Construction and Building Materials 72 (2014) 333–339

Contents lists available at ScienceDirect

Construction and Building Materials

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

Carbon footprint of coarse aggregate in Brazilian construction

Efigênia Rossi, Almir Sales ⇑
Department of Civil Engineering, Federal University of São Carlos, Via Washington Luís, km 235, São Carlos, SP 13565-905, Brazil

h i g h l i g h t s

 The carbon footprint was 1.50 kg CO2e per ton of coarse aggregate in Brazil.
 The stages of CO2e contribution were extraction, transport and crushing process.
 The critical stage was the crushing process because of electric power consumption.
3
 The diesel total consumption was 2.91E01 kg per m of coarse aggregate.
3
 The electric power consumption was 10.77 kW h per m of coarse aggregate.

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

Article history: This research measures the carbon dioxide equivalent emissions (CO2e) of the extraction, transport and
Received 27 June 2014 crushing process of coarse aggregate in Brazil. The goal is to quantify the carbon footprint and identify the
Received in revised form 23 August 2014 critical stages of those activities. The quantification of CO2e was determined by the emissions of carbon
Accepted 27 August 2014
dioxide (CO2), carbon monoxide (CO), nitrous oxide (N2O), methane (CH4) and non-methane volatile
Available online 8 October 2014
organic compound (NMVOC). The results show that the crushing process is the critical stage because
of its high electric power consumption. These factors should be addressed in future Brazilian environ-
Keywords:
mental studies in order to find alternative solutions prior to exploring new quarries.
Global warming
Construction materials
Ó 2014 Elsevier Ltd. All rights reserved.
Carbon dioxide equivalent
Power consumption
Coarse aggregate

1. Introduction Studies show that during the process of manufacturing cement,

1.1. Background
In terms of volume used, concrete is one of the most consumed
materials in the world, totalling more than one cubic meter per
person per year [1]. Concrete production commonly uses
aggregates, cement and water in duly adjusted proportions. These
aggregates can either be sand, coarse aggregate or gravel. Specifi-
cally, coarse aggregate varies between 30% to 60% of concrete,
depending on the desired compressive strength, modulus of
elasticity and other properties [2,3].
These aggregates are abundant in Brazil, and their consumption in
2011 was about 3.50 ton per inhabitant—a low rate compared with
developed countries. For example, the average consumption in
United States is about six to seven tons per inhabitant per year [4].
Nevertheless, the aggregates sector in Brazil is growing due to high
investments in infrastructure and habitation, as showed in Fig. 1.
⇑ Corresponding author. Tel.: +55 16 33518659; fax: +55 16 33518262.
E-mail address: almir@ufscar.br (A. Sales).
and consequently in concrete, that a large amount of CO2 is emit-
ted [5–7]. The Intergovernmental Panel on Climate Change (IPCC)
‘‘has estimated that cement and ceramic manufacture were
responsible for more than 20% of the world’s industrial carbon
dioxide (CO2) production’’ [6]. But none of these studies quantifies
the carbon emission of each material used in concrete mixtures,
particularly the coarse aggregate and the critical stages to obtain it.
The anthropogenic GHGs intensify the phenomenon of global
warming. According to the Intergovernmental Panel on Climate
Change (IPCC), ‘Carbon dioxide (CO2) is the most important anthro-
pogenic GHG. Its annual emissions have grown between 1970 and
2004 by about 80%, from 21 to 38 Gigatonnes (Gt), and represented
77% of total anthropogenic GHG emissions in 2004’ [8].
Carbon footprint is defined as ‘a measure of the climate change
impact of the product where all the greenhouse gas emissions emit-
ted during the product life cycle are taken into account’ [9]. This
total amount of other GHG is calculated by carbon dioxide equiva-
lent emissions (CO2e) to establish an equal dimensional unit.
The life cycle incorporates product and process analyses before
they can be dismissed or treated linearly from an environmental

http://dx.doi.org/10.1016/j.conbuildmat.2014.08.090
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
334 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339

Fig. 1. Aggregates consumption in Brazil, in millions of tons [4].

Fig. 2. System boundary.

perspective. In this view, production is seen as an end but also as
an essential part of the product composition and must therefore
be consistent with the earlier stages of extraction and subsequent
final disposal [10]. The life cycle in this study is a product system
from a cradle-to-gate view, considering since the extraction to
the manufacturing process. But there are other ways such as cradle
to cradle, considering the recycling in the manufacturing process
and cradle to grave, analysing since the extraction of raw materials
until the final disposal.
Some studies have been done to quantify the environmental
impact of using concrete. A Life-Cycle Carbon dioxide (LCCO2)
method was used to quantify emissions in different types of con-
cretes used in civil construction [11]. The life cycle inventory of a
road must include the construction, operation and final disposal
[12]. The Life Cycle Assessment (LCA) could include the results of
different building materials [13]. The CO2 emissions could also be
quantified as factors applied in various materials [14]. However,

Table 1 Table 2
GWP factors related to diesel. GWP in kg CO2.

Substance g substance/L diesel [15] Substance kg CO2e/kg substance [10]

CH4 0.182 CH4 4.00E02

CO2 2799.000 CO2 1.00E+00
CO 60.500 CO 5.00E01
NMVOC 14.720 NMVOC 6.67E01
N2O 0.022 N2O 3.13E03
 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339 335

identified the electric power and the diesel consumption in the

process and compared them to earlier studies.

2. Materials and methods

2.1. Functional unit and system boundaries

The life cycle of coarse aggregate to be applied in civil construction has six
stages: extraction, transport to the crushing process, crushing process, storage,
transport from the crushing process to the trucks and transport to the consumer
centres. The inputs and outputs were obtained by a technical visit to a basalt
quarry. An inventory was done to store data. An inventory analysis was done in
order to identify which stages contribute to the CO2e emissions. The stages with
CO2e contributions were extraction, transport to the crushing process, crushing
process, transport from the crushing process to the trucks and transport to the con-
sumer centres. The system’s boundaries are summarized in Fig. 2. The functional
unit adopted was 1 m3 of coarse aggregate.

2.2. Methodology

The quantification of CO2e was determined by the emissions of CO2, CO, N2O,
CH4 and NMVOC adjusted to 1 m3 of coarse aggregate. In this paper, the carbon
footprint emissions were calculated from the consumption of diesel with some
factors shown in Table 1. As for the consumption of diesel, this was determined
by the equipment’s operation hours.
The contribution of each substance was determined by CO2e, as shown in
Eq. (1) [10], where Q is the quantity of substance emitted (kg substance/m3 coarse
aggregate) and Global Warming Potential (GWP) is the emission factor in Table 2.

CO2e ¼ Q =GWP ð1Þ

Fig. 3. The drill used to insert explosives into the basalt. 2.3. Assumptions

For this paper, the following assumptions were made:

none of these studies analyses the critical stages of the processing
used to obtain the raw materials.
1.2. Aims
This paper aims to quantify the carbon footprint of 1 m3 of
coarse aggregate used in Brazilian civil construction and to identify
the critical stage production. The CO2e emissions were quantified
by considering the cradle-to-gate view. The stone analysed was
basalt, and the processing stages were extraction, transport to
the crushing process, crushing process, transport from the crushing
process to the trucks and transport to the consumer centres.
1.3. Research significance
The contribution of this study is identifying the critical stages
of CO2e emissions considering the cradle-to-gate view. We also
Table 3
The list of substances emitted per m3 of coarse aggregate in activities for stage 1.
(a) The emissions were determined for 1m3 of coarse aggregate that ranged in
size from 4.8 to 12.5 mm [16].
(b) The diesel density adopted was an average value of 0.835 kg/L [17].
(c) Basalt’s specific gravity was 2630 kg/m3 [18].
(d) Basalt mining average daily production calculated as 1475.5 m3 of coarse
aggregate.
(e) Average equipment operation hours were determined as 10 h in mining.
(f) In Brazil, the electric energy mix is 69% hydropower, 28% fossil, 2% nuclear
and other renewable sources 2% [19].
(g) The CO2 emission factors for Brazilian electricity generation were 271.3 kg
CO2 per MW h for 2013, as calculated by the Brazilian Ministry of Science,
Technology and Innovation [20].
(h) Electrical power loss quantified in mining was 30%.
(i) The storage stage was not considered for CO2e emissions because it is
temporary and only material or energy flows were considered.
(j) In stage 4 (transport to the consumer centres), the average distance
between the mining and the centres is 50 km. The minimum distance used
is 1 km and the maximum is 150 km. The process is not viable if the
distance is further than 150 km.

Description Quantities Dimensional unit

CH4 4.20E05 kg CH4/m3 coarse aggregate
CO2 from ANFO 1.28E04 kg CO2/m3 coarse aggregate
CO2 6.46E01 kg CO2/m3 coarse aggregate
CO 1.40E02 kg CO/m3 coarse aggregate
NMVOC 3.40E03 kg NMVOC/m3 coarse aggregate
N2O 5.08E06 kg N2O/m3 coarse aggregate

Table 4
Substances emitted per m3 of coarse aggregate in activities for stage 2.

Description Quantities Dimensional unit

CH4 3.01E06 kg CH4/m3 coarse aggregate
CO2 4.63E02 kg CO2/m3 coarse aggregate
CO 1.00E03 kg CO/m3 coarse aggregate
NMVOC 2.44E04 kg NMVOC/m3 coarse aggregate
N2O 3.64E07 kg N2O/m3 coarse aggregate
Fig. 4. The excavator transports the basalt to the truck.
336 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339

(k) The truck in stage 4 has 250 horsepower (HP), made in 2010 with a capacity
of 12 m3. This means that this vehicle can carry 14–15 tons of coarse
aggregate, while consuming 3.4 km/L of diesel. Thus, the total calculated
fuel was 8.32E03 kg diesel per m3 coarse aggregate, considering a half a
distance load and half an unloaded load.
the material to trucks, as shown in Fig. 4. The diesel consumption
of the excavator was 9.07 L diesel per hour.
The carbon dioxide emission factor per litre of diesel is 2799 g
[15], and the total of emissions by diesel combustion used in this
stage are shown in Table 3.

3. Results and discussion

3.2. Transport to the crushing process

3.1. Extraction
The basalt was obtained by explosions in the stone. The
explosives were made by using a drill, as shown in Fig. 3. The diesel
consumption for the drill was 25 L for each hour of equipment
work.
The explosive used was Ammonium Nitrate and Fuel Oil
(ANFO). Eq. (2) summarizes this reaction in a balanced environ-
ment [21]:
3NH4 NO3 þ CH2 ! 3N2 þ CO2 þ 7H2 O þ 912 cal=g
For each 1 m3 of coarse aggregate, 700 g of ANFO is necessary.
The esteometry of Eq. (1) shows the emission of 1.28E04 kg
CO2 per m3 of basalt. After the explosion, an excavator transports
This stage is mainly characterised by transporting the materials
to the crushing process, which is about one kilometre from the
quarry. Some trucks transport the material, consuming 1.07 km/L
of diesel. Considering the same combustion process and the
emission factor mentioned above (2799 g CO2 per litre of diesel),
the CO2 total emission in combustion is as shown in Table 4.
Even in the diesel combustion process, there are other sub-
stances to consider. All the factors used for these substances were
mentioned in [15] and the emissions are shown in Table 4. For
carbon monoxide emissions, the factor used was 60.5 g CO per litre
ð2Þ
of diesel. The nitrous oxide factor is 0.022 g N2O per litre of diesel.
The factor of methane is 0.182 g/L diesel. The last substance
studied was a Non-methane Volatile Organic Compound (NMVOC),
with a factor of 14.72 g NMVOC per litre of diesel.

Fig. 5. Fluxograms for equipment used in crushing process are described in Table 5.
 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339 337

3.3. Crushing process

The crushing process is characterised by the fragmentation of

the basalt and its sift, resulting in transport by the desired com-
mercial trucks. Moreover, several crushing and screening processes
are powered by electric energy.
Fig. 5 illustrates the crushing process, including the equipment
used, as detailed in Table 5. The electrical power consumption was
determined by the equipment operation hours, and the results are
shown in Table 5, along with the capacity of each equipment.
The CO2 emission factors for Brazilian electricity generation
were 271.3 kg CO2 per MW h for 2013, as calculated by the
Brazilian Ministry of Science, Technology and Innovation [15,22].
For stage 3, the total consumption of energy was 10.77 kW h
per m3 of coarse aggregate, so the CO2 emissions 2.92E+00 kg
CO2 per m3 of coarse aggregate.

Fig. 6. Wheeled loaders transport the coarse aggregate to trucks.

Table 5
The electrical power consumption for each equipment in Fig. 5.
3.4. Transport from the crushing process to the trucks
Item Equipment Capacity Equipment Potência Electrical
(operation power (HP) (kW) power The transport from the crushing process to the trucks is done by
hours/m3 consumption
the wheeled loaders shown in Fig. 6.
basalt) (kW h)
The consumption of this equipment is 13.47 l of diesel per hour.
1 Vibrating 0.0070 40 29.44 0.21
The total found was 7.62E02 kg diesel per m3 coarse aggregate.
feeder
2 Primary jaw 0.0070 200 147.2 1.03 For this total, the emissions of CO2, CO, CH4, N2O and NMVOC,
crusher calculated by [15] factors, are shown in Table 6.
3 Conveyor belt 0.0070 10 7.36 0.05
4 Waste scrap
5 Conveyor belt 0.0070 50 36.8 0.26
Table 6
6 Coarse aggregate Substances emitted per m3 of coarse aggregate in activities for stage 4.
7 Feeder 0.0070 10 7.36 0.05
channel Description Quantities Dimensional unit
8 Conveyor belt 0.0105 20 14.72 0.15
CH4 2.01E06 kg CH4/m3 coarse aggregate
9 Grid 0.0105 15 11.04 0.12
CO2 2.56E01 kg CO2/m3 coarse aggregate
10 Secondary 0.0105 125 92 0.97
CO 5.52E03 kg CO/m3 coarse aggregate
crusher
NMVOC 1.34E03 kg NMVOC/m3 coarse aggregate
11 Conveyor belt 0.0105 20 14.72 0.15
N2O 1.66E05 kg N2O/m3 coarse aggregate
12 Conveyor belt 0.0105 15 11.04 0.12
13 Coarse aggregate 0
14 Conveyor belt 0.0105 15 11.04 0.12
15 Vibrating 0.0061 15 11.04 0.07
screen 1
Table 7
16 Conveyor belt 0.0006 7.5 5.52 0.00
Substances emitted per m3 of coarse aggregate in activities for stage 5.
17 Crusher 1 0.0061 400 294.4 1.78
18 Conveyor belt 0.0061 10 7.36 0.04 Description Quantities Dimensional unit
19 Conveyor belt 0.0061 30 22.08 0.13
20 Crusher 2 0.0061 500 368 2.23 CH4 2.19E07 kg CH4/m3 coarse aggregate
21 Vibrating 0.0061 30 22.08 0.13 CO2 2.79E02 kg CO2/m3 coarse aggregate
screen 2 CO 6.03E04 kg CO/m3 coarse aggregate
22 Conveyor belt 0.0061 15 11.04 0.07 NMVOC 1.47E04 kg NMVOC/m3 coarse aggregate
23 Conveyor belt 0.0061 7.5 5.52 0.03 N2O 1.81E06 kg N2O/m3 coarse aggregate
24 Conveyor belt 0.0061 10 7.36 0.04
25 Conveyor belt 0.0061 15 11.04 0.07
26 Conveyor belt 0.0061 20 14.72 0.09
27 Coarse aggregate 3
28 Conveyor belt 0.0071 10 7.36 0.05
29 Vibrating 0.0071 30 22.08 0.16
screen 3
30 Conveyor belt 0.0071 10 7.36 0.05
31 Conveyor belt 0.0071 10 7.36 0.05
32 Conveyor belt 0.0071 10 7.36 0.05
33 Coarse aggregate 1
34 Gravel
35 Stone dust
36 Crushed stone
37 Coarse aggregate 2
Total 8.29
Total with 10.77
loss
Fig. 7. Carbon footprint.
338 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339

The CO2, CO, CH4, N2O and NMVOC emissions calculated by [15]
factors are shown in Table 7.

3.6. Characterisation

In order to standardise the dimensional units, all of the

substances were transformed in terms of CO2e. For each stage,
the emissions of CO2e are showed in Fig. 7. Moreover, this graph
shows that the critical stage is the crushing process (stage 3). In
this paper, the total carbon footprint was 3.95 kg CO2e per m3
coarse aggregate or 1.50 kg CO2e per m3 coarse aggregate.
In each stage, diesel consumption was calculated as shown in
Fig. 8. Diesel consumption for each stage. Fig. 8, and the diesel total consumption was 2.91E01 kg per m3
coarse aggregate. The critical stage was the extraction because
the of the high consumption of the drill.
Regarding the source of the CO2e emissions, most of these
emissions come from electric power consumption and from diesel
and explosives, as shown in Fig. 9.
Diesel
26.062% 3.7. Comparison with earlier studies

Studies from [12] and [23] offer some results similar to this
paper (Table 8). Every substance emitted has the same dimensional
ANFO magnitude, excluding the N2O.
0.003% Although [12] adopted a cradle-to-grave view to coarse
aggregate, they considered only the production. Thus, these similar
results could be explained because both studies adopted a
Electric
cradle-to-gate view.
73.935%
Comparing this paper with Marinković et al. [23], there are
similarities with CO2 and CO values, but differences in the other
values. The study from Parka et al. [11] showed 0.40E03 kg CO2
per kg coarse aggregate as an emission value.
Regarding to the source of the CO2e emissions, most of these
Fig. 9. Sources of CO2e emissions. emissions come from electric power consumption and from diesel
and explosives. Fig. 10 shows the comparison between this paper
(a) and [14] (b) in terms of the sources of CO2e emissions. More-
Table 8 over, the majority GWP came from electricity and in granite slabs
The list of substances emitted per m3 of coarse aggregate in activities for stage 5.
production, thereby confirming the same pattern [24].
Substance kg Substance emitted/kg coarse aggregate
Present study [12] [23] 4. Conclusions
CH4 1.99E09 3.82E09 1.30E06
This study quantified the carbon dioxide equivalent (CO2e)
CO2 1.48E03 1.42E03 1.38E03
CO 2.71E06 1.49E06 3.48E06 emissions generated by the Brazilian activities with the cradle-
N2O 7.14E09 3.61E08 5.50E08 to-gate view of the coarse aggregate life cycle. The carbon footprint
shows the most critical stage as is the crushing process because of
the high electric power consumption. Even the other stages have a
3.5. Transport to the consumer centres significant impact, but the CO2 emission in the generation of elec-
trical power could explain the increased values of CO2e in stage 3.
In this stage, the coarse aggregate was transported by trucks to Electric power consumption represents about 74% of these
the consumer centres. The average distance was 50 km and the emissions, regardless of the equipment used in stage 3. The diesel
diesel consumption was 3.4 km/L. consumption was 2.91E01 kg per m3 of coarse aggregate and the

Fig. 10. Comparison between this paper (a) and [14] (b) about the source emissions.
 E. Rossi, A. Sales / Construction and Building Materials 72 (2014) 333–339
total carbon footprint was 1.50 kg CO2e per ton of coarse aggre-
gate in Brazil. This value most be considered in environmental
studies due to quantifying these impacts and seeking new alterna-
tives before exploring new quarries.
The comparison with earlier studies showed some important
similarities in emissions and their sources. For future studies, it
will be important to compare the recycled aggregates’ emissions
in order to analyse if they could reduce the carbon footprint of
raw materials.
Acknowledgments
The authors acknowledge CAPES and MCTI/CNPq for their finan-
cial support.
References
[1] Scrivener KL, Kirkpatrick RJ. Innovation in use and research on cementitious
material. Cem Concr Res 2008;38:128–36.
[2] Nikbin IM, Beygi MHA, Kazemi MT, Amiri JV, Rahmani E, Rabbanifar S, et al. A
comprehensive investigation into the effect of aging and coarse aggregate size
and volume on mechanical properties of self-compacting concrete. Mater Des
2014;59:199–210.
[3] Nikbin IM, Beygi MHA, Kazemi MT, Amiri JV, Rahmani E, Rabbanifar S, et al.
Effect of coarse aggregate volume on fracture behavior of self compacting
concrete. Constr Build Mater 2014;52:137–45.
[4] IBRAM Brazilian Mining Association. Brazilian Mineral Overview: aggregates,
<http://www.ibram.org.br/sites/1400/1457/00000369.pdf/>; 2012.
[5] Higuchi T, Morioka M, Yoshioka I, Yokozeki, K. Development of a new
ecological concrete with CO2 emissions below zero. Const Build Mater.
[17.02.14].
[6] Heath A, Paine K, McManus M. Minimising the global warming potential of
clay based geopolymers. J Clean Prod, <http://dx.doi.org/10.1016/
j.jclepro.2014.04.046>; 2014.
[7] Thiel CL, Campion N, Landis AE, Jones AK, Schaefer LA, Bilec MM. A materials
life cycle assessment of a net-zero energy building. Energies 2013;6:1125–41.
[8] IPCC Intergovernmental Panel on Climate Change. Climate change 2007:
synthesis report. Spain, <http://www.ipcc.ch/publications_and_data/ar4/syr/
en/mains2-1.html#footnote5>; 2007.
339
[9] ISO/TS 14067:2013, 5.15.2013. Greenhouse gases e carbon footprint of
products e requirements and guidelines for quantification and
communication. 1st ed. Technical Specification.
[10] Wenzel H, Hauschild M, Alting L. Environmental assessment of
products. Bonton/Dordrecht/London: Kluwer Academic Publishers; 1997.
[11] Parka J, Taea S, Kima T. Life cycle CO2 assessment of concrete by compressive
strength on construction site in Korea. Renew Sust Energy Rev
2012;16:2940–6.
[12] Stripple H. Life cycle assessment of road: a pilot study for inventory analysis.
In IVL Swedish Environmental Research Institute. Sweden, <http://www.ivl.se/
download/18.7df4c4e812d2da6a416800071481/1350484198080/
B1210E.pdf>; 2001.
[13] Kellenberger D, Althaus H. Relevance of simplifications in LCA of building
components. Build Environ 2009;44:818–25.
[14] Flower DJM, Sanjayan JG. Green house gas emissions due to concrete
manufacture. Int J Life Cycle Assess 2007;12:282–8.
[15] MCTI Ministry of Science, Technology and Innovation, Emissões de Gases de
Efeito Estufa no Setor Energético por Fontes Móveis (Greenhouse gases
emission emission factors by mobile sources in energy sector), Brazil, <http://
www.mct.gov.br/index.php/content/view/17352.html>; 2006 [in Portuguese].
[16] ABNT – NBR 7225 – Materiais de pedra e agregados naturais (Stone materials
and raw aggregates). Rio de janeiro; 1993 [in Portuguese].
[17] Barabás I. Predicting the temperature dependent density of biodiesel–diesel–
bioethanol blends. Fuel 2013;109:563–74.
[18] Sales A, Rodrigues de Souza F, Nunes dos Santos W, Zimer AM, Almeida FCR.
Lightweight composite concrete produced with water treatment sludge and
sawdust: thermal properties and potential application. Constr Build Mater
2010; 24: 2446–53.
[19] Nogueira LPP, Lucena AFP, Rathmann R, Rochedo PRR, Szklo A, Schaeffer R. Will
thermal power plants with CCS play a role in Brazil’s future electric power
generation? In J Greenhouse Gas Control 2014;24:115–23.
[20] MCTI Ministry of Science, Technology and Innovation, CO2 emission factors,
<http://www.mct.gov.br/index.php/content/view/74689.html>; 2013 [in
Portuguese].
[21] Borg DG, Chiappetta RF, Morhard RC, Sterner VA. Explosives and rock
blasting. Dallas, USA: Atlas Powder Company, Maple Press; 1987.
[22] Calili RF, Souza RC, Galli A, Armstrong M, Marcato ALM. Estimating the cost
savings and avoided CO2 emissions in Brazil by implementing energy efficient
policies. Energ Policy 2014;67:4–15.
[23] Marinkovic’ S, Radonjanin V, Malešev M, Ignjatovic’ I. Comparative
environmental assessment of natural and recycled aggregate concrete.
Waste Manage 2010;30:2255–64.
[24] Mendoza JF, Oliver-Solà J, Gabarrell X, Josa A, Rieradevall J. Life cycle
assessment of granite application in sidewalks. Int J Life Cycle Assess
2012;17:580–92.