Biomass and Bioenergy 129 (2019) 105334

Contents lists available at ScienceDirect

Biomass and Bioenergy

journal homepage: www.elsevier.com/locate/biombioe

Research paper

Comparative energy and environmental analysis of agro-pellet production T

from orchard woody biomass
⁎
Sara Rajabi Hamedania, Andrea Colantonia, , Francesco Galluccib, Mariangela Salernob,
Cristian Silvestria, Mauro Villarinia
a
Department of Agriculture and Forest Science, Università Degli Studi Della Tuscia Viterbo, Italy
b
Consiglio per la Ricerca in Agricoltura e l'analisi dell'Economia Agraria (CREA), Italy

A R T I C LE I N FO A B S T R A C T

Keywords: Woody biomass has a considerable potential for energy purposes specially after densification processes which
LCA convert these materials into solid biofuels with higher energy density. This study focuses on the energy and
Pellet environmental assessment of pellet production system from vine and olive grove woody biomass. The life cycle
Energy return ratio assessment (LCA) from grape and olive cultivation up to packed pellet production, ready for delivery to final
Sensitivity analysis
users was conducted to quantify the eco-profile of agro-pellet production. The results of comparative analysis
Woody biomass valorization
revealed that pellet production from both biomasses is dominated by cultivation phase. Vine pellet production
was more dependent on chemical fertilizers specially potassium sulfate, but olive pellet production was more
dependent on copper oxide pesticide. Primary energy use per MJ energy of vine pellet was greater than that of
olive pellet (0.6 vs. 0.19). On the other hand, energy return ratio of olive pellet was three times of that of vine
pellet (5.22 vs. 1.7). Nevertheless, olive pellet had higher environmental impacts in the forms of human toxicity,
freshwater ecotoxicity and marine ecotoxicity. Consequently, sensitivity analysis investigates the impact of
variation in input parameters of fertilizers, pesticides, machinery, diesel fuel, transport and pelletizing process
on LCA results.

1. Introduction pruning residues represent a cost for farmers. The disposal of vineyard
residues, for example, usually consists in shredding, burying or burning
The European Union's Renewable Energy Directive (RED) 2009/28/ them.
EC commits the EU to increase the share of renewable energy con- The low bulk and energy densities of biomass persuade operators to
sumption in gross final energy consumption from 8.5% (in20041) to conduct densification processes (pelleting and briquetting) prior to its
20% by 2020 [1,2]. Following this directive, the total contribution of energy use [10]. This process transfers biomass into homogeneous and
biomass energy to the renewable energy resources' mix, is foreseen to automate solid biofuel that additionally has higher energy density.
represent 45.2% of renewable energy mix in 2020 [3]. This represents Agro-pellets are mainly used as fuels for heating and power generation
an increase from around 3002 PJ (PJ) in 2010 to about 4636 PJ in 2020 and has become very attractive in the European market in the mid-
[4]. 2000s [11].
Biomass has a fundamental role to develop renewable energy The largest national production markets of pellets are Sweden and
sources [5]. Nonetheless, there are a few environmental burdens of Germany followed by Italy. The utilization rates of pellet plants in these
first-generation biofuels (the final output of biomass), such as: land use three most important producers in 2008 were 64%, 56%, and 87%,
change and damage to biodiversity, impact on food availability and real respectively [12]. An important factor limiting the utilization of the
global warming potential [6–8]. The use of more sustainable biomass, pellet plants is the low availability of the feedstock to produce the
in particular agricultural residues such as pruning residues unravels pellets [13]. At the same time, despite the production capacity of pellet
these environmental burdens. In addition to the environmental benefits, plants, the level of production is not adequate to satisfy the domestic
the energetic use of agriculture residues, is also beneficial under the demand [14]. According to Ref. [15], Italy, for example, with pellet
farmers’ perspective [9]. In most situations, the proper management of production capacity of about 0.8 Mt in 2010 would not meet the

⁎
Corresponding author.
E-mail addresses: [email protected] (A. Colantoni), [email protected] (M. Villarini).
1
Source: Eurostat Database, Energy Statistics, https://ec.europa.eu/eurostat/data/database.

https://doi.org/10.1016/j.biombioe.2019.105334
Received 8 January 2019; Received in revised form 7 August 2019; Accepted 12 August 2019
0961-9534/ © 2019 Elsevier Ltd. All rights reserved.
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

internal demand. At the EU level, production is not keeping up with the assumption of the present study is that biomass residues are considered
faster growth rate of pellet consumption in recent years. Cocchi et al. as a coproduct as stated in our assumptions. Furthermore, this as-
[15] reported that the gap between the production and consumption of sumption has been also considered in previous works like the one of
pellets (in tons) had an 8-fold increase in the period 2008–2010. Boschiero et al. [19] and in several previous works by some authors
The growing demand for woody biomass to produce pellets [16] [22,29]; therefore, they have a contribution to all emissions and im-
and, on the other hand, their significant influence on environmental pacts corresponding to cultivation of the main crop. By the way, also
policies as renewable energy leads to a critical need to measure the Moreno and Dufour [30] used the same concept of economic allocation
environmental consequences connected to pellet production from to distribute environmental impact between main product and the other
woody biomass residues in orchards [17]. coproducts of the same unique cultivation process.
Boschiero et al. [18,19] investigated the use of woody residues from Regarding pellet production from orchard residues, some studies
apple orchards to generate heat and power in a gasification unit using considered the potential of these residues for pellet production [10] and
the life cycle assessment methodology. Picchi et al. [20] compared the their use in domestic boilers [31]. However, there is a lack of study on
combustion of vineyard pruning residues and wood chips in a domestic the environmental impact of pellet production from orchard residues. In
boiler to assess flue gases and heavy metal emissions. Torquati et al. addition, management of woody biomass supply chain to produce pellet
[21] estimated energy balance and CO2 emissions derived from energy has not been considered in previous studies and on this point an added
production chain from pruning residues of vineyards and olive groves in value has been generated by the present work.
the Italian Umbria region. Environmental profile of hydrogen produc- By the way, the objective of this LCA study is to quantify and
tion, as an energy carrier, from almond shell evaluated in Ref. [22]. As compare the environmental impacts of pellet production from orchard
these studies do not concern palletization, they are not comparable with woody biomass namely, olive and vine biomass. This comparison ad-
pellet production studies. Indeed, in these cases, the analysis does not dresses not only different kinds of biomass but also different handling
include the production process of pellets which represents an essential ways to biomass collection that have already been left untouched in
part of the system presented in Fig. 1. Several studies have also assessed relevant studies. To achieve this target, following a cradle-to-gate
environmental impact of pellet production from different feedstocks, perspective, three subsystems are defined to cover all processes corre-
mostly from short rotation forestry [23–25], forestry residues [26,27] sponding to biomass cultivation, different approaches to collecting
and olive pomace [28]. biomass in field and converting to woodchips and, finally, pellet pro-
Contrary to some of the previously cited studies, an important duction at the gate of pellet factory. The LCA applies 5 life cycle impact

Fig. 1. The diagram of the system boundary of pellet production.

2
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

categories, including energy demand category, to compare environ- caused by application of inputs, i.e. fertilizer, pesticide, and diesel fuel
mental burdens. This study signifies the first comprehensive LCA study consumption for farm operations during the cultivation of grape and
of the pellet production from orchard pruning biomass in Italy. Owing olive were estimated using IPCC guidelines [35]. The life-cycle
to booming market of wood pellet in Europe and requirement for raw inventory data are distributed between the main crops and pruning
material supply [12] and, on the other hand, potential opportunities for residues in conformity with economic allocation. According to
developing the use of agricultural residues in bioenergy production literature, the economic allocation of vine and olive orchard biomass
[32], the methodology applied in this article follows the directive of the accounts for 10% [36] and 1% [19] of main crops’ price, respectively.
European parliament on the bioenergy generation [33] in which co- Due to the lack of data on price of olive pruning, economic value of
products should be taken into account for the purposes of the calcula- apple pruning biomass has been designated for olive biomass. To
tion. Due to the fact that orchard residues contribute to energy gen- estimate biomass production, pruning material was weighted by
eration and they possess an economic value, they should be considered setting up a data acquisition form according to UNI EN 14778:2011
as co-products consisting of a corresponding production phase. Hence, guideline, which governs the methods for taking samples of solid
this study provides worthwhile insights towards sustainable develop- biofuels [37]. The data required includes cultivation areas of the raw
ment of pellet production, in particular, from agricultural residues. The material and relevant features like inter-row line of vineyard, distance
main audience of this study are policymakers and bio-economists who on the row between screw and screw, number of homogeneous rows in
promotes bioenergy for sustainable development. the sampled length, the length of the rows as well as year of planting,
variety of vines, vine nursery, manual or mechanical pruning, the ratio
2. Material and methods between the number of rows and the number of swaths where the
prepared branches are stacked.
2.1. LCA methodology The average vine biomass yield per hectare was determined as dry
weight via the following formula (Formula (1)) [37]:
LCA generally consists of four phases including goal and scope de-
1000
finition, life cycle inventory (LCI), life cycle impact assessment and Rss = Mss* ⎛ ⎞*10−3
⎝ L*I*2 ⎠ (1)
interpretation of results [34].
The target of this study is to assess the environmental footprints where Rss is the production of shoots in dry matter (t/ha); Mss is the
associated with all steps throughout the pellet production system. average weight of the dried samples collected in the vineyard (kg); L is
Special attention is given to orchard biomass collection activities to the length of the inter-row segment (m); I is the width of the inter-row
evaluate the real contribution of biomass provision to the global en- in the sampled section (m); 2 represents the ratio of the number of rows
vironmental burdens. to the number of swaths.
The functional unit selected is 1 MJ of pellet produced as a unit In the case of olive groves, biomass produced by pruning of olive
capable of comparing pellet production from two kinds of biomass. Low trees varies depending on age of plant and the number of plants per
calorific value of olive and vine pellet were considered 16740 MJ hectare. The average pruning material from vineyard and olive groves
tpellet−1 and 16272 MJ tpellet−1, respectively. were estimated as 2.9 and 4.5 t ha−1, respectively.
Table 1 presents the inventory of inputs and on-farm emissions from
2.1.1. System boundary description biomass production.
The bioenergy production chain under assessment was divided in
three principal subsystems illustrated in Fig. 1: 2.1.2.2. Subsystem ‘woodchips production’—SS2. Biomass collection and
pruning into woodchips were managed by two different approaches. In
• SS1 (subsystem 1) involving all farming operations carried out for one approach, olive pruning biomass previously raked and accumulated
orchards biomass production, into rows was shredded into fragments in various sizes by a tractor
• SS2 (subsystem 2) including all the activities involved in woodchips equipped with rotary cutter and simultaneously carried on a storage
production and, then unloaded on an agricultural trailer.
• SS3 (subsystem 3) covering all activities in the factory under as- In second approach, vine branches previously placed in rows were
sessment. packed, by a towed round baler. It took material from ground and
packed it into cylinders 1.2 m wide and 1.5 m in diameter. Once
Waste and disposal protocols are excluded from the boundary. In formed, bale was expelled in the row and subsequently collected by a
addition, the use phase (i.e. combustion of pellets produced) is not tractor with loader and transported out of the vineyard.
included as the system boundary ends at the point of leaving the fac- In both approaches, storage phase lasted 7 months and took place
tory. during the spring-summer period to take advantage of environmental
temperatures and solar radiation. Bales were stored in an open shed
2.1.2. Inventory data collection while woodchips were held in an indoor silo.
2.1.2.1. Subsystem ‘biomass production’—SS1. In this study, the data for The water content of vine and olive grove biomass was measured
foreground system related to biomass production were collected by ranging from 40% to 50% and from 35% to 45%, respectively. On the
visiting and interviewing with the growers in 21 orchards, consisting of other hand, the humidity content of woodchips must be lower than 14%
12 vineyards in provinces of Treviso, Pordenone and Udine and 9 olive prior to comminution (grinding) process for pelletizing. Therefore, this
groves in province of Viterbo. These farms were selected for their can be achieved either through collecting biomass into round bales and
availability for collaboration in this study and because representative natural drying or by drying process in factory.
sample of plantation in terms of homogeneous surface areas. A variety In the area studied, the vine biomass was baled and dried in open
of farm sizes, ranging 0.5–4 ha of olive groves and 0.2–5 ha of air. Through this process the water content of bales dropped to 12–14%
vineyards, were included in samples. The total sample accounted for and therefore, it could directly enter the grinding phase without further
13.8 and 22 ha of the area covering olive groves and vineyards, drying requirements. However, olive woodchips were required to be
respectively. The data included the full annual use of physical inputs dried in the factory. After the natural drying, round bales were chipped
and outputs. For instance, electricity consumption is estimated on by a mobile chipper. In both approaches, around 2% of the total woody
requirements for irrigation operation. Olive is cultivated on non- biomass chipped was directly emitted to the atmosphere as wood dust
irrigated orchards; therefore, no electricity is allocated to this [38].
plantation. In addition, on-farm emissions to the water, soil and air In woodchips production process, the foreground data collected

3
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Table 1 Table 3
Inventory data related to biomass production per functional unit (1 MJ energy Inventory table for pellet production factory per functional unit (1 MJ energy of
of pellet). pellet).
Items Unit Quantity (Unit MJ−1) Items Unit Quantity (Unit MJ−1)

Vine biomass Olive grove biomass Vine woodchips Olive woodchips

A. Output kg 0.064 0.062 A. Output MJ 1

B. Input from technosphere B. Input from technosphere
1- Pesticides kg 1- Woodchips kg 0.062 0.06
Insecticide 2- Electricity kWh
Organophosphorus-compound 5.4E-5 2.6E-5 Drying – 1.25E-3
Fungicide Comminution 2.3E-3 2.25E-3
Captan 9E-5 – Pelletizing 1.4E-3 1.36E-3
Copper oxide – 7.7E-5 Cooling 8.23E-5 8E-5
Herbicide 3- Natural gas MJ – 0.016
Glyphosate – 9E-6 4- Diesel kg 6E-4 5.9E-4
2- Fertilizers kg 5- Plastics kg 5.4E-8 5.2E-8
Compost 6.8E-3 – C. Output to environment
Nitrogen fertilizer, as N 5.7E-4 6.3E-4 Emissions to air
Urea 1.3E-3 – 1- Carbon dioxide (CO2) kg 2.5E-3 2.4E-3
Potassium fertilizer, as K2O 0.002 5.8E-4 2- Wood dust kg 1.24E-3 1.2E-3
Phosphate fertilizer, as P2O5 1.3E-3 3E-4 3- Methane (CH4) kg 3.3E-7 3.2E-7
Sulfur 8.6E-4 – 4- Dinitrogen monoxide (N2O) kg 2.08E-8 2.03E-8
3- Diesel kg 2.5E-3 2E-4
4- Electricity kWh 0.0008 –
C. On-farm emissions
SS3. Secondary data were considered for the production of fuels, elec-
1- Emissions to air kg
Ammonia (NH3) 5.7E-5 6E-5
tricity, plastics and machines, and they were also taken from the
Dinitrogen monoxide (N2O) 2E-5 7E-6 Ecoinvent database® [40,45]. Table 4 shows the infrastructures con-
Nitrogen oxides (NOx) 4.4E-6 1.4E-6 sidered, the materials and their lifetimes.
Carbon dioxide (CO2) 0.010 8.8E-4
Methane (CH4) 1.3E-6 7E-8
2.1.3. Impact assessment
2- Emissions to water kg
Nitrate (NO3−) 0.16 0.2 In this study, environmental results are characterized and normal-
Phosphorus 0.066 0.014 ized using Recipe midpoint (H) v1.13 [46] in SimaPro software (version
3- Emissions to soil kg 8.3.0), in the line with the same approach used in other LCA studies
Pesticides 1.3E-4 1.1E-4
[26,47]. In this approach, normalized results are assessed to identify
potential impacts. Afterwards, these impact categories selected are
characterized for further assessment. Among a total of eighteen impact
were related to fuel consumptions for all of the machines (chipper,
categories assessed, those categories with negligible effects have been
tractor equipped with rotary cutter, trailer, baler and loader).
omitted from displayed results. Subsequently, the following six cate-
Background data related to emissions from production of fossil fuel
gories were assessed in this study:
consumed in the agricultural machinery [39] as well as the production
and use of machinery [40] were extracted from Ecoinvent database®
i. Aquatic eutrophication dividing into two different impact cate-
[41,42].
gories: one for nitrogen concentration in seawater and one for
Table 2 reports the chief data regarding the machines used in the
phosphor concentration in freshwater. For seawater eutrophication,
different operations included in this subsystem.
reference substance can be kg of N emission equivalent from N
fertilizers to freshwater (which will reach coastal waters) and kg of
2.1.2.3. Subsystem ‘pellet factory’—SS3. Data regarding construction P emission equivalent from P fertilizers to freshwater for freshwater
and operation of pelleting plant was obtained from literature [25]. eutrophication.
Therefore, the productivity of plant was assumed as 2000 kg h−1 with ii. Aquatic ecotoxicity including freshwater ecotoxicity (FEc) and
2000 h working hours per year [25]. marine ecotoxicity potential (MEc), measured in kg 1,4-di-
Table 3 summarizes the most relevant inventory data managed in chlorobenzene equivalent (also named 1,4-DBeq in the following

Table 2
Characteristics of machines involved in collecting, transport and chipping activities.
Operation Machinery Power (kW) Weight (kg) Capacity Diesel consumptiona Operation hours Lifespan Average distance (m)

−1 −1
Collecting Tractor 80 2400 – 20 kg ha 1.17 h ha 10000 hb

Rotary cutter 50 1650 3200 kg h−1 – 1.10 h ha−1 1000 hb

Transport to storage Tractor 100 4700 – 20.8 kg h−1 0.25 h 10000 hb 10000
Trailer – 3500 18 m3 – –
Baling Tractor 48.7 4480 – 12 kg ha−1 1.5 h ha−1 10000 hb
Baler 33 2250 1760 kg ha−1 – 1.45 h ha−1 1000 hb
Bale loading Tractor 50 2500 – 13 kg ha−1 1 h ha−1 10000 hb
Loader 31 500 – – –
Transport to storage Tractor 82 4500 – 20 kg h−1 0.25 h 10000 hb 10000
Trailer – 1200 6 m3 – – –
Chipping Chipper 397 4000 18000 kg ha−1 8.6 kg ha−1 0.16 h ha−1 10000 hc

a
Referred to productive machine hours without delays.
b
[43].
c
[44].

4
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Table 4
Characteristics and lifetimes of the pellet plant infrastructures [25].
Sections Infrastructure type Materials Lifetime (y)

Drying Cup elevator, magnetic separator Steel low-alloyed: 700 kg 50

Rotary drum Aluminum wrought alloyed: 640 kg 10
Aluminum sheet rolling: 640 kg
Exhaust fan Aluminum: 1000 kg 50
Steel low-alloyed: 1000 kg
Natural gas boiler Refractory: 70 kg 20
Cast iron: 4200 kg
Chromium steel: 230 kg
Steel low-alloyed: 190 kg
Rock wool: 40 kg
Comminution Cup elevator, screw conveyor Steel low-alloyed: 700 kg 50
2 hammermills Reinforces steel: 2500 kg 10
Steel sheet rolling: 2500 kg
Pelleting 2 feed screw, 2 screw extractors Steel low-alloyed: 700 kg 50
2 presses Steel low-alloyed: 4000 kg 10
Steel rolling: 4000 kg
Cooling 2 feed hoppers, 2 screw conveyors Steel low-alloyed: 700 kg 50
Cooler Steel low-alloyed: 200 kg 15
Screw extractor Steel low-alloyed: 210 kg 50

Fig. 2. Comparison of CED in Vine and Olive pruning pellet production.

5
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

sections of the article): this indicator calculates the effects of ha- the line with those previously reported by Fantozzi and Buratti [25]
zardous chemicals on freshwater, sea and ocean compartments for who investigated energy consumption and environmental footprints of
an infinite time horizon. pellet production from poplar within a Short Rotation Coppice (SRC) in
iii. Human toxicity (HT), measured in kg 1,4-dichlorobenzene equiva- Italy. They reported poplar cultivation (42.1%) and wood pelleting
lent: This category describes fate, exposure and effects of the toxic (37.9%) were main processes which contribute to energy consumption.
chemicals on human health for an infinite time horizon. The ex- Table 5 summarizes the results of energy analysis of vine, olive and
posure routes, for instance, are air and drinking water. poplar wood pellet productions for cumulative energy demand (CED)
iv. Natural land transformation (NLT), measured in m2 natural land: and energy return ratio (ERR) which are metrics used to assess energy
This category characterizes land transformation from natural to technologies [53]. Cumulative energy demand signifies how much
non-natural state and vice versa. The natural land can be recognized primary energy input is demanded to deliver 1 MJ of pellet. The ERR is
as the land use types ‘forest’, ‘tropical forest’ and ‘sea/ocean’. Non- the inverse of CED and defined as the ratio of total usable energy
natural land however is designated as those land types exposed by produced from the process analyzed to the life cycle primary energy
high human intervention (agricultural and urban lands). input (measured as CED). A value of ERR greater than 1 is favourable
for renewable energy sources. Higher ratios represent improved energy
Transformation from natural land is characterized by a positive efficiency of processes. Results indicated that the ERR of olive pellet
characterization factor (1), signifying environmental damage. Whereas, was three times the ERR of vine pellet due to the lower energy re-
negative characterization factor (−1) expresses environmental benefit quirements in biomass cultivation (5.22 vs. 1.7). Higher ERR of olive
allocated to transformation to natural land. pellet production can be interpreted by higher energy delivered from
olive pellet and less life cycle primary energy demand. ERR of wood
2.1.4. Uncertainty and sensitivity analysis pellet was estimated as 3.25 [25]. CED of vine pellet was more than that
Uncertainty issues are always relevant in LCA studies. However, of olive pellet (0.6 vs. 0.19). The differences are interpreted by different
they deem to be more critical when developing comparative models. primary energy use and different energy content of theses pellets.
Therefore, uncertainty analysis is necessary to support results of com-
parative studies [48]. 3.2. Environmental perspective
In this study, two types of uncertainty, data and method un-
certainties are assessed. Data uncertainty is quantified through a de- 3.2.1. Comparative life cycle impact assessment and interpretation
fined range as ± 10%. Sensitivity analysis is also performed to estimate Table 6 presents the distribution of the characterization results
the effect of impact assessment method choice on the outcome of the among the different subsystems. The results show that the environ-
study and compare results of vine and olive pellet by using ReCipe mental impact is mainly due to biomass cultivation (90–99% of the
midpoint (H) and CML-IA methods. total impact) and pelletizing (0.5–7%) [25]. reported similar results for
pellet production from poplar.
2.2. Net energy analysis As it is seen in Table 6, aquatic ecotoxicity (MEc, FEc) of olive
cultivation was higher than that of vine production (5E-4 vs. 3.7E-4 kg
Net energy analysis is a methodology to quantify the energy demand 1,4-DBeq MJ−1). It refers to impacts of toxic substances on freshwater
to manufacture a product or create a service [49,50]. It computes the and marine ecosystems. Similarly, HT of olive cultivation was higher
gross energy requirement considering both direct and indirect energy than that of vine cultivation (1.7E-2 vs. 8E-3 kg 1,4-DBeq MJ−1). The
consumption [51]. To determine primary energy required, it is neces- slight contribution of pelletizing process to these impact categories is
sary to track life cycle of energy through the relevant industrial system owing to the production of electricity requirements in this activity.
[52]. Net energy analysis is based on the same system boundary used in However, if the emissions related to the biomass cultivation were ex-
the LCA [53]. The cumulative energy demand (CED) [54] was applied cluded from the pellet production system, the major contribution to all
to calculate gross energy requirement. categories would be affected by electricity consumption in pelletizing
phase similarly to other studies [26,28].
3. Results and discussions A comparative overview of characterization results for vine and
olive pellet are presented in Fig. 3. Evaluation of environmental in-
3.1. Energy perspective dicators indicated that impact of HT, FE and MEc categories for olive
pellet production were higher than those of vine pellet production and,
Fig. 2 shows the comparative shares of different activities for vine therefore, vine pellet was more environmentally compatible than olive
and olive pellet. For vine pellet production, the primary energy use was pellet for most of impact categories. Nevertheless, production of vine
found to be 0.6 MJ MJ−pellet
1
, from which biomass cultivation, pelletizing pellet generates higher environmental loads in the forms of FEu and
and baling were the main contributors and they were consumed as NLT. The difference in MEu of vine and olive pellet production was
0.48 MJ MJ−pellet
1
, 0.06 MJ MJ−pellet
1
and 0.027 MJ MJ−pellet
1
respectively. negligible. Higher MEc, FEc and HT of olive pellet production were
This results also revealed that chipping phase and transportation were caused by emissions in production phase of copper oxide applied as
the least demanding energy activities. Life cycle primary energy re- pesticide in olive growing. The optimum application of this pesticide or
quirement for olive pellet production was 0.19 MJ MJ−pellet 1
, which was replacement with alternative pesticide are possible pathways to reduce
less than that of vine pellet. Biomass cultivation (0.09 MJ MJ−pellet
1
) and associated environmental emissions. FEu of olive pellet production was
pelletizing (0.06 MJ MJ−pellet
1
) were the main contributors to the total 28% of that of vine pellet (Fig. 3). This means that 1 MJ energy from
energy inputs in olive pellet production. Furthermore, energy needs olive pellet causes 72% less eutrophication than 1 MJ energy from vine
from chipping (0.02 MJ MJ−pellet
1
) was relatively high. These results are in pellet. FEu, expressed as kg Peq per kg emission, encompasses all

Table 5
Energy analysis result (CED, ERR) for orchard pruning pellet and wood pellet.
Vine pruning pellet Olive pruning pellet Poplar wood pellet [25]

Cumulative energy demand (CED) (MJ primary MJ delivered−1) 0.6 0.19 0.3
Energy return ratio (ERR) (MJ delivered MJ primary−1) 1.7 5.22 3.25

6
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Table 6
Characterization results associated with the packed pellet production (ReCipe, FU: 1 MJ energy of pellet).
Impact category Unit Vine pellet (Unit MJ−1) Olive pellet (Unit MJ−1)

SS1 (%) SS2 (%) SS3 (%) SS1 (%) SS2 (%) SS3 (%)

FEu kg Peq 1E-4 (99) 2.4E-7 (0.43) 6E-7 (0.57) 2.8E-5 (97) 2.9E-7 (1) 5.9E-7 (2)
MEc kg 1,4-DBeq 1.8E-4 (89) 8.6E-6 (4.5) 1.4E-5 (6.5) 3E-4 (91) 9.8E-6 (4) 1.6E-5 (5)
FEc kg 1,4-DBeq 1.9E-4 (89) 9.2E-6 (4.3) 1.5E-5 (6.7) 2E-4 (90) 1E-5 (4.1) 1.8E-5 (5.9)
HT kg 1,4-DBeq 8E-3 (90) 3E-4 (4.8) 4E-4 (5.2) 1.7E-2 (94) 4E-4 (2.8) 5E-4 (3.2)
NLT m2 8.6E-6 (78) 1.2E-6 (12) 1.1E-6 (10) 1E-6 (34) 7.7E-7 (28) 1.1E-6 (38)
MEu kg Neq 6.4E-5 (99) 3E-7 (0.5) 3.5E-7 (0.5) 6.2E-5 (99) 1.8E-7 (0.4) 4.2E-7 (0.6)

Freshwater eutrophication (FEu), Marine ecotoxicity (MEc), Freshwater ecotoxicity (FEc), Human toxicity (HT), Natural land transformation (NLT), Marine eu-
trophication (MEu).

impacts owing to immoderate levels of macro-nutrients in the en- 3.2.2. Uncertainty and sensitivity analysis
vironment caused by emissions of nutrients to water. It principally In order to show the changes of comparative eco-profiles of olive
depends on the production and application of fertilizers, chiefly urea and vine pellet by application of different LCA models, the CML-IA
and phosphate. Nitrogen emissions (especially due to NO−3 1) and phos- baseline model was used and LCA results of olive and vine pellet are
phate emissions owing to fertilizer consumption are mainly contributed comparatively presented in Table 7. Impact categories of MEc, FEc, HT
to FEu [55]. and EP were found to be main contributors to environmental emissions
To sum it up, comparative LCA of vine and olive pellet showed that profile of both vine and olive pellet by normalizing the result in CML-IA
for most of the impact categories, the environmental loads of olive model. Other impacts were found to be immaterial and are therefore
pellet were more than those of vine pellet, and this was greatly owing to excluded from analysis.
higher consumption of pesticides, especially copper oxide, for olive Using CML model, human toxicity impact for vine and olive pellet
cultivation, which entails mainly toxic substances such as copper and was found to be 0.01 and 0.016 kg 1,4-DBeq MJ−pellet
1
, respectively. High
zinc emitted to water during manufacturing process. In fact, environ- value of eutrophication potential from vine pellet is considerably re-
mental impacts of products connected to agricultural systems depend sulted from emissions by fertilizer use especially urea in the farm. By
on a large extent on farmer production practices. using CML model, HT, FEc and MEc for olive pellet were higher than
Fig. 4 illustrates the percentage contribution of normalized impact those of vine pellet. The similar comparative results were obtained in
categories from olive cultivation in the region. The greatest effects on ReCipe midpoint (H) model presented in the previous section; however,
most of the impact categories were derived from production of inputs the magnitude of some of the impact categories was different in two
such as pesticides, fertilizers and diesel, followed by emissions from models which can be interpreted by discrepancies in approaches to
application of fertilizers such as nitrate, phosphorus, CO2, NH3, N2O characterization modeling. For example, magnitude of MEc is con-
and NOx. Normalized impact categories for vine biomass production are siderably different in CML-IA and ReCipe, but when analyzing the
presented in Fig. 5. Among vine biomass production inputs, the main percentage contribution of different inputs on MEc, pesticides and
contributors to total emissions were fertilizers, particularly potassium chemical fertilizers are main contributors in both of the models.
sulfate and urea, followed by emissions from their applications. Ap- Sensitivity analysis of different emission sources on impact cate-
plication of nitrogen fertilizer contributes to MEu impact; therefore, gories of vine and olive pellet are presented in Fig. 6. Since it is ex-
replacing nitrogen fertilizer by manure will reduce MEu in vine biomass pected that variations create the symmetrical increasing and decreasing
production in the region. Hence, the optimum application of nitrogen impacts in average values of the individual input parameters on out-
fertilizer is encouraged to decrease nutrient leaching and NOx emissions come of model, only the increasing orientation is presented in the
causing aquatic eutrophication. diagrams. These diagrams show the variables which have the greatest
effects on each of the impact categories. As seen, in olive pellet

Fig. 3. Comparative characterization results of Olive and Vine pruning pelletizing (ReCipe, FU: 1 MJ energy of pellet).

7
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Fig. 4. Percentage contribution of normalized impact categories in Olive biomass cultivation (FU: 1 MJ energy of pellet).

production, the most sensitive variable on HT, FEc and MEc was copper Table 7
oxide. However, HT, FEc and MEc in vine pellet production showed the Impact assessment results at characterization for vine and olive pellet produc-
highest sensitivity on potassium sulfate. This is interpreted by high tion using (CML-IA, FU: 1 MJ energy of pellet).
dependence of vineyards on this fertilizer use. Phosphate fertilizer, Impact category Unit Vine pruning pellet Olive pruning pellet
diesel and nitrogen fertilizer contributed the most to FEu, NLT and
MEu, respectively, in vine and olive pellet productions. As magnitude of HT kg 1,4-DBeq 0.01 0.016
FEc kg 1,4-DBeq 0.006 0.008
MEc in CML-IA and ReCipe models was considerably different, the
MEc kg 1,4-DBeq 21.2 24.9
contribution of emission sources on MEc is analyzed further in detail. In EP kg PO4eq 3E-4 1E-4
environmental profile of vine pellet by ReCipe model, MEc was con-
tributed by potassium sulfate (41.7%), urea (24.4%), diesel (6.8%), Human toxicity (HT)Fresh water aquatic ecotox. (FEc), Marine aquatic ecotox.
pesticides (5.6%), phosphate fertilizer (5%), and pelletizing (4.8%); (MEc), Eutrophication (EP).
similarly, in CML-IA model MEc of vine pellet was contributed by po-
tassium sulfate (42%), urea (20.7%), pesticides (6.7%), pelletizing copper oxide, potassium sulfate, pelletizing, and nitrogen fertilizer were
(5.9%), phosphate fertilizer (5.4%), and diesel (5%). These results re- found to be substantial contributors to MEc in the Recipe (Fig. 6, O5)
vealed that, in vine pellet production, potassium sulfate and urea are and CML-IA models.
dominated by other emission sources in both ReCipe (Fig. 6, V5) and Overall, the results of uncertainty and sensitivity analysis signified
CML-IA models. However, in environmental assessment of olive pellet, that fertilizers and pesticides had high impacts on all impact categories.

Fig. 5. Percentage contribution of normalized impact categories in Vine biomass cultivation (FU: 1 MJ energy of pellet).

8
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Fig. 6. Sensitivity analysis of input parameters on impact categories for olive and vine pellet production systems.

Efficient use of pesticides and fertilizers in olive groves and vineyards transformation. Moreover, energy analysis revealed that ERR of olive
can reduce corresponding environmental emissions. pellet was higher than that of vine pellet production. Therefore, the
olive pellet biomass chain has a more favorable energy balance than
4. Conclusion that of vine pellet.
Sensitivity analysis results also revealed that fertilizers and pesti-
This study presented the results of a comparative life cycle assess- cides had the highest contributions to all impact categories. Hence,
ment of the whole process of pellets production from the cultivation to efforts should focus on reduction and suitable use of nutrients and ef-
the packaging for both vine and olive. The novelty of this investigation ficient management of pesticide which lead to decrease in environ-
can be ascribed to the lack of previous analyses on palletization of these mental impacts and to enhance ERR of pellet production systems from
biomasses within the selected boundaries including cultivation and orchard biomass.
pellet production and based on the essential assumption that biomass Systematic application of fertilizers and pesticides are strongly en-
residue is not a byproduct but a coproduct. couraged within the framework of sustainable development of pro-
As regards results, the LCA study indicated that all environmental duction systems basically targeted on main crops. However, since re-
burdens are mostly imposed by agricultural operations. Olive pellet had sidues are coproducts of fruit production system, they contribute to all
higher human toxicity, marine ecotoxicity, freshwater ecotoxicity and emissions and impacts corresponding to cultivation of the main crop.
marine eutrophication. Nonetheless, vine pellet production system was And on the other hand, cultivation-related impacts dominate all en-
characterized by higher freshwater eutrophication and natural land vironmental categories, therefore, amelioration of production practices

9
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

Fig. 6. (continued)

can mitigate overall impacts of pellet production system. In this line, Author contributions
application of fertilizers at an appropriate rate and a decent timing
through an efficient method count as possible pathways for more en- The contribution to the programming and executing of this special
vironmentally friendly pellet production from orchard biomass. must be equally divided by the authors.
These specific findings as a pattern can be applied by managers and
decision makers to recognize the most energy-efficient and environ-
Conflicts of interest
mental-cleaner pellet production systems in Italy.
From global perspective, however, this study reveals that taking
The authors declare no conflicts of interest.
into account the production phase of biomass residues in life cycle as-
sessment studies leads to delving deeply into agricultural practices and
as a result it can be able to provide a comprehensive insight towards Acknowledgments
sustainability assessment of biomass-based energy systems. In fact, this
conclusion is not limited to pellet production in Italian territory, but it The research was partially supported by MIUR (Italian Ministry for
can be, to a large extent, generalized to all types of bioenergy systems. education, University and Research) in the context of the initiative
“Department of excellence”, Law 232/216, and AGRENER project by
MIPAAF (Ministry of Agriculture, Food and Forestry, Rome, Italy).
We thank all authors of the special issue.

10
S. Rajabi Hamedani, et al. Biomass and Bioenergy 129 (2019) 105334

References pelleting using life cycle assessment, Biomass Bioenergy 84 (2016) 107–117,
https://doi.org/10.1016/j.biombioe.2015.11.018.
[29] S. Rajabi Hamedani, L. Del Zotto, E. Bocci, A. Colantoni, M. Villarini, Eco-efficiency
[1] V. Montola, N. Colonna, V. Alfano, M. Gaeta, S. Sasso, V. De Luca, et al., Censimento assessment of bioelectricity production from Iranian vineyard biomass gasification,
Potenziale Energetico Biomasse, Metodo Indagine, Atlante Biomasse Su WEB-GIS, Biomass Bioenergy 127 (2019) 105271, https://doi.org/10.1016/j.biombioe.2019.
RSE/2009/1 (2009). 105271.
[2] M. Gerritsen, Renewable energy policy in the European Union is there a firm [30] J. Moreno, J. Dufour, Life cycle assessment of hydrogen production from biomass
foundation for the EU's 2030 renewable energy targets? Eur. Policy Rev. 1 (2015) gasification. Evaluation of different Spanish feedstocks, Int. J. Hydrogen Energy 38
3–30, https://doi.org/10.1080/026782900202480. (2013) 7616–7622, https://doi.org/10.1016/j.ijhydene.2012.11.076.
[3] Eurostat, Share of renewable energy in gross final energy consumption n.d, https:// [31] D. Prando, M. Boschiero, D. Campana, R. Gallo, S. Vakalis, M. Baratieri, et al.,
ec.europa.eu/eurostat/tgm/table.do?tab=table&init=1&language=en&pcode= Assessment of different feedstocks in South Tyrol (Northern Italy): energy potential
t2020_31&plugin=1. and suitability for domestic pellet boilers, Biomass Bioenergy 90 (2016) 155–162,
[4] M. Szabó, A. Jäger-waldau, F. Monforti-ferrario, N. Scarlat, H. Bloem, https://doi.org/10.1016/j.biombioe.2016.03.039.
M. Quicheron, et al., Tech. Assess. Renew. Energy Action Plans 47 (2011), https:// [32] Y. Iqbal, I. Lewandowski, A. Weinreich, B. Wippel, B. Pforte, O. Hadai, et al.,
doi.org/10.2788/57968. Maximising the yield of biomass from residues of agricultural crops and biomass
[5] N.L. Panwar, S.C. Kaushik, S. Kothari, Role of renewable energy sources in en- from forestry-final report European commision, (2016).
vironmental protection: a review, Renew. Sustain. Energy Rev. 15 (2011) [33] Council of the European Union, Directive of the European parliament and of the
1513–1524, https://doi.org/10.1016/j.rser.2010.11.037. council on the promotion of the use of energy from renewable sources, Brussels
[6] J. Popp, Z. Lakner, M. Harangi-Rákos, M. Fári, The effect of bioenergy expansion: (2018).
food, energy, and environment, Renew. Sustain. Energy Rev. 32 (2014) 559–578, [34] ISO 14040, Environmental Management - Life Cycle Assessment - Principles and
https://doi.org/10.1016/j.rser.2014.01.056. Framework, (2006).
[7] E. Gallagher, The gallagher review of the indirect effects of biofuels production, [35] IPCC, IPCC guidlines for national greenhouse gas inventories, Available at: http://
Renew. Fuels Agency (2008). www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html.
[8] T. Searchinger, R. Heimlich, R.A. Houghton, F. Dong, A. Elobeid, J. Fabiosa, et al., [36] J. Moreno, J. Dufour, Life cycle assessment of hydrogen production from biomass
Use of U.S. croplands for biofuels increases greenhouse gases through emissions gasification. Evaluation of different Spanish feedstocks, Int. J. Hydrogen Energy 38
from land-use change, 423 (2008). (2013) 7616–7622, https://doi.org/10.1016/j.ijhydene.2012.11.076.
[9] M. Spöttle, S. Alberici, G. Toop, D. Peters, L. Gamba, S. Ping, et al., Low ILUC [37] UNI EN 14778/2011, Solid biofuels - sampling guideline, (2011), p. 1275.
potential of wastes and residues for biofuels, (2013), p. 168. [38] C. Whittaker, N. Mortimer, R. Murphy, R. Matthews, Energy and greenhouse gas
[10] J. Sánchez, M. Dolores Curt, M. Sanz, J. Fernández, A proposal for pellet production balance of the use of forest residues for bioenergy production in the UK, Biomass
from residual woody biomass in the island of Majorca (Spain), AIMS Energy 3 Bioenergy 35 (2011) 4581–4594, https://doi.org/10.1016/j.biombioe.2011.07.
(2015) 480–504, https://doi.org/10.3934/energy.2015.3.480. 001.
[11] N. Audigane, M. Bentele, J.M. Ferreira, Á. Gyurik, J.-M. Jossart, A.-C. Mangel, et al., [39] R. Spinelli, N. Magagnotti, Determining long-term chipper usage, productivity and
European Pellet Report-PellCert Project, (2012). fuel consumption, Biomass Bioenergy 66 (2014) 442–449, https://doi.org/10.
[12] R. Sikkema, M. Steiner, M. Junginger, W. Hiegl, M.T. Hansen, A. Faaij, The 1016/j.biombioe.2014.04.016.
European wood pellet markets: current status and prospects for 2020, Biofules, [40] T. Nemecek, T. Kagi, Life cycle inventories of Agricultural Production Systems,
Bioprod. Biorefining 5 (2011) 250–278. ecoinvent report No. 15, Final Rep. Ecoinvent V20 (2007) 1–360.
[13] R. Pokharel, R.K. Grala, D.L. Grebner, S.C. Grado, Factors affecting utilization of [41] B.P. Weidema, C. Bauer, R. Hischier, C. Mutel, T. Nemecek, J. Reinhard, et al.,
woody residues for bioenergy production in the southern United States, Biomass Overview and Methodology. Data Quality Guideline for the Ecoinvent Database,
Bioenergy 105 (2017) 278–287, https://doi.org/10.1016/j.biombioe.2017.07.002. (2013) version 3.
[14] W. Mauser, G. Klepper, F. Zabel, R. Delzeit, T. Hank, B. Putzenlechner, et al., Global [42] R. Frischknecht, N. Jungbluth, H.J. Althaus, G. Doka, R. Dones, T. Heck, et al., The
biomass production potentials exceed expected future demand without the need for ecoinvent database: overview and methodological framework, Int. J. Life Cycle
cropland expansion, Nat. Commun. (2015), https://doi.org/10.1038/ncomms9946. Assess. 10 (2005) 3–9, https://doi.org/10.1065/lca2004.10.181.1.
[15] IEA Bioenergy, Global Wood Pellet Industry Market and Trade Study, (2011). [43] W. Edwards, Estimating Farm Machinery Costs. Machinery Management, Iowa State
[16] M. Guo, W. Song, J. Buhain, Bioenergy and biofuels: history, status, and perspec- University extention and outreach, 2015 Available at: https://www.extension.
tive, Renew. Sustain. Energy Rev. 42 (2015) 712–725. iastate.edu/agdm/crops/html/a3-29.html.
[17] C.R.R. Pleguezuelo, V.H.D. Zuazo, C. Bielders, J.A.J. Bocanegra, F. PereaTorres, [44] R. Spinelli, L. Eliasson, N. Magagnotti, Value retention, service life, use intensity
J.R.F. Martínez, Bioenergy farming using woody crops. A review, Agron. Sustain. and long-term productivity of wood chippers as obtained from contractor records,
Dev. 35 (2014) 95–119, https://doi.org/10.1007/s13593-014-0262-1. Forests 8 (2017) 1–13, https://doi.org/10.3390/f8120503.
[18] M. Boschiero, F. Cherubini, C. Nati, S. Zerbe, Life cycle assessment of bioenergy [45] R. Dones, C. Bauer, R. Bolliger, B. Burger, T. Heck, A. Röder, et al., Life cycle in-
production from orchards woody residues in Northern Italy, J. Clean. Prod. 112 ventories of energy systems: results for current systems in Switzerland and other
(2016) 2569–2580, https://doi.org/10.1016/j.jclepro.2015.09.094. UCTE countries. Ecoinvent report No. 5. Dübendorf, CH: Paul scherrer institut
[19] M. Boschiero, M. Kelderer, A.O. Schmitt, C. Andreotti, S. Zerbe, Influence of agri- villigen, Swiss Cent. Life Cycle Inventories (No. 5) (2007).
cultural residues interpretation and allocation procedures on the environmental [46] M. Goedkoop, R. Heijungs, M. Huijbregts, A. De Schryver, J. Struijs, R. Van Zelm,
performance of bioelectricity production - a case study on woodchips from apple ReCiPe 2008-a Life Cycle Impact Assessment Method Which Compares Harmonised
orchards, Appl. Energy 147 (2015) 235–245, https://doi.org/10.1016/j.apenergy. Category Indicators at the Midpoint and the Endpoint Level. Report I:
2015.01.109. Characterisation, Ministry of Housing, Spatial Planning and Environment (VROM),
[20] G. Picchi, S. Silvestri, A. Cristoforetti, Vineyard residues as a fuel for domestic The Netherlands, 2009 10.029/2003JD004283.
boilers in Trento Province (Italy): comparison to wood chips and means of polluting [47] C. Buratti, M. Barbanera, F. Testarmata, F. Fantozzi, Life Cycle Assessment of or-
emissions control, Fuel 113 (2013) 43–49, https://doi.org/10.1016/j.fuel.2013.05. ganic waste management strategies: an Italian case study, J. Clean. Prod. (2014)
058. 1–12.
[21] B. Torquati, D. Marino, S. Venanzi, P.R. Porceddu, M. Chiorri, Using tree crop [48] ISO 14044, Environmental Management -life Cycle Assessment - Requirements and
pruning residues for energy purposes: a spatial analysis and an evaluation of the Guidelines, (2006).
economic and environmental sustainability, Biomass Bioenergy 95 (2016) 124–131, [49] R. Frischknecht, F. Wyss, S.B. Knöpfel, T. Lützkendorf, M. Balouktsi, Cumulative
https://doi.org/10.1016/j.biombioe.2016.09.017. energy demand in LCA : the energy harvested approach, (2015), pp. 957–969,
[22] S. Rajabi Hamedani, M. Villarini, A. Colantoni, M. Moretti, E. Bocci, Life cycle https://doi.org/10.1007/s11367-015-0897-4.
performance of hydrogen production via agro-industrial residue gasification—a [50] G.P. Hammond, Engineering sustainability : thermodynamics, energy systems, and
small scale power plant study, Energies 11 (2018) 675, https://doi.org/10.3390/ the environment, Int. J. Energy Res. 639 (2004) 613–639, https://doi.org/10.1002/
en11030675. er.988.
[23] P.W.R. Adams, J.E.J. Shirley, M.C. Mcmanus, Comparative cradle-to-gate life cycle [51] F. Roberts, The aims, methods and uses of energy accounting, Appl. Energy 4 (1978)
assessment of wood pellet production with torrefaction, Appl. Energy 138 (2015) 199–217, https://doi.org/10.1016/0306-2619(78)90003-X.
367–380, https://doi.org/10.1016/j.apenergy.2014.11.002. [52] E. Pohl, D. Diarra, A method to determine primary energy savings of CHP plants
[24] D. Ruiz, G.S. Miguel, B. Corona, F.R. López, LCA of a multifunctional bioenergy considering plant-side and demand-side characteristics, Appl. Energy 113 (2014)
chain based on pellet production, Fuel 215 (2018) 601–611, https://doi.org/10. 287–293, https://doi.org/10.1016/j.apenergy.2013.07.038.
1016/j.fuel.2017.11.050. [53] P.W.R. Adams, M.C. McManus, Small-scale biomass gasification CHP utilisation in
[25] F. Fantozzi, C. Buratti, Life cycle assessment of biomass chains : wood pellet from industry: energy and environmental evaluation, Sustain. Energy Technol.
short rotation coppice using data measured on a real plant, Biomass Bioenergy 4 Assessments 6 (2014) 129–140, https://doi.org/10.1016/j.seta.2014.02.002.
(2010) 0–8, https://doi.org/10.1016/j.biombioe.2010.07.011. [54] R. Hischier, B. Weidema, Althaus, C. Hans-Jörg Bauer, G. Doka, R. Dones,
[26] A. Laschi, E. Marchi, S. Gonz, Environmental performance of wood pellets ’ pro- Frischknecht, S. Rolf Hellweg, et al., Implementation of Life Cycle Impact
duction through life cycle analysis, Energy (2016) 103, https://doi.org/10.1016/j. Assessment Methods Data, (2010), https://doi.org/10.1007/s40265-016-0641-9
energy.2016.02.165. v2.2.
[27] M. Roder, C. Whittaker, P. Thornley, How certain are greenhouse gas reductions [55] V. Fantin, S. Righi, I. Rondini, P. Masoni, Environmental assessment of wheat and
from bioenergy ? Life cycle assessment and uncertainty analysis of wood pellet-to- maize production in an Italian farmers' cooperative, J. Clean. Prod. 140 (2017)
electricity supply chains from forest residues, Biomass Bioenergy (2015) 9, https:// 631–643, https://doi.org/10.1016/j.jclepro.2016.06.136.
doi.org/10.1016/j.biombioe.2015.03.030.
[28] A. Kylili, E. Christoforou, P.A. Fokaides, Environmental evaluation of biomass

11