
Harrowing
Pre-sowing weeding
Pre-sowing fertilization
Sowing
Pre-.emergence weeding
Post-emergence weeding
Top-dressing fertilization
Harvesting

e
e
96.15
5.34
e
2.12
6.36
4.59
2.12
e
e
40.04

Rapeseed
31.38
16.96
e
5.34
e
2.12
6.36
4.59
2.12
e
6.36
40.04

Soybean
31.38
e
e
e
13.81
e
6.36
4.59
e
2.12
e
40.04

4.

Inventory analysis

4.1.

Cultivation step

2 kg ha1 of Metazachlor at pre-emergence [18]. Sowing is

planned in the mid September with 7 kg ha1 of seeds [6]. Yield
(2820 kg ha1) was estimated considering the overall rapeseed
production in Umbria during 2009 (282,000 kg) and the relative
cultivated area (100 ha) [7].
Table 1 shows the agricultural operations considered and
the related diesel consumption such as in [19].
As for sunflower, seeds drying was considered from 10% of
moisture to 9%.

4.1.1.

Sunflower

4.1.3.

According to Umbrian Integrated Production Guidelines, the

nitrogen input calculated was 122.4 kg ha1 whereof 40%
applied in pre-sowing fertilization as diammonium phosphate
(311 kg ha1, 18% of N and 46% of P2O5) and 60% distributed in
top-dressing fertilization as urea (161.5 kg ha1, 46% of N) [17].
Two herbicide treatments were also considered: the first one
in pre-sowing time, using 1 kg ha1 of Trifluralin, and the
other one in pre-emergence period, applying 2 kg ha1 of
Oxadiazon [18].
Seeds yield (2530 kg ha1) was estimated from the overall
sunflower seeds production in Umbria during 2009
(99,667,100 kg) and the relative cultivated area (39,366 ha) [7].
Table 1 shows the agricultural operations and the related
diesel consumption such as in [19].
According to RED, seeds drying was included in the cultivation step assuming that moisture content was reduced from
10% to 9% to guarantee the preservation. The energetic
consumptions of this phase were drawn from [19]:
e diesel: 5 MJ kg1 of evaporated water;
e electricity: 3.6 MJ kg1 of evaporated water.

4.1.2.

Rapeseed

Nitrogen input was estimated in 158.8 kg ha1 of N divided in

40% at pre-sowing fertilization and 60% at top-dressing
fertilization. Fertilizers employed in Umbria are represented
by ternary fertilizer (8% N, 24% P2O5, 24% K2O) at pre-sowing
(790 kg ha1) and urea at top-dressing (157.5 kg ha1), satisfying the nutritional needs of the crop in terms of potash and
phosphorus [6]. As regards weeding treatment, two operations
were assumed: 1.5 kg ha1 of Trifluralin at pre-sowing and

Soybean

In Table 1 typical agricultural operations in the Veneto region

are shown [17,20e22]. Irrigation was not assumed because in
Italy only 36.3% of cultivated area is irrigated [23]. Soybean
requires moderate phosphorus and potassium fertilizer input
(70 kg ha1 of P2O5, 100 kg ha1 of K2O), while nitrogen fertilization was not considered because it can inhibit the biological fixation [17].
Only one herbicide treatment was considered, with
1 kg ha1 of imazamox, 5 g ha1 of thifensulfuron-methyl and
2 kg ha1 of fluazifop-p-butyl at post-emergence [21]. The yield
was assumed equal to 3470 kg ha1 of moist seeds (moisture
content: 15%); it was chosen to take into account the Veneto
statistical data for soybean cultivation in 2009 [7]. For soybean,
seeds drying was considered from 15% of moisture to 14% [24].
In Table 2, the GHG emissions of the cultivation step for
sunflower, rapeseed and soybean crops are reported, together
with the emission factors used to count the GHG emissions of
the materials life cycle production. In order to carry out
a meaningful comparison with the default values reported in
the RED, the same emission factors were assumed. GHG
emissions are obtained assuming 26.4 MJ kg1 (sunflower and
rapeseed) and 23.6 MJ kg1 (soybean) of dry matter as Lower
Heating Value of the seeds [25].

4.2.

Processing step

Extraction phase is different for the analyzed chains in terms

of energy inputs and oil yields, while the other processes are
the same. In the extraction process, steam produced by
a natural gas boiler with efficiency of 90% is used. Seeds
characteristics taken into account are the following:

31

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Table 2 e Emissions (unit: g MJL1 of seed) of the cultivation and drying process phases [25].
Phase

Input

Cultivation

Drying

Diesel
N fertilizer
K2O fertilizer
P2O5 fertilizer
Pesticides
Seeds
N2O emissions
CO2 emissions
Total Cultivation
Diesel
Electricity
Total Drying

Emission factor
1

87.5 g MJ
6065.3 g kg1
583.2 g kg1
1017.8 g kg1
17257.6 g kg1
200 g kg1

87.5 g MJ1
127.6 g MJ1

e moisture: 9% rapeseed and sunflower, 14% soybean;

e impurities: 1.5% rapeseed and sunflower [data from
Italian plants], 0.9% soybean [26];
e oil: 45% sunflower [data from Italian plants], 40% rapeseed
[data from Italian plants], 19% soybean [26];
e dry matter: 44.5% sunflower, 49.5% rapeseed, 66%
soybean.
The following properties of crude oil were taken into
account for the mass balance of the refining process:
e gums: 0.5%;
e impurities: 0.5%;
e free fatty acids: 1%.
Processing step comprises also the purification process of
crude glycerol; it requires steam and the related natural gas
consumption amounts to 13.66 MJ kg1 of glycerol [25]. GHG
emissions of glycerol purification amounts to 2.59 g MJ1 of
biodiesel.
Data about energy and mass flows of the processing step
are reported in Table 3. GHG emissions are reported in Table 4
with the emission factors used to count the GHG emissions of
the materials life cycle production. GHG emissions are
expressed on the basis of the output of each phase and the
following Lower Heating Values were assumed [25]:
e crude oil: 36 MJ kg1;
e refined oil: 36 MJ kg1;
e biodiesel: 37.2 MJ kg1.

4.3.

Transport step

Transport step comprises two phases: biomass transport from

the field to the processing plant and biodiesel transport from
the processing plant to the filling station. The same hypotheses of the RED [25] were assumed because considered
acceptable for the Italian chains. Biomass transport was
considered long 50 km with a 40 Mg truck, characterized by an
actual payload of 22 Mg. Biodiesel transport was divided into
two parts: transport from processing plant to depot, where
biodiesel is blended with diesel, and from depot to the filling
station. Each transport develops long 150 km with a 40 Mg

Sunflower

Rapeseed

Soybean

8.51
12.35
e
2.12
0.43
0.02
8.20
1.97
33.60
0.18
0.19
0.37

5.40
14.37
1.66
2.89
0.40
0.02
11.29
2.24
38.28
0.16
0.19
0.35

4.44
e
0.84
1.03
0.75
e
3.57
e
10.63
0.18
0.23
0.41

truck and electricity consumption was considered for fuel

handling both at depot and at filling station. In Table 5 are
shown energy consumptions and GHG emissions.

4.4.

Coproducts allocation

According to the RED, allocation by energy content was the

procedure chosen for the coproducts accounting; the method
is based on the evaluation of the lower heating values of the
main product (biodiesel) and its coproducts (meal and glycerol); the following values were assumed [25]:
e biodiesel: 37.2 MJ kg1;
e glycerol: 16.6 MJ kg1;
e meal: 15.5 MJ kg1.
Then, the energy amounts per one kg of biodiesel were
evaluated considering the coproducts yields:
e
e
e
e
e

biodiesel: 37.2 MJ kg1 of biodiesel (Eb);

glycerol: 1.7 MJ kg1 of biodiesel (Eg);
sunflower meal: 18.4 MJ kg1 of biodiesel (Efsu);
rapeseed meal: 22.9 MJ kg1 of biodiesel (Efr);
soybean meal: 65.4 MJ kg1 of biodiesel (Efso)

In this way, the percentage (sunflower; Ps 65%; rapeseed:

Pr 60%; soybean: Pso 36%) of GHG emissions of the cultivation, seeds drying, biomass transport and extraction
processes assigned to biodiesel was calculated with the
following [Eqs. (6), (7) and (9)]:



Ps Eb = Eb Eg Efsu  100

(6)




Pr Eb = Eb Eg Efr  100

(7)




Pso Eb = Eb Eg Efso  100

(8)

Whereas, the percentage (Pt) of GHG emissions of the

refining and transesterification processes allocated to biodiesel (96% for all chains) was determined with the Eq. (9):



Pt Eb = Eb Eg  100

(9)

For the other processes (glycerol purification and biodiesel

transport), all GHG emissions were assigned to biodiesel.

32

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Table 3 e Mass and energy flows of the processing phase.

Step
Extraction

Input

Electricity
Natural gas
Hexane
Meal

Output

Unit

Sunflower

Rapeseed

Soybean

kJ L of crude oil
kJ L1 of crude oil
g L1 of crude oil
g kg1 of seeds

478.8
1598.4
1.3
517

554.4
1897.2
1.6
574

1288.8
4784.4
5
756

1

Refining
Degumming and neutralization

Input

Bleaching

Input

Deodorization

Input

Soapstock treatment

Input

Transesterification

Input

Electricity
Natural gas
NaOH
Phosphoric acid
Electricity
Natural gas
Fullers earth
Electricity
Natural gas
Electricity
Natural gas
Electricity
Natural gas
NaOH
HCl
Methanol
Citric Acid
Glycerol

Output

5.

Case studies

In addition to the reference scenario, three case studies were

analyzed in order to verify if higher GHG savings could be
obtained. The first two case studies have been considered
since they provide alternative farming practices in Umbria
while the third is an interesting opportunity to improve the
energy efficiency of biodiesel chains.
The following assumptions for each case study are reported.
1. the term F in the Eq. (2) was considered different from 0 for
rapeseed. It was hypothesized that an organic fertilization
(150 kg ha1) was carried out during the cultivation of the
previous crop (wheat). According to [14], only a fraction of

Chemical refining

Physical refining

23.9
16
6
1
11.7
167.4
9.2
11.7
435.6
0.14
1.98
31.68
662.4
0.9
8.8
95.7
0.6
91.7

16.5
157.7
e
e
11.6
166.3
9.2
11.6
496.8
e
e
31.68
662.4
0.9
8.8
95.7
0.6
91.7

1

kJ L of refined oil
kJ L1 of refined oil
g L1 of refined oil
g L1 of refined oil
kJ L1 of refined oil
kJ L1 of refined oil
g L1 of refined oil
kJ L1 of refined oil
kJ L1 of refined oil
kJ L1 of refined oil
kJ L1 of refined oil
kJ L1 of biodiesel
kJ L1 of biodiesel
g L1 of biodiesel
g L1 of biodiesel
g L1 of biodiesel
g L1 of biodiesel
g L1 of biodiesel

organic nitrogen is really available for rapeseed due to the

mineralization of organic fertilizers residues; this fraction
amounts to 20% and the term F was equal to 30 kg ha1 of N.
Therefore the amount of synthetic nitrogen fertilizers was
128.8 kg ha1 of N, distributed as ternary fertilizer at presowing (644 kg ha1) and as urea at top-dressing fertilization (168 kg ha1);
2. sunflower yield of 3180 kg ha1 was assumed, drawn from
a study in the Perugia Province [5] related to a wheatsunflower cropping system taken off the straw from the
field; in this case the term E in the Eq. (2) was equal to
10 kg ha1 of N [14]. The amount of synthetic nitrogen
fertilizers was 130 kg ha1 of N distributed as diammonium
phosphate at pre-sowing (288 kg ha1) and as urea at topdressing fertilization (170 kg ha1);

Table 4 e GHG emissions of the processing step.

Input

Electricity
Natural gas
n-Hexane
Fullers earth
Phosphoric Acid
NaOH 50%
Sulfuric Acid
HCl 36%
Methanol
Citric Acid
Total

Emission
factors

1

127.6 g MJ [25]
67.7 g MJ1 [25]
3933.4 g kg1 [25]
436 g kg1 [27]
1435.3 g kg1 [28]
933 g kg1 [28]
207.9 g kg1 [25]
431.6 g kg1 [29]
1981.4 g kg1 [25]
963.1 g kg1 [28]

Refining
(g MJ1 of refined oil)

Extraction
(g MJ1 of crude oil)
Sunflower

Rapeseed

Soybean

Chemical

Physical

1.89
3.33
0.16
e
e
e
e
e
e
e
5.38

2.18
3.97
0.19
e
e
e
e
e
e
e
6.34

5.00
9.84
0.54
e
e
e
e
e
e
e
15.37

0.18
1.60
e
0.12
0.04
0.90
0.005
e
e
e
2.40

0.16
1.72
e
0.12
e
e
e
e
e
e
2.00

Transesterification
(g MJ1 of biodiesel)

0.12
1.37
e
e
e
0.03
e
0.12
5.80
0.02
7.46

33

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Table 5 e Energy consumptions and GHG emissions of the transport step.

Biomass transport
Sunflower
Diesel
Electricity
GHG emissions

2.1 J MJ

Biodiesel transport

Rapeseed

1

of seed
e
0.17 g MJ1 of seed

2.1 J MJ

Soybean

1

2.5 J MJ1 of seed

e
0.20 g MJ1 of seed

of seed
e
0.17 g MJ1 of seed

3. Combined Heat and Power (CHP) plant at the service of

refining, transesterification and glycerol purification
processes was considered; oil extraction phase was
excluded because, in Italy, it is generally performed in
a separate plant. Natural gas CHP plant was assumed with
electric yield of 35% (he) and thermal yield of 45% (ht). The
term eee in the RED methodology takes into account emission saving from excess electricity from a CHP plant.
According to RED, in accounting for that excess electricity,
the size of the cogeneration unit shall be assumed to be the
minimum necessary to supply the heat needed to produce
the fuel. The GHG saving associated with that excess electricity shall be taken to be equal to the amount of greenhouse gas that would be emitted when an equal amount of
electricity was generated in a power plant using the same
fuel as the cogeneration unit. In this way, heat amounts for
each processing phase were determined:
 refining: 21 kJ MJ1 of biodiesel;
 transesterification: 18 kJ MJ1 of biodiesel;
 glycerol purification: 38 kJ MJ1 of biodiesel.
Therefore the total heat demand was equal to 77 kJ MJ1 of
biodiesel (ET). Electricity (EC) required by the three phases was
equal to 2.3 kJ MJ1 of biodiesel (refining: 1.4 kJ MJ1 of biodiesel, transesterification: 0.96 kJ MJ1 of biodiesel). Electricity
deliverable by the CHP plant (ECG) was obtained with the Eq.
(10):
ECG ET  he =ht 59:9 kJ MJ1 of biodiesel

(10)

Excess electricity (ECC) was obtained with the Eq. (11):

ECC ECG  EC 57:6 kJ MJ1 of biodiesel

(11)

Considering that the production of 1 MJ of electricity

requires 2.857 MJ of natural gas (electric yield: 35%), the

8 J MJ1 of biodiesel
1.2 J MJ1 of biodiesel
1.26 g MJ1 of biodiesel

amount of natural gas related to excess electricity was equal

to 164.6 kJ MJ1 of biodiesel, equivalent to a carbon credit of
11.1 g MJ1 of biodiesel. This value has to be subtracted to the
overall GHG emissions of the chain.

6.

Results

Table 6 shows the allocated GHG emissions of each process

referred to MJ of biodiesel, which was chosen as the functional
unit. GHG emissions of each process were initially referred to
MJ of its output (e.g. MJ of seed for cultivation process);
therefore, in order to refer GHG emissions to MJ of biodiesel,
they were multiplied for the yields of the subsequent
processes.
In Table 7 GHG emissions splitted among the three main
steps of the chains are reported; values, rounded off to the
unit, are compared to those defined in the RED. Moreover, GHG
savings are shown calculated applying Eq. (1).
Results showed that GHG emissions were higher than
those reported in the RED for sunflower and rapeseed (49-49
gCO2eq MJ1 (for chemical and physical refining respectively)
for sunflower, 56-55 gCO2eq MJ1 for rapeseed) and lower for
soybean (31-30 gCO2eq MJ1).
The step characterized by the highest environmental
impact was cultivation for sunflower and rapeseed chains,
with a burden of 67% and 69% respectively, and processing
(57%) for soybean.
In the cultivation step, the major contributions to the GHG
emissions are represented by nitrogen fertilizers production,
diesel consumption and N2O emissions from nitrogen fertilizers application, which contribute to the environmental
impact of the step for 37%, 25% and 24% respectively for
sunflower, for 38%, 15% and 30% for rapeseed. GHG emissions
of soybean cultivation are mostly due to diesel consumption

Table 6 e Calculated emissions for the different processes of the chains (g MJL1 of biodiesel).
Process

Cultivation
Seeds drying
Biomass transport
Oil extraction
Oil refining
Transesterification
Glycerol purification
Biodiesel transport
Total

Sunflower

Rapeseed

Soybean

Chemical
refining

Physical
refining

Chemical
refining

Physical
refining

Chemical
refining

Physical
refining

32.0
0.4
0.2
3.5
2.1
7.2
2.6
1.3
49.3

31.7
0.3
0.2
3.4
1.8
7.2
2.6
1.3
48.5

38.1
0.4
0.2
3.8
2.1
7.2
2.6
1.3
55.7

37.8
0.3
0.2
3.8
1.8
7.2
2.6
1.3
55.0

11.5
0.3
0.2
5.4
2.1
7.2
2.6
1.3
30.6

11.4
0.3
0.2
5.3
1.8
7.2
2.6
1.3
30.1

34

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

Table 7 e Total emissions (g MJL1 of biodiesel) and GHG savings (%) for the different steps of the chains: comparison
between calculated and RED typical and default data.

Sunflower

Rapeseed

Soybean

Chemical refining
Physical refining
RED typical value
D (%)
RED default value
D (%)
Chemical refining
Physical refining
RED typical value
D (%)
RED default value
D (%)
Chemical refining
Physical refining
RED typical value
D (%)
RED default value
D (%)

Cultivation

Processing

Transport

Total emissions

GHG saving

32
32
18
44
18
44
38
38
29
24
29
24
12
12
19
37
19
37

15
15
16
7
22
32
16
15
16
0/7
22
27/32
17
17
18
6
26
35

1
1
1
0
1
0
1
1
1
0
1
0
1
1
13
92
13
92

49
49
35
29
41
16
56
55
46
18/16
52
7/5
31
30
50
38/40
58
47/48

41
42
58
17/16
51
20/18
34
34
45
11
38
4
63
64
40
23/24
31
32/33

(42%) and N2O emissions from decomposition of crop

residues (34%).
As shown in Table 7, GHG emissions of cultivation step are
quite different from values reported in the RED, mostly due to
the fact that RED considers different values of seeds yield
(sunflower: 2440 kg ha1, rapeseed: 3113 kg ha1, soybean:
2798 kg ha1), diesel consumption (sunflower: 82.6 L ha1,
rapeseed: 82.6 L ha1, soybean 58.6 L ha1), nitrogen fertilizers
application (sunflower: 39 kg ha1, rapeseed: 137.4 kg ha1,
soybean 8.0 kg ha1) and N2O emissions (sunflower:
2430 g ha1, rapeseed: 3010 g ha1, soybean 2238 g ha1). Data
in the present study were obtained by using Tier 1, while the
RED applied Tier 3 (using the DNDC model) of IPCC methodology. In particular, for soybean chain RED data are referred to
cultivation in the US assuming 74 g kg1 of N as nitrogen
content of below-ground residues dry matter instead of
8 g kg1 of N as reported in the IPCC Guidelines.
As regards the processing step, the most impacting process
is transesterification (47% for sunflower, 46% for rapeseed and
41% for soybean), due above all to the methanol production
(78%) for all the chains. The difference between chemical and
physical refining is very small (about 0.3 gCO2eq MJ1 of biodiesel), mostly due to the NaOH consumption in the chemical
process. Overall GHG emissions of the processing step are
slightly lower (maximum 7%) than typical values reported in
the RED. However there are some differences by analyzing
single processes; in fact the emissions of the oil extraction and
refining are higher than RED values for sunflower and rapeseed chains (extraction: 2.52 g MJ1 of biodiesel for sunflower
and 2.59 g MJ1 of biodiesel for rapeseed; refining: 0.71 g MJ1
of biodiesel for both crops) while only the refining process has
a noticeable variation for soybean chain. The main differences
concern the higher consumption of electricity and natural gas,
mostly for the oil refining step (electricity: 19.4 kJ L1 of refined
oil; natural gas: 296 kJ L1 of refined oil). Instead, the transesterification process has lower emissions than those of the
RED, mainly due to lower consumption of natural gas
(1528 kJ L1 of biodiesel) and HCl (17.8 g L1 of biodiesel).

GHG emissions of transport step are almost entirely

allowable to biodiesel transport from depot to filling station
(about 87% of the transport process). The main difference with
the RED concerns the transport of soybeans because in the
Directive transport of soybeans via truck over a distance of
700 km and maritime transport via ship over a distance of
10 186 km were considered.
GHG savings results showed that the current target of RED
(35%) was reached by sunflower (41e42%) and soybean
(63e64%) chains, but not by rapeseed chain (34%). These
percentages may be significantly increased by assuming the
hypotheses of the case studies (Fig. 2), reaching the target
value of 50% in 2017 also for biodiesel from sunflower. The
first two case studies reveal that small changes to the reference cultivation technique may help to improve the sustainability of the supply chain. In the first case study, GHG
emissions of the sunflower cultivation were 35 g MJ1 of biodiesel, obtaining, for the entire chain, 52 g MJ1 of biodiesel
and a GHG saving of 38%, that it makes sustainable the supply
chain. The second case study allows to obtain 26 g MJ1 of
biodiesel for the rapeseed cultivation and 43 g MJ1 of biodiesel for the entire chain, with a GHG saving of 49% for

Fig. 2 e Case studies: variation of the GHG savings (%).

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

chemical refining and 50% for physical refining. However the

best result can be obtained through the use of a CHP plant,
reaching a GHG saving of 55%, 47%, 77%, respectively for
sunflower, rapeseed and soybean.

7.

Conclusions

The aim of the present paper was to calculate the typical GHG
emissions of biodiesel from sunflower, rapeseed and soybean
produced in Italy, according to the calculation rules defined in
the RED. The analysis was based on agricultural data related to
the Umbria region for sunflower and rapeseed and to the
Veneto region for soybean. Standard cultivation techniques
for the considered regions were adopted. Data about the
processing step were drawn from the main Italian biodiesel
plants. The assessment of GHG emissions in the agricultural
phase shows that a value less than the typical reported in the
Directive could be obtained. Therefore, as determined by the
RED, if Italy communicated this value to the European
Commission it may be taken as the default value for the
Veneto region, simplifying the certification process of the
biodiesel chain. Values much higher than those reported in
the RED were achieved for sunflower and rapeseed; this fact
will oblige the economic operators to adopt necessarily the
actual values. GHG emissions related to the biodiesel
production step agree with the typical values given in the RED,
showing that for this phase there are no substantial differences between the Italian and the European scenario.
However, even for this step, the economic operators will be
taken to calculate the actual values since the default values
are increased by 40% compared to the typical ones. The last
consideration is that the transport phase of the soybean, for
which the actual value must be calculated as the default
value, takes into account the importation from the United
States.
In synthesis, using field data for Italian conditions we find
the RED default values for total biodiesel production GHG
emissions to be an underestimate for sunflower and rapeseed,
while for soya the RED values strongly overestimate the GHG
emissions. By moving to natural gas fired combined heat and
power plants for the processing step, even higher reductions
in GHG emissions are possible.

references

[1] Gnansounou E, Dauriat A, Villegas J, Panichelli L. Life cycle

assessment of biofuels: energy and greenhouse gas balances.
Bioresour Technol 2009;100(21):4919e30.
[2] Directive 2009/28/EC. On the promotion of the use of energy
from renewable sources and amending and subsequently
repealing Directives 2001/77/EC and 2003/30/
EC.2009.04.23.Off J EU 2009. L 140:16e62.
[3] Directive 2003/30/EC. On the promotion of the use of biofuels
or other renewable fuels for transport.2003.05.17. Off J EU
2003. L 123:42e6.
[4] Renewable Energy [Internet]. Brussels: European
Commission. c2012-[accessed 01.10.2012]. Available at:
http://ec.europa.eu/energy/renewables/biofuels/emissions_
en.htm.

35

[5] Monotti M, editor. Manuale di corretta prassi per la

produzione integrata di girasole [Internet]. 3A e Parco
tecnologico agroalimentare dellUmbria [accessed
20.09.2010]. Available at: http://www.parco3a.org; 2001 [in
Italian].
[6] Ciriciofolo E, editor. Manuale di corretta prassi per la
produzione integrata di colza [Internet]. 3A e Parco tecnologico
agroalimentare dellUmbria [accessed 20.09.2010]. Available at:
http://www.parco3a.org; 2001 [in Italian].
[7] ISTAT [Internet]. Rome: Agricoltura e zootecnia. c2009[accessed 03.09.2010]. Available at: http://agri.istat.it/sag_is_
pdwout/jsp/Introduzione.jsp [in Italian].
[8] Schmidt JH, editor. Life cycle assessment of rapeseed oil and
palm oil, PhD thesis, Department of development and
planning, Aalborg University; 2007.
[9] Mosier AR, Crutzen PJ, Smith KA, Winiwarter W. Nitrous
oxides impact on net greenhouse gas savings from biofuels:
life-cycle analysis comparison. Int J Biotechnol 2009;11:60e74.
[10] Smeets EMW, Bouwman LF, Stehfest E, Van Vuuren DP,
Posthuma A. Contribution of N2O to the greenhouse gas balance
of first-generation biofuels. Glob Change Biol 2009;15:1e23.
[11] Intergovernmental Panel on Climate Change (IPCC). Guidelines
for national greenhouse gas inventories. In: Agriculture,
forestry and other land use, vol. 4. Japan: IGES; 2006.
[12] Decreto del Ministero dellAmbiente e della Tutela del
Territorio e del Mare. Sistema nazionale di certificazione per
biocarburanti e bioliquidi; 23 Gennaio 2012. Gazzetta
Ufficiale della Repubblica Italiana n. 31 del 07/02/2012, Italy
[in Italian].
[13] University of New Hampshire. Users guide for the DNDC
model. Available at: http://www.dndc.sr.unh.edu/?page_
id4; 2007 [accessed 19.07.2010].
[14] Regione Umbria. Disciplinari di produzione integrata,
Allegato A: Indicazioni e norme generali. Available at: http://
www.agriforeste.regione.umbria.it/canale.asp?id962; 2010
[in Italian].
[15] CRA [Internet]. Rome: Banca Dati Agrometeorologica
Nazionale. c2010-[accessed 02.07.2010]. Available at: http://
old.politicheagricole.it/ucea/forniture/index3.htm [in
Italian].
[16] Masoni A, Mariotti M, Lulli L, editors. Supporto tecnico per la
gestione dei piani di concimazione delle colture erbacee
[Internet]. Universita` di Pisa [accessed 13.07.2010]. Available
at: http://risorseidriche.arsia.toscana.it/UserFiles/File/idrico/
Divulgazione%202009/SupportoTecnicoErbacee.pdf; 2007 [in
Italian].
[17] Benvenuti A, Vannozzi GP. Girasole (Helianthus annuus L.).
In: Baldoni R, Giardini L, editors. Coltivazioni erbacee. Piante
oleifere, da zucchero, da fibra, orticole e aromatiche,
Bologna. Pa`tron; 2001. p. 59e95 [in Italian].
[18] Bonciarelli F. In: Coltivazioni erbacee da pieno campo.
Calderini Edagricole; 2001 [in Italian].
[19] Nemecek T, Heil A, Huguenin O, Meier S, Erzinger S, Blaser S,
et al. Life cycle inventories of agricultural production
systems. Dubendorf: Swiss Centre for Life Cycle Inventories;
2003. Report No.: 15.
[20] Meriggi P, Bartolini D. Avvicendamento, lavorazione del
terreno, concimazione e semina. In: Regione Emilia
Romagna. Agricoltura - anno 33, vol. 2; 2005. p. 86 [in Italian].
[21] Meriggi P, Bartolini D. Scelta varietale e controllo delle
infestanti. In: Regione Emilia Romagna. Agricoltura - anno
33, vol. 2; 2005. p. 87e90 [in Italian].
[22] Meriggi P, Bartolini D. Irrigazione e raccolta. In: Regione
Emilia Romagna. Agricoltura - anno 33, vol. 2; 2005. p. 90 [in
Italian].
[23] Bonati G, Liberati G. Uso irriguo dellacqua e principali
implicazioni di natura ambientale. Istituto Nazionale
Economia Agraria; 2007 [in Italian].

36

b i o m a s s a n d b i o e n e r g y 4 7 ( 2 0 1 2 ) 2 6 e3 6

[24] Centro Ricerche Produzioni Vegetali. In: Disciplinare di

Produzione Integrata. Girasole, soia, mais. Bologna: Regione
Emilia Romagna - Assessorato Agricoltura; 2005 [in Italian].
[25] Edwards R, Larive JF, Mahieu V, Rouveirolles P. Well-ToWheels analysis of future automotive fuels and powertrains
on the European context. Brussels: Joint Research Centre;
2007. Report No.: version 2c.
[26] AA.VV. Progetto per lo sviluppo della filiera dei
biocombustibili [Internet]. Azienda Regionale Veneto
Agricoltura [accessed 08.10.2010]. Available at: http://www.

venetoagricoltura.org/upload/pubblicazioni/
Biocombustibili_SC26/ azione_A.pdf; 2004 [in Italian].
[27] Kellenberger D, Althaus HJ, Kunniger T, Jungbluth N. Life
cycle inventories of building products. Ecoinvent report No.
7, EMPA Dubendorf; 2003.
[28] Althaus HJ, Chudacoff M, Hellweg S, Hischier R, Jungbluth N,
Osses M, et al. Life cycle inventories of chemicals. Ecoinvent
report No. 8, Dubendorf; 2003.
[29] Dubois JL. Selective oxidation of hydrocarbons and the global
warming problem. Catal Today 2005;99:5e14.