Carbon footprint of a pale lager packed in different formats: Assessment and sensitivity analysis based on transparent data

Full text

Carbon footprint of a pale lager packed in different formats:
assessment and sensitivity analysis based on transparent data
Alessio Cimini, Mauro Moresi
*
Department for Innovation in Biological, Agro-Food, and Forest Systems, University of Tuscia, Via S. C. de Lellis, 01100 Viterbo, Italy
article info
Article history:
Received 5 September 2014
Received in revised form
29 May 2015
Accepted 13 June 2015
Available online xxx
Keywords:
Aluminum can
Carbon footprint
Glass bottle
Pale lager beer
Sensitivity analysis
Steel keg
abstract
Energy and water consumption, waste generation, and emissions to air are the main environmental
issues of the brewing industry. Several strategies have been so far proposed to reduce its impact on the
global climate. This study assessed the environmental impact of the industrial production and distri-
bution of 1 hL of a pale lager, as packed in different formats by the Italian brewery Birra Peroni Srl (Rome,
Italy) over the period April 2012eMarch 2013, in compliance with the Publicly Available Specification
2050 standard method. The estimated carbon footprint of 1 hL of lager beer packaged in 66-cL glass
bottles, 33-cL glass bottles assembled in cardboards or cluster packs, 33-cL aluminum cans, or 30-L steel
kegs was of the order of 57, 67, 74, 69, or 25 kg CO
2e
, respectively. Such a difference in the overall carbon
footprint values was due to the diverse contributions of packaging materials and transportation. In
particular, the impact of packaging materials was minimum in the case of kegs, in virtue of the high reuse
coefficient, and maximum in the case of the 33-cL glass bottle cluster packs. The estimated carbon
footprint values were considerably lower than those reported in the most recent literature, probably
because of the large production scale and short distribution chain of Birra Peroni brewery, utilization of
beer co-products as feed and anaerobic digestion of liquid wastewaters. Owing to the linearity of the
mathematical model of the carbon footprint, its sensitivity to the change of one-emission factor-at-a-
time allowed the main hot spots in the life cycle of beer (i.e., glass bottle production and barley culti-
vation) to be identified and targeted for mitigating the carbon footprint of any pale lager, both of them
being not related to the brewery production scale examined here. The scientific value of this work relies
on the choice of estimating the carbon footprint using wholly transparent data in order to allow its direct
comparison to other estimates, as well as its straightforward re-calculation using better quality data, as
available.
©2015 Elsevier Ltd. All rights reserved.
1. Introduction
In spite of the main commitments to reduce the current
greenhouse gas (GHG) emissions in the Earth atmosphere and
mitigate climate change under the Kyoto Protocol targets, the GHG
emissions associated to the food supply chain are significantly
contributing to the Earth global warming. For instance, Tukker et al.
(2006) estimated that the food, drink, tobacco, and narcotics area of
consumption accounted for 22e31% of the so called Global
Warming Potential (GWP) impact category referred to a series of
products (i.e., cars, food, heating, and house building) consumed in
the EU-25. Within this area of consumption, meat and meat
products had the greatest environmental impact with an estimated
contribution to GWP in the range of 4e12 % of all products, whereas
the soft-drinks and alcoholic beverages accounted for the 0.9 and
0.6% of the GWP of total products, respectively. Thus, the beverage
sector has started implementing strategies to reduce its impact on
the global climate, as focused for instance by the Beverage Industry
Environmental Roundtable on the basis of the sensitivity of the
beer GWP to variations in material or process practice aspects (such
as packaging material selection, distribution logistics, recycling
rates, etc.) in either Europe or North America (BIER, 2012). Despite
its ancient tradition, the brewing industry is currently pushed to
perform within the constraints of product quality, process safety,
economic viability, and limited environmental damage.
The main environmental issues associated with brewing include
material, water, and energy consumption, as well as waste pro-
duction. Brewery processes are intensive users not only of both
*Corresponding author. Tel.: þ39 0761 357494; fax: þ39 0761 357498.
E-mail address: moresi@unitus.it (M. Moresi).
Contents lists available at ScienceDirect
Journal of Cleaner Production
journal homepage: www.elsevier.com/locate/jclepro
http://dx.doi.org/10.1016/j.jclepro.2015.06.063
0959-6526/©2015 Elsevier Ltd. All rights reserved.
Journal of Cleaner Production xxx (2015) 1e18
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

electrical and thermal energy, but also of good-quality water. By
referring to some European breweries, their corresponding specific
consumption yields range from 8 to 12 kWh, from 100 to 200 MJ,
and from 4 to 7 hL per hL of beer produced, respectively (Olajire,
2012). Additionally, wastewater generation ranges from 1 to 5 %
of total beer production. Several studies have been sofar carried out
to understand and verify the environmental impact generated
through the entire life cycle of some lager beers in Chile (Mu~
noz
et al., 2012), Denmark (Frees and Pedersen Weidema, 1998),
Estonia (Talve, 2001), Greece (Koroneos et al., 2005), Italy (Cordella
et al., 2008), Japan (Takamoto et al., 2004), and Spain (Hospido
et al., 2005). Also a few environmental product declarations (EPD,
2010, 2011a,b) provided reliable quantification and certification of
the environmental performance of lager-type beers packed in
disposable 33-cL glass bottles, 25-L steel kegs, or 20-L plastic
drums, according to the Life Cycle Assessment methodology (ISO,
2006).
By accounting for the beer life cycle, Hospido et al. (2005) and
Talve (2001) assessed that the most relevant environmental impact
of the agricultural subsystem regarded the eutrophication impact
category, owing to the leakage of nitrogen- and phosphorous-rich
compounds from agricultural fields, production of fertilizers, etc.
The packaging system used greatly affected the GHG emissions,
these ranging from about 106 or 149 kg CO
2e
hL
1
of beer barreled in
plastic or steel kegs, up to 190 kg CO
2e
hL
1
of beer bottled in glass
bottles (EPD, 2011a). Whereas the GHG emissions associated with
the production process represented as little as 14.8e17.7 % of the
overall ones, those issued from the upstream production or down-
stream processes were the greatest ones when including glass bottle
production or steel keg distribution, respectively (EPD, 2011a).
Another study was performed by the Climate Conservancy
(2008) in cooperation with New Belgium Brewing Company to
assess the GHG emissions across the full life cycle of Fat Tire
®
Amber Ale (FT) using the British Standard Publicly Available Spec-
ification (PAS 2050) standard method (BSI, 2008a), as detailed in
the Guide to PAS 2050 (BSI, 2008b). The system boundaries of the
life cycle study included acquisition and transport of raw materials,
brewing operations, business travel, employee commuting, trans-
port and storage during distribution and retail, use and disposal of
waste. The estimated carbon footprint (CF) was about 150 kg CO
2e
per hL of beer packaged in glass bottles of 12-fluid ounce capacity
(1 flounce z29.57 mL) in a 6-bottle selling unit, this figure being
quite smaller than the values mentioned above (EPD, 2011a).
Owing to the numerous factors, assumptions, and performance
variables that can impact the above calculated carbon footprints,
current carbon quantification exercises cannot in principle be used
to compare not only different beverages, but also the same category
of beverages. Despite several comparative LCA studies of beer
production have been published over the last decade, the use of
qualitative, often incomplete, data and, what is more, unknown
emission factors allow no direct comparison among the environ-
mental scores reported.
Among the numerous international standards (i.e., PAS 2050;
Bilan Carbone
®
;Environmental Product Declaration, EPD
®
;Green-
house Gas Protocol;Australian Wine Carbon Calculator, AWCC)
currently available to assess product and service environmental
impact, the great majority of them, except the EPD
®
procedure,
focuses on the single impact category of climate change only, and
does not assess other potential social, economic and environmental
impacts (Moresi, 2014). Despite the carbon footprint is a “reduced
scope”tool, which does not provide an indicator of the overall
environmental impact for the activity examined, it appears to be an
opportunity for small and medium enterprises to make their ac-
tivities more sustainable, as well as to satisfy the interest of the
market towards eco-labeling initiatives.
To calculate the carbon footprint, it is crucial to rely on appro-
priate emission factors. Similarly, the estimation of all the envi-
ronmental impact categories, considered for instance by the Eco-
indicator 99 method, asks for an impressive number of prefixed
parameters in order to reduce the 11 impact categories to just three
damage categories (Goedkoop and Spriensma, 2001). Even the use
of a well known LCA software, such as Simapr
o (PR
e Consultants,
Amersfoort, NL), does not help to understand concealed material,
energy, and emission interrelationships. In the circumstances, it is
difficult to make comparisons among the CF studies reported in the
literature, the typical weakness of which being their non-
transparent nature regarding the use of data sources and limita-
tions. Only the Bilan Carbone
®
and AWCC procedures refer to
manuals reporting a series of emission factors (ADEME, 2007; NGA,
2013), and thus make the GHG emissions estimated by different
companies for a given food or drink not only comparable, but also
transparent. On the contrary, the quite numerous EPD
®
studies
openly available at <www.environdec.com>provide no informa-
tion about the emission factors used, this making such environ-
mental assessment exercises no way repeatable and, thus,
ineffective from a scientific point of view.
The first aim of this study was to develop a life-cycle assessment
model to estimate the carbon footprint (CF) of the industrial pro-
duction and distribution of 1 hL of lager beer in different packaging
formats (i.e., 66- or 33-cL glass bottles, 33-cL aluminum cans, 30-L
stainless steel kegs) and selling units (i.e., carton, tray, or cluster-
multipack), by indicating clearly not only all the processing and
packaging consumption yields, but also all the emission factors
used. The second aim of this study was to carry out a sensitivity
analysis of CF to assess the influence of different parameters (such
as, origin of raw materials and their cultivation methods, GHG
emissions per kWh of electric energy generated by fossil and/or
renewable sources, transportation by road or railway, etc.) by
changing one-emission factor-at-a-time and keeping the others at
their nominal values in order to identify the most promising
strategy to mitigate the GHG emissions associated to the produc-
tion and distribution of the pale lager of concern.
2. Methods
The life-cycle analysis was performed in compliance with the
PAS 2050 standard method (BSI, 2008a, 2008b), the stages of which
being the following: goal and scope definition, inventory analysis,
impact assessment, and interpretation of results. The scope of this
study was to assess the environmental impact from cradle to beer
distribution centers, this complying with a business-to-business
study in accordance with PAS 2050.
2.1. Goal and scope
The goal for this study was to develop an LCA model to assess
the carbon footprint of a pale lager beer, made of malted barley,
maize grits and hop pellets, and produced from the Italian brewery
Birra Peroni Srl (Rome, Italy), and consumed in Italy, as well as to
identify its life-cycle hot spots. The purpose of the study was also to
provide information for carbon labeling so as to inform the con-
sumer on the total carbon footprint of lager beer, as well as to aid
environmentally conscious decision-making.
2.2. Functional unit
The analysis was based on a functional unit defined as 1 hL of
lager beer packaged in different packaging formats and selling
units.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e182
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

2.3. System boundary
The system boundary for this study was not restricted to the
brewing process; and, in accordance with the LCA approach, it also
included the upstream and downstream phases. The whole system
is represented in Fig. 1 and included:
i) the agricultural processes of barley, corn, and hop
cultivation;
ii) the production of malt, maize grits, and beer;
iii) the production of the packaging materials (glass bottles;
aluminum cans; stainless steel kegs; closures; labels; card-
board cartons, trays or cluster-multipacks; wrap films; etc.),
as well as the auxiliary ones (oxygen; carbon dioxide;
phosphoric acid; calcium chloride; diatomaceous earth, DE;
polyvinyl-polypyrrolidone, PVPP; cleaning products);
iv) the transport of raw, process, and packaging materials from
their production sites to the brewery gate, as well as that of
packaged beer to the distribution centers;
v) the disposal of packaging material waste generated in the
brewery;
vi) the treatment of wastewaters, including the production of
methane by anaerobic digestion, and
vii) the production of heat, as well as electricity utilization from
the Italian grid.
All identification items used in Fig. 1 are given in the List of
symbols.
Since roots, spent grains, and yeast surplus are by-products that
occur during malting and brewing processes, they were considered
as an avoided production of cattle feed. This effect was accounted
for in the life cycle assessment by means of CO
2
credits.
Contrary to PCR (2013), this case study neglected the beer
consumption phase (i.e., beer refrigeration, dispensing, and losses;
consumer displacement and wastewater treatment). Use phase
scenarios are indeed extremely variable and a precise determina-
tion would have required an onerous investigation. Consequently,
the waste generation in the consumption stage was not included;
while the distribution stage was limited to the distribution centers,
the consumer transport to and from the retail shop being excluded
by the PAS 2050 standard method (see Section 6.5). As stipulated by
PAS 2050, the production of capital goods (machinery, equipment
and energy wares) was also excluded from the system boundary.
Finally, all CO
2
credits from recycling of renewable and non-
renewable materials were included.
2.4. Data gathering and data quality
According to PAS 2050 (Section 7.2), the following was stated:
i) Geographic scope: this LCA study focused on the production,
and distribution of Peroni lager differently packaged in Italy.
Fig. 2 shows the maps of raw materials purchased and Peroni
lager sales distributed by the Birra Peroni brewery (Rome,
Italy).
ii) Time scope: the reference time period for assessing the car-
bon footprint values was April 2012eMarch 2013. According
to Assobirra (2013), the overall beer production by the three
breweries owned by Birra Peroni Srl in Rome, Bari and Padoa
(Italy) in the years 2009e2013 was approximately constant
and equal to 329,460 ±6736 m
3
. Thus, the beer production
was regarded as about steady over the above quinquennium,
the change in beer demand being of the order of 2%.
iii) Technical reference: the process technology underlying the
datasets used in this study reflects process configurations, as
well as technical and environmental levels, which are typical
for industrial-scale lager beer processing in the reference
period.
iv) Primary data for this PAS 2050-compliant study were
collected from the Italian brewery Birra Peroni Srl (Rome,
Italy).
v) Secondary data were sourced from the Italian Institute for
Environmental Protection and Research as concerning the
GHG emissions associated to the Italian electric energy pro-
duction by renewable and non-renewable sources (ISPRA,
2012), an LCA software (i.e., Simapro 7.2 v.2, Pr
e Consul-
tants, Amersfoort, NL), and several databases (such as
BUWAL 250, Ecoinvent, ETH-ESU 96, Franklin USA 98, Idemat
2001, LCA Food-DK) using the method developed by IPCC
(2007), as well other technical reports (ADEME, 2007; BIER,
Fig. 1. Beer system boundary (TR ¼transport).
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 3
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

2012; BSI, 2008b; Chicago Manufacturing Center, 2009;
Cimini and Moresi, 2015; Climate Conservancy, 2008; Coca
Cola, 2010; Field To Market, 2012; International Aluminium
Institute, 2013; ISPRA, 2012; Kanokkantapong et al., 2009;
Ma et al., 2012; Manfredi et al., 2009).
Table 1 summarizes the emission factors, expressed in kg CO
2e
emitted over a time horizon of 100 years, associated to the manu-
facture of the raw and packaging materials, ingredients and pro-
cessing aids, as well as to the delivery of the main energy sources,
aforementioned materials, final product in the tertiary packages of
concern, and solid wastes generated during the brewing process
under study. Since the final carbon footprint values are heavily
dependent on the values chosen for such factors, these were
carefully checked to provide perspectives on the key drivers to the
product environmental impact and were deliberately shown to
make transparent the CF estimation.
2.5. Inventory analysis
In this stage of the LCA procedure all the inputs of resources and
energy and yield outputs for raw material production; beer pro-
cessing, packaging, and transportation; waste management; and
refrigerant losses were gathered as reported below.
2.5.1. Raw materials
The agricultural stage included barley, maize, and hop cultiva-
tion and postharvest operations (i.e., cleaning; de-stoning; grain
germination; bran, grit and germ separation; drying; storage).
Barley cultivation gives rise to several GHG emissions (as due to
seed production, farm machinery operations, irrigation, use of
fertilizers, pesticides, and soil amendments, as well as emissions
from the soil) totaling ~0.638 kg CO
2e
kg
1
of barley (Climate
Conservancy, 2008). Malting consists of barley steeping in water
and germination, followed by malt drying and roasting. Each of
these steps requires both electric and thermal energy. According to
the Chicago Manufacturing Center (2009), the GHG emissions
associated to malting amounted to (0.292 ±0.084) kg CO
2e
kg
1
of
malted barley, while the average ratio of barley-to-malt was
approximately equal to 4:3 (Climate Conservancy, 2008). Thus, the
default emission factor for malted barley was estimated as 1.143 kg
CO
2e
kg
1
. In this study, malted barley was produced at the malt-
house Saplo Spa (Pomezia, Rome, Italy) from barley cultivated in
Central and South Italy, the overall cultivation area being about
16,880 ha (Birra Peroni, 2012). Fig. 2a shows the cultivation map of
barley purchased by Birra Peroni Srl. By accounting for the amounts
of barley supplied by the regions of Italy differently colored in the
map, it was possible to estimate the kilometric distance traveled by
any lot of barley from its cultivation site to the Saplo processing
plant (Pomezia, Italy) and then to the brewery using the web site
http://www.viamichelin.it, these average distances being about
236 and 33 km, respectively.
As concerning the maize grits, two hybrid varieties of corn (that
is, Nostrano Peroni PR and ME) were used as raw materials (Birra
Peroni, 2012). Such varieties were cultivated in the provinces of
Parma and Messina (Italy), respectively, as shown by the areas
marked in yellow in Fig. 2a. Thus, their average transport distance
was of about 608 km. Based on the USA GHG emissions per bushel
of corn for grain from 1980 to 2011 (Field to Market, 2012), corn
growth and harvesting would result in ~0.227 kg CO
2e
kg
1
of grain
corn, its moisture mass fraction being generally 0.155. Grain corn
conversion into maize grits at an average moisture content of 12% is
of the order of 0.5 kg per kg of rawcorn (Gresser, 2010; Mejía, 2003)
and involves several operations, such as grain conditioning,
degermination, drying, grading, and milling (Mejía, 2003). By
assuming that such a process would take almost the same energy of
Fig. 2. Maps of (a) barley and maize grits supply and (b) all-format Peroni beer sales distribution, both referred to the brewery examined in this work. The different colors used to
mark the regions of Italy refer to the mass fraction percentages of barley supplied to the malthouse Saplo Spa (Pomezia, Italy) in order to be converted into malt and then
transported to the brewery Birra Peroni Srl (Rome, Italy), or to the beer sales volume (BS) expressed in m
3
, respectively. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e184
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Table 1
Minimum, maximum and default values of the GHG emission factors for the energy sources; means of transport; raw and packaging materials, detergents, and processing aids;
refrigeration fluids; waste disposal; and byproduct uses, applied in this carbon footprint exercise, as extracted from several databases (i.e., BUWAL250, Ecoinvent, ETH-ESU 96,
Franklin USA 98, Idemat 2001, Industry data 2.0, LCA Food-DK, USA Input Output database 98) of the LCA software Simapro 7.2 v.2 (Pr
e Consultants, Amersfoort, NL) and
technical references.
Emission factor Minemax values Default value Unit Ref.
Energy source
Electricity, medium voltage, at grid/IT 0.582e0.818 0.3931 kg CO
2e
kWh
1
BUWAL 250; ISPRA (2012)
Electricity, fossil fuels, at grid/IT 0.545 kg CO
2e
kWh
1
ISPRA (2012)
Photovoltaic electricity 0.055 kg CO
2e
kWh
1
ADEME (2007)
Diesel oil (upstream þcombustion) 3.487 kg CO
2e
kg
1
ADEME (2007)
Natural gas (upstream þcombustion) 0.064 kg CO
2e
MJ
1
ADEME (2007)
Means of transport
Transport, lorry >32 Mg, EURO5-3 0.104e0.117 kg CO
2e
Mg
1
km
1
Ecoinvent
Transport, lorry 16e32 Mg, EURO5-3 0.154e0.168 kg CO
2e
Mg
1
km
1
Ecoinvent
Transport, lorry 7.5e16 Mg, EURO5-3 0.268e0.291 kg CO
2e
Mg
1
km
1
Ecoinvent
Transport, lorry 3.5e7.5 Mg, EURO5-3 0.657e0.635 kg CO
2e
Mg
1
km
1
Ecoinvent
Transport, freight, rail/RER S 0.0393e0.0577 0.0393 kg CO
2e
Mg
1
km
1
Ecoinvent; ETH-ESU 96
Raw materials, ingredients, detergents, processing aids
Barley (produced in Europe, or in the
USA, transport not included)
0.19e0.62
0.53e0.79
0.26
0.66
kg CO
2e
kg
1
BIER (2012)
Barley Malting 0.199e0.362 0.292 ±0.084 kg CO
2e
kg
1
Chicago Manufacturing Center (2009)
Italy-grown malted barley 1.143 kg CO
2e
kg
1
This work
Foreign malted barley 0.545e1.345 kg CO
2e
kg
1
This work
Corn grain 0.20e0.55 0.227 kg CO
2e
kg
1
Ma et al. (2012); Field to Market (2012)
Maize grits 0.746 kg CO
2e
kg
1
This work
Hop Pellets (transport not included) 2.39 kg CO
2e
kg
1
Climate Conservancy (2008)
PVPP 2.41 kg CO
2e
kg
1
Cimini and Moresi (2015)
Compressed air (>30 kW, 8 bar) 0.0627 kg CO
2e
m
3
Ecoinvent
Oxygen 0.228 kg CO
2e
kg
1
BUWAL250
Water, deionized, at plant/CH S 1.03 kg CO
2e
m
3
Ecoinvent
Tap water, at user/CH S 0.154 kg CO
2e
m
3
Ecoinvent
Lime (burnt) ETH S, equiv. to DE 1.39 kg CO
2e
kg
1
LCA Food-DK
Gypsum 0.27 kg CO
2e
kg
1
ETH-ESU 96
CalCl
2
0.931 kg CO
2e
kg
1
Ecoinvent
ZnSO
4
.H
2
O 1.85 kg CO
2e
kg
1
Ecoinvent
H
3
PO
4
, 85% in H
2
O 1.46 kg CO
2e
kg
1
Ecoinvent
HNO
3
, 50% in H
2
O 0.308e3.20 3.2 kg CO
2e
kg
1
Ecoinvent
NaOH 50% in H
2
O 1.12 kg CO
2e
kg
1
Ecoinvent
Acetic acid, 98% in H
2
O 1.58 kg CO
2e
kg
1
Ecoinvent
H
2
O
2
, 50% in H
2
O 1.21 kg CO
2e
kg
1
Ecoinvent
CO
2
0.266 kg CO
2e
kg
1
BUWAL 250
Refrigeration fluids
Ethylene glycol 1.61 kg CO
2e
kg
1
Ecoinvent
Ammonia liquid 2.12 kg CO
2e
kg
1
Ecoinvent
Packaging materials
Glass bottles, 100% recycled evirgin 0.546e0.936 kg CO
2e
kg
1
Franklin USA 98
Glass bottles, 10% recycled 0.57 Climate Conservancy (2008)
Al cans, 50% recycled 8.96 kg CO
2e
kg
1
International Aluminium Institute (2013)
Al cans, unspecified recycling rate 9.38 kg CO
2e
kg
1
Coca Cola (2010)
Aluminum, 99% purity 11.5 kg CO
2e
kg
1
Idemat 2001
Stainless steel AISI 316 4.02 kg CO
2e
kg
1
Idemat 2001
Stainless steel kegs 1.70 kg CO
2e
kg
1
Chicago Manufacturing Center (2009)
Steel crowns, 28% recycled 2.81 kg CO
2e
kg
1
Climate Conservancy (2008)
PVC injection molding 2.87 kg CO
2e
kg
1
Industry data 2.0
Paper Labels 0.306e4.86 1.17 kg CO
2e
kg
1
Ecoinvent; Climate Conservancy (2008)
Adhesive 2.35 kg CO
2e
kg
1
Climate Conservancy (2008)
Solvents, organic, unspecified 2.31 kg CO
2e
kg
1
Ecoinvent
Printing ink 1.45 kg CO
2e
kg
1
USA Input Output database 98
Packaging LDPE film 1.74e2.6 2.6 kg CO
2e
kg
1
Ecoinvent
Kraft unbleached 100% rec.FAL 1.76e4.86 2.33 kg CO
2e
kg
1
Franklin USA 98
Waste disposal
Landfill
PE packaging waste 0.113e0.491 0.113 kg CO
2e
kg
1
Ecoinvent; BUWAL 250
Paper packaging waste 0.021e1.34 1.34 kg CO
2e
kg
1
BUWAL 250; Ecoinvent
Cardboard packaging waste 0.02e1.73 1.73 kg CO
2e
kg
1
BUWAL 250; Ecoinvent
Wood untreated 0.0857 0.086 kg CO
2e
kg
1
Ecoinvent
Urban and biological wastes 0.561e0.786 0.8 kg CO
2e
kg
1
Manfredi et al. (2009); BSI (2008b)
Disposal, zeolite or concrete, 5% water 0.00708 kg CO
2e
kg
1
Ecoinvent
Copper (inert) 0.000557 kg CO
2e
kg
1
ETH-ESU 96
Incineration
Digester sludges 0.0695 kg CO
2e
kg
1
Ecoinvent
Used lubricating oils or fuel wastes 3.4e3.76 3.5 kg CO
2e
kg
1
ADEME (2007); Kanokkantapong et al. (2009)
Waste oil 2.88 kg CO
2e
kg
1
ETH-ESU 96
Recycling
paper 0.0635-0 0.0635 kg CO
2e
kg
1
BUWAL 250; Ecoinvent
(continued on next page)
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 5
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

malting, the GHG emissions associated to corn grit production may
be roughly estimated as 0.746 kg CO
2e
kg
1
maize grits.
In accordance with Climate Conservancy (2008), hop agriculture
was regarded as characterized by the same impact categories
accounted for barley, this leading to an estimated value of GHG
emissions of about 2 kg CO
2e
kg
1
hops processed. In the major hop
growing regions, harvest is generally targeted when cones reach
approximately 23% dry matter, while hops are dried down to 8e12%
moisture for packaging and storage (Madden and Darby, 2012), this
yielding further 0.391 kg CO
2e
per kg of hops processed. In this case,
hop pellets were sourced in Germany, this resulting in an average
distance of 1509 km.
It was also assumed that barley and corn were cultivated on land
which had been used for agricultural purposes for longer than 20
years (PAS 2050: Section 5.5); therefore, the GHG emissions arising
from land use change were not considered.
2.5.2. Processing
There are several steps in the brewing process, these including
mashing, lautering, boiling, cooling, oxygenation, fermentation,
conditioning, yeast removal, kieselguhr filtering, PVPP stabilization,
and packaging. The lager beer under study was obtained with the
high-gravity brewing method (Eßlinger, 2009). To arrest biological
contamination, the chill haze-free lager beer was pasteurized using
the conventional tunnel pasteurization method to heat progres-
sively either glass bottles or aluminum cans up to 60

C for
10e20 min, or the flash pasteurization method to heat the product
up to circa 70

C for 60e90 s before its aseptic filling in steel kegs.
In the schematic brewing process flow sheet shown in Fig. 1, all
input materials and solid waste generated, as well as the needs for
electric energy (EE) to drive all process equipment, machines, and
refrigeration system (i.e., malted barley and corn grit conveyors,
millers, mixers, filter, pumps, compressors, etc.) and for thermal
energy (Q) during the mashing and wort boiling phases, were
pointed out.
By referring to the overall production of 1,160,835 hL of Birra
Peroni Srl in the reference time period examined, Tabl e 2 shows
the specific consumption yields for raw materials (malted barley,
maize grits, hop pellets); brewing coadjutants and processing
aids (i.e., oxygen; compressed air; calcium chloride and sulfate;
phosphoric acid at a mass fraction of 0.75; DE; PVPP); chemicals
for plant cleaning and PVPP regeneration (30% caustic soda, 53%
nitric acid, Oxonia active, etc.); and refrigerants (ammonia and
ethylene glycol). Table 2 also lists the specific formation yields
for spent grains and surplus yeast, i.e., the main by-products that
occur during brewing, these being regarded as an avoided pro-
duction of cattle feed. In particular, hop pellet consumption was
estimated on the basis of an average iso-
a
-acid mass fraction of
0.055, whereas the average raw protein mass fraction of spent
grains was assumed as equal to 0.08 (Platto Feeding Company,
2014). Moreover, Oxonia active (Ecolab Inc., St. Paul, MN, USA)
is an acidic colorless liquid sanitizer for food processing equip-
ment, composed of hydrogen peroxide (mass fraction: 0.275) and
peracetic acid (CH
3
eCOeOeOH, mass fraction: 0.058). No infor-
mation about the lubricant consumption was available, thus their
consumption per each functional unit was regarded as negligible
in agreement with other CF estimates (Moresi and Paone, 2012).
The weak wort, resulting from the final rinsing of spent grains
with hot water and having a low sugar content of 1e2

P(de-
grees Plato), and hot trub, separated from the hopped wort in the
centre of the whirlpool tank, represented about the 6.7% and
4.0% by volume of any wort batch (i.e., 120 0 hL), respectively. The
wastewaters had an average COD value of about 60 00 ppm (mg
O
2
L
1
) and were firstly submitted to anaerobic digestion, the
resulting liquid digestate being aerobically treated up to a final
COD value of 69 ppm and then disposed of in the municipal
sewer system. The resulting sludge was firstly anaerobically
treated and finally composted.
All the other materials used in minimum quantities (that is,
yeast starter cultures, minor chemicals and wastes, etc.) were not
included in the system boundaries since their potential influence
on the analysis results was assumed as negligible, being smaller
than 1% (PAS 2050: Section 6.3).
2.5.3. Packaging
Beer packaging varied with the format selected. In the reference
time period examined, the 88.9% of the overall Peroni beer was
packaged in glass bottles (the 66.6% of which having a volume of
66 cL and the remainder 22.3% of 33 cL), the 6.9% in 33-cL
aluminum cans and the residual 4.2% in 30-L steel kegs. More-
over, the 56% of the lager beer packaged in 33-cL glass bottles was
assembled in cartons, each one containing 24 33-cL bottles; while
the remainder in 3-packs and then in cartons, each one containing
8 clusters. Fig. 3 shows the block diagrams for their primary, sec-
ondary and tertiary packaging processes in order to identify all the
packaging materials and aids needed, the solid waste generated, as
well as the electric and thermal energy needs.
Beer packaging is carried out using counter-pressure fillers,
where any beer package is purged with CO
2
before being filled with
beer. The gaseous stream used by Birra Peroni Srl was mainly that
leaving the beer fermentors, while as little as the 5% of the overall
amount needed was of fossil origin. This fraction was assumed to be
totally dispersed in the Earth atmosphere. The utilization of the
carbon dioxide exiting the beer fermentors guarantees perfect
quality control, since residual oxygen in the commercial carbon
dioxide has a detrimental effect on beer flavor stability. The gaseous
streams leaving the fermentation tanks are firstly fed to a foam
separator and then stored in low-pressure gas storage balloons.
From these, the gas is firstly cleaned in a scrubber by counter flow
of water, compressed to one sixteenth of its original volume, dried
via molecular sieves, deodorized by activated carbon, and finally
condensed at 20

C and 18 bar.
Table 3 lists all the packaging materials and aids (i.e., glass
bottles, aluminum cans, steel kegs, crown and can open closures,
ball lock keg couplers, labels, adhesive, ink, cartons, cluster packs,
trays, stretch and shrink film, pallet, etc.) used to prepare the beer
Table 1 (continued )
Emission factor Minemax values Default value Unit Ref.
Plastics (incl. PE and excl. PVC) 0.332-0 0.332 kg CO
2e
kg
1
BUWAL 250; Ecoinvent
Wood 0kgCO
2e
kg
1
Ecoinvent
Glass 0.376 kg CO
2e
kg
1
BUWAL 250
aluminum 10.6 kg CO
2e
kg
1
BUWAL 250
Steel 1.69 kg CO
2e
kg
1
BUWAL 250
Byproduct uses
Animal feed production
(low protein) ≡equiv. spent grains þsurplus yeast
0.637 kg CO
2e
kg
1
LCA Food-DK
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e186
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

primary, secondary and tertiary packages. Particularly, the primary
packages consisted of 33- or 66-cL amber glass bottles, 33-cL
aluminum cans, or 30-L stainless steel kegs. The secondary pack-
ages were made of cartons, each one containing 15 66-cL bottles,
24 33-cL bottles, or 8 clusters (any of the latter assembling
333-cL bottles), or trays of 24 33-cL cans.
Table 4 shows the specific consumption of the packaging ma-
terials used to package 1 hL of lager beer in the different formats
examined.
2.5.4. Waste management
Fig. 4 shows the block flow diagram of solid waste and gaseous
effluent formation during the Birra Peroni packaging and pallet
management in the distribution centers. In particular, all wastes
arising from the beer making were disposed of as follows:
- glass bottles, aluminum cans, or stainless steel kegs rejected
during cleansing, sterilization and filling steps were collected
and recycled to glass, aluminum, or steel manufacturing,
respectively;
- body, neck and bar code labels discarded during primary pack-
aging, as well as pallet tapes and labels rejected during tertiary
packaging, were gathered and used as feedstock for recycled
paper;
- cartons and trays refused during secondary packaging were
amassed for recycling and reusing;
- stretch wrap films discarded during the secondary packaging of
cans or tertiary one of bottles and cans, as well as keg valves
during the primary packaging of kegs, were collected and
recycled to make other polyethylene products;
- wooden pallets, either discarded during tertiary packaging or
gathered for collection at the distribution centers, were repaired
and reused.
On the contrary, the fossil carbon dioxide used to pilot the beer
filling machines was assumed as totally released in the air.
Table 5 lists the different solid wastes formed during the
manufacturing process as referred to the functional unit. Each
packaging material waste can be extracted from the data listed in
Table 4.
2.5.5. Transport
The only transport modality for raw materials, processing aids
and detergents from their production sites to the Birra Peroni fac-
tory gate in Rome (Italy) was by road via the means of transport
listed in Table 6. Their corresponding emission factors were
dependent on the European emission standards and were sourced
from the software Simapr
o 2 (PR
e Consultants, Amersfoort, NL).
According to Birra Peroni (2012), the average emission factor for
any means of transport was estimated by assuming that the 30% of
all materials was transported by using Euro 5 means, while the
remainder 70% by Euro 3 ones.
The average distance traveled from any production site to the
brewery gate by all raw and packaging materials, processing aids,
and detergents, as well as any transport modality, is shown in
Tables 2 and 4.
As regarding the final product packaged in the formats under
study, such packages were generally delivered to numerous dis-
tribution centers located in quite all the provinces of Italy (see
Fig. 2). According to Birra Peroni (2012), about the 30% of the overall
production was delivered by means of heavy rigid trucks (HRT),
while the remainder 70% by articulated trucks (AT), their average
emission factors being listed in Table 6. Thus, the average emission
factor resulted to be equal to:
½0:3ð0:30:268 þ0:70:291Þþ0:7
ð0:30:154 þ0:70:168Þz0:2 kgCO
2e
ðMg kmÞ
1
:
Moreover, since 2009 Birra Peroni Srl has rationalized its dis-
tribution network so as to increase the average truck loading rate
and minimize the distance traveled by any format. According to the
data sourced by the Sales Office of Birra Peroni Srl, it was possible to
identify the sales volume of Birra Peroni in any of the formats under
study, as well as the average distance traveled to reach each dis-
tribution center using the web site http://www.viamichelin.it, as
shown in Table 7. These data referred to an overall sales volume of
950,163 hL and were regarded as valid for all the production ca-
pacity registered in the reference time period of this study. In
particular, for any tertiary package the GHG emissions were
calculated by accounting for the transport of a single pallet and,
Table 2
Specific consumption yields of raw materials, processing aids, brewing coadjutants, detergents, refrigerants, and by-products per hL of Peroni lager together with the transport
means used (see Table 6) and average distance traveled from their production site to the brewery gate.
Inventory Consumption yield Unit Means of transport Distance [km]
Raw materials and processing aids
Barley 14.36 kg hL
1
AT 236
Malted barley 10.77 kg hL
1
AT 33
Maize grits 4.71 kg hL
1
AT 608
Hop pellets 91.60 g hL
1
HRT 1509
Oxygen 1.43 g hL
1
RT 91
Compressed air 2.58 M
3
hL
1
ee
Calcium chloride 24.78 g hL
1
RT 210
Calcium sulfate 19.91 g hL
1
RT 210
Brewing coadjutants
Diatomaceous earth 0.112 kg hL
1
RT 110
PVPP 0.110 g hL
1
RT 445
Phosphoric acid (75% w/w) 14.21 g hL
1
RT 445
Detergents
Nitric acid (53% w/w) 17.22 g hL
1
RT 445
Caustic soda (30% w/w) 0.606 kg hL
1
RT 445
Oxonia active 17.055 g hL
1
RT 589
Trimeta LPC 40.256 g hL
1
RT 589
Refrigerants
Ammonia 0.052 g hL
1
RT 175
Ethylene glycol 0.956 g hL
1
RT 62
Byproducts
Spent grains 17.44 kg hL
1
RT 150
Surplus yeast 1.45 kg hL
1
RT 150
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 7
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

then, referred to 1 hL of the final product packaged in the formats
examined (Table 3).
According to Birra Peroni (2012), the 30-L stainless steel kegs are
generally reused 3.6 times per year for as long as 20 years. Thus, the
GHG emissions associated to their transport were estimated by
splitting the transport contribution of the new kegs from that of the
reused ones. The former was assessed by accounting for the average
distance traveled from the manufacturer to the brewery gate (i.e.,
1500 km) and the amount of kegs yearly replaced, this being equal
to 1/(3.6 20) ¼1/72-th of the keg consumption yield listed in
Fig. 3. Schematic diagram of the packaging process for Peroni lager in (a) 33- or 66-cL amber glass bottles; (b) 33-cL aluminum cans; (c) 30-L stainless steel kegs. All the iden-
tification items for the input and output materials are reported in the List of symbols.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e188
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Table 4. This was line with the LCA report by the Chicago
Manufacturing Center (2009), having any steel keg an expected
life of 100 uses. On the contrary, the keg return system contribution
was estimated by referring to the average distance traveled by the
filled kegs, that is 143.6 km (Table 7).
Once the lager beer had been delivered on wooden pallets, the
distributor or trader collected the empty pallets to allocate them
back to the original producer, where defected pallets were
repaired according to specification and made available again to
the brewery. Since the pallet operators serving Birra Peroni Srl
were located at distances in the range of 25e335 km, the average
distance traveled by the repaired pallets was estimated as equal
to about 150 km by accounting for both the mass of pallets
transported and relative transport distance. Altogether, the
average distance traveled by the empty pallets was fixed at circa
300 km, the average distance traveled by the final product
independently of its primary packaging being of about 153 km
(Table 7).
The impact of solid waste transportation was assessed on the
basis of an average distance between the brewery gate and each
disposal site of about 400 km, in agreement with previous assess-
ment (Moresi and Paone, 2012) by means of lightemedium rigid
trucks, the average emission factor of which being equal to 0.65 kg
CO
2e
(Mg km)
1
, as shown in Table 6. Finally, the main byproducts
of brewing, that is spent grains and surplus yeast (Table 2), were
used as cattle feed and transported to about 150 km from the
brewery.
2.5.6. Energy sources
The energy sources used by Birra Peroni Srl to manufacture the
pale lager packed in the aforementioned formats were of the
electric or thermal (i.e., methane and diesel) type.
Electric energy was used to pilot process machines and equip-
ment, to refrigerate process streams and final product during the
production and packaging phases, as well as to run plant utilities
and electric forklifts. Their specific consumption needs were
collected from the industrial brewery Birra Peroni Srl (Rome, Italy),
and listed in Table 8. By referring to the Italian thermoelectric
production in 2011, the overall electric energy dissipation rate was
about 3% (ISPRA, 2012); thus, the overall consumption of electric
energy was assumed as equal to the 97% of that effectively absorbed
by the grid network.
Methane-fired boilers were used to generate either the steam or
hot water needed during the phases of mashing, lautering, and
wort boiling, or sanitization of pipes, equipment and filling ma-
chines. A fraction of the methane required consisted of digester
methane gas. All thermal energy consumption needs, expressed in
MJ hL
1
, are shown in Table 8. The overall thermal energy re-
quirements during the brewing process were estimated on the
basis of an average steam boiler efficiency of 88%, whereas the
overall volumetric consumption of methane was guessed on ac-
count of its lower heating value of 37.76 MJ m
3
(STP). Thus, the
only GHG emissions deriving from the burning of fossil methane
were accounted for.
Table 3
Mass of any component of primary, secondary and tertiary packagestogether with their corresponding overall masses and GHG emissions associated to the transport of a pallet
or a functional unit (FU) of beer packaged in different formats.
Format Amber glass bottle Aluminum can Stainless steel keg Unit
Primary packaging
Volume 0.66 0.33 0.33 30 L
Mass 290 185 12.3 9600 g
Diameter height 74.8 267.6 61 213.2 66.2 115.2 408 510 mm
Adhesive for labels 0.293 0.154 ee g
Ink diluent 0.0015 0.0007 0.0000 0.0023 g
Ink 0.0002 0.0001 0.0000 0.0017 g
Crown closure 1.99 1.99 ee g
Can open closure ee 3.8 eg
Plastic liner for crowns 0.11 0.11 ee g
Body label 0.63 0.29 ee g
Neck Label 0.48 0.40 ee g
Plastic ball lock keg coupler ee e5g
Primary packaging overall mass 0.9568 0.5196 0.3478 39.755 g
Secondary packaging Carton Carton Cluster Pack Tray e
No. of primary packages 15 24 8 24 ee
Length width depth 383 227 273 369 253 218 369 253 218 287 523 58 emm
Carton or tray mass 282.76 254.5 240.5 91.08 eg
3-bottle cluster mass ee27.91 ee g
Adhesive for cartons or trays 16 15 15 ee g
Stretch and shrink film eee20 eg
Beer volume per carton 9.9 7.92 7.92 7.92 eL
Secondary packaging overall mass 14.651 12.740 12.934 8.472 ekg
Tertiary packaging
No. secondary or primary packages 11 9 9 9 6 e
No. layer per pallet 6 8 8 13 4 e
Height of pallet 1.782 1.888 1.888 1.640 2.290 m
Bar code label 1 1.08 1 1.08 1 1.08 1 1.08 5 1.08 g
Pallet label 23.108 2 3.108 2 3.108 2 3.108 eg
Pallet tape 2 0.29 2 0.29 2 0.29 2 0.29 2 0.29 g
Stretch &shrink film 483 385 385 555.6 eg
Pallet mass 18 18 18 18 4 22 kg
Beer volume per pallet 653.4 570.2 570.2 926.6 720 L
Tertiary packaging overall mass 985.4 935.7 949.6 1009.8 1042.1 kg
GHG emission per pallet 28.9 28.2 37.0 34.8 29.9 kg CO
2e
GHG emissions per FU 4.42 4.94 6.48 3.76 4.16 kg CO
2e
hL
¡1
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 9
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Two out of the nine forklifts dedicated to pallet transport within
the brewery were of the diesel type, the other 7 forklifts being of
the electric one. No information about their diesel oil consumption
was available, thus such consumption per each functional unit was
regarded as negligible, in agreement with previous estimates
(Moresi and Paone, 2012).
2.5.7. Refrigerant losses
Refrigerant losses in the life cycle of beer take place during the
manufacturing process and use phase (i.e., refrigerated storage of
beer at pubs, bars, and home), the latter being disregarded in this
study. In the brewery examined a safety refrigerant as ammonia,
known as R 717, was used. Although it is synthetically produced,
ammonia is considered a natural refrigerant because it occurs in
nature's material cycles. Moreover, such a colorless gas has no
ozone depletion potential (ODP ¼0) and no direct global warming
potential (GWP ¼0). Its indirect greenhouse effect contribution is
also very limited owing to its high energy efficiency (its latent heat
of evaporation being equal to 1262 J g
1
)(Eurammon, 2011). Its
characteristic odor makes its leakage in the air easy to detect, thus
no leakage was accounted for.
An aqueous solution of ethylene glycol was used as secondary
refrigerant, this being cooled via the primary refrigerant system
and circulated for use during wort fermentation and beer filtration
in the temperature range of 0e12

C. The loss of such refrigerant
per each hL of beer is given in Table 2 and was assumed to
contribute to the organic load of wastewaters, these being disposed
off in accordance with the antipollution norms by combining
anaerobic pre-treatment and aerobic post-treatment.
2.5.8. Life-cycle impact assessment
To assess the Carbon Footprint of the functional unit chosen (1 hL
of packed beer), all GHG emissions associated to the production of
raw and packaging materials, processing aids and detergents, to
their transportation and that of the final product and processing
wastes, to the consumption of thermal and electric energy sources,
were estimated by multiplying the entity
J
i
of any activity
parameter (expressed in mass, energy, massekm basis) by its cor-
responding emission factor EF
i
:
CF ¼X
i
ðJ
i
EF
i
Þ(1)
Since all activity data were referred to the aforementioned
functional unit, the resulting carbon footprint was expressed as kg
CO
2e
hL
1
.
3. Results
By referring to the default option for the emission factors listed
in Table 1, the impact assessment phase resulted in the calculation of
the carbon footprint (CF) of 1 hL of lager beer packaged in 66-cL
glass bottles (~57 kg CO
2e
), 33-cL glass bottles assembled in
either cardboards (~67 kg CO
2e
) or cluster packs (74 kg CO
2e
), 33-cL
Table 4
Specific consumption yield for all the packaging materials used to package 1 hL of Peroni lager in the different formats examined together with their transport modality and
average distance traveledfrom their production site to the brewery gate, as well as the overall GHG emissions associated to the transport of the packaging materials allocated in
the functional unit (FU).
Packaging materials Packaging formats Unit Means of transport Distance [km]
Primary packaging 66-cL GB 33-cL GB 33-cL GB 33-cL ALC 30-L SSK
Secondary packaging Carton Carton Cluster þCarton Tray e
Beer volume used 1.022 1.022 1.022 1.022 1.022 hL hL
1
No. of bar code labels 0.162 0.189 0.189 0.189 0.667 hL
1
Mass of bar code labels 0.17 0.20 0.20 0.12 3.6 g hL
1
RT 608
Carton adhesive 37.37 45.45 45.45 22.73 0.0 g hL
1
RT 599
Label adhesive 44.44 46.72 46.72 0.0 0.0 g hL
1
RT 599
Ink diluent 0.23 0.20 0.20 0.71 0.0 g hL
1
RT 577
Ink 0.03 0.03 0.03 0.51 0.0 g hL
1
RT 577
No. of bottles, cans or kegs 152.12 305.76 305.76 304.24 3.33 hL
1
Mass of bottles 44.12 56.57 56.57 eekg hL
1
AT 166
Mass of cans eee 3.74 ekg hL
1
AT 470
Mass of new kegs eee e 0.44 kg hL
1
AT 1500
Mass of reused kegs eee e 32.00 kg hL
1
AT 144
No. of crowns or can open closures 152.73 305.45 305.45 306.67 ehL
1
Mass of crowns 320.7 641.5 641.5 eeghL
1
HRT 300
Mass of can open closures eee 1.17 0 kg hL
1
HRT 470
No. of plastic ball log keg coupler eee e 3.33 hL
1
Mass of plastic ball log keg coupler eee 16.67 g hL
1
RT 1500
No. of paper body and neck labels 152.27 304.24 304.24 eehL
1
Mass of paper body labels 95.93 70.58 70.58 eeghL
1
HRT 1370
Mass of paper neck labels 73.09 121.70 121.70 eeghL
1
RT 1370
No. of cluster multi-packs ee101.01 eehL
1
Mass of cluster multi-packs ee2.82 eekg hL
1
AT 380
No. of cartons and trays 10.16 12.69 12.69 12.69 ehL
1
Mass of cartons 2.87 3.23 3.052 eekg hL
1
AT 380
Mass of trays eee 1.16 0.00 kg hL
1
AT 421
Stretch and Shrink Film per trays eee 252.5 0 g hL
1
RT 502
No of pallets 0.15 0.16 0.16 0.11 0.57 hL
1
Mass of pallets 2.73 2.95 2.95 2.05 12.47 kg hL
1
AT 300
Mass of pallet labels 2.01 2.35 2.35 1.41 eghL
1
RT 608
Pallet tape 0.09 0.10 0.10 0.07 0.33 g hL
1
RT 608
Wrap film per pallet 73.23 63.13 63.13 63.13 eghL
1
HRT 502
Carbon dioxide
a
0.63 0.62 0.62 0.48 1.32 kg hL
1
RT 85
Overall GHG emissions per FU 1.69 2.12 2.29 0.73 1.50 kg CO
2e
hL
1
a
It refers to the CO
2
of fossil origin used to package the clarified and stabilized beer; it represents just the 5% of the overall amount needed, the remainder being recovered
from the gaseous streams leaving the beer fermentors.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e1810
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Fig. 4. Block flow diagram of solid waste and gaseous effluent formation during Peroni lager packaging and pallet management in tertiary packaging and distribution centers. All
symbols used are reported in the List of symbols.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 11
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Table 5
Specific formation of solid wastes associated to the production of a functional unit (FU) of Peroni lager,and overallGHG emissions associated to their transport under the
assumptions reported in the text.
Solid waste type Consumption yield [g hL
1
] Specific GHG emissions [kg CO
2e
hL
1
]
Urban wastes 19.2 0.0050
Spent DE sludges 336.0 0.0874
Digester sludges 183.7 0.0478
Waste fuels 202.1 0.0526
Paper and cardboard packaging waste 218.9 0.0569
Plastic packaging waste 169.4 0.0441
Wood packaging waste 39.5 0.0103
Waste toner and printer cartridges 159.2 0.0414
Glass packaging waste 143.2 0.0373
Hydrocarbon-rich wastes 13.3 0.0035
Concrete structures to be demolished 5.2 0.0014
Aluminum waste 7.4 0.0019
Iron and steel waste 15.4 0.0040
Waste copper wire 7.0 0.0018
Spent grains 17,442.6 1.7017
Surplus yeast 1449.1 0.1433
Overall GHG emissions per FU 2.2420
Table 6
Means of transport used to delivery all the materials involved in the brewing process under study together with their corresponding European GHG emission standards on a
kmemass basis as sourced from Simapr
o 7.2 v.2 (Pr
e Consultants, Amersfoort, NL).
Means of transport Load capacity [Mg] European emission standards [kg CO
2e
(Mg km)
1
] Average emission factor [kg CO
2e
(Mg km)
1
]
Euro 5 Euro 4 Euro 3
Articulated truck (AT) 16e32 0.154 0.152 0.168 0.164
Heavy rigid truck (HRT) 7.5e16 0.268 0.265 0.291 0.284
Lightemedium rigid truck (RT) 3.5e7.5 0.635 0.626 0.657 0.650
Table 7
Sales volume of Peroni lager sold in different formats in the time period examined (i.e., glass bottles, GB; aluminum cans, ALC; stainless steel kegs, SSK) and corresponding
average distance traveled from the brewery gate to the distribution centers.
Primary package Secondary package Volume [hL] Average distance traveled [km]
66-cL GB 15 pieces per carton 672,178 146.5
33-cL GB 24 pieces per carton 128,998 150.6
33-cL GB 8 clusters per carton 100,968 194.6
33-cL ALC 24 pieces per tray 17,972 172.5
30-L SSK e30,047 143.6
Overall packages 950,163 152.6
Table 8
Consumption yields for electric (EE) and thermal (Q) energy; well (AP), tap (AA) and process (APR) water referred to 1 hL of Peroni lager produced and packed in the formats
examined, together with the partial and overall utility consumption yield (UCY) per hL of beer produced.
Utility consumption yield (UCY) EE [kWh hL
1
] Q [MJ hL
1
]AP[LhL
1
] APR [L hL
1
]AA[LhL
1
]
Beer processing (BRP)
Well water distribution 0.021 eee e
Process water production eeee242.3
Thermal energy generation 0.093 e2.9 ee
Refrigeration 2.237 e116 ee
Compressed air 0.602 e2.1 ee
CO
2
0.676 e10.5 ee
Wort production 0.638 24.5 0.0 128.2 e
Beer making 1.359 8.9 50.2 53.2 12.5
Forklifts 0.158 eee e
General Plant Utilities 0.739 3.0 e0.30 8.8
Biogas e10.0 ee e
Energy dissipation 0.202 6.3 ee e
Subtotal UCY referred to BRP 6.73 52.72 77.2 181.6 263.7
Beer packaging (PP)
Packaging Line for 0.66-cL GBs 1.317 24.9 43.8 28.5 e
Packaging Line for 0.33-cL GBs 2.516 17.7 66.2 6.8 e
Packaging Line for 0.33-cL ALCs 3.035 11.7 5.5 122.0 7.7
Packaging Line for 30-L SSKs 1.604 20.9 e177.4 e
Overall UCY referred to BRP and PP 8.61 73.7 118.7 226.9 264.5
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e1812
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

aluminum cans (~69 kg CO
2e
), or 30-L stainless steel kegs (~25 kg
CO
2e
), as shown in Fig. 5.
The percentage contribution of the different life cycle phases to
CF is illustrated in Fig. 6 and depends on the packaging format of
choice. According to the data shown in Table 9, minimum and
maximum percentage impacts of raw materials and processing aids
(RPM), as well as brewing processing (BRP) and packaging (PP),
were associated to the beer packaged in a 3-pack of 33-cL glass
bottles and 30-L stainless steel kegs, respectively. As concerning the
packaging materials (PM) or transportation (TR), the corresponding
contribution varied from 48 to 58 or from 10 to 14 % of CF for the
beer bottled or canned to as little as 5 or 25% of CF in the case of
kegs, owing to their high reuse coefficient. The contribution of
wastes generated in the brewery was of the order of 1e2 % of CF for
all the formats examined (Table 9).
The GHG emissions associated to raw materials, brewing pro-
cessing and packaging amounted to ~25.3 kg CO
2e
hL
1
indepen-
dently of the package used (Table 9). On the contrary, the
66.6
74.4
69.3
24.9
56.8
0
20
40
60
80
100
120
GB66-cL GB33-cL GBC33-cL ALC33-cL SSK30-L
CF [kg CO2e hL-1]
Fig. 5. Carbon footprint (CF) of a functional unit (1 hL) of Peroni lager packed in 66- or 33-cL glass bottles (GB), assembled in cartons either as loose or multipack (C) bottles, 33-cL Al
cans (ALC), or 30-L stainless steel kegs (SSK).
0
20
40
60
80
RPM BPR PM PP TR WD
%
GB66- cL
GB33- cL
GBC33 -cL
ALC33-cL
SSK30-L
Fig. 6. Percentage contribution of the different life cycle phases to the carbon footprint of a functional unit (1 hL) of Peroni lager packed in 66- or 33-cL glass bottles (GB), assembled
in cartons either as loose or multipack (C) bottles, 33-cL Al cans (ALC), or 30-L stainless steel kegs (SSK): RPM, raw materials and processing aids; BPR, brewing processing; PM,
packaging materials; PP, packaging; TR, transportation; WD, waste disposal.
Table 9
Percentage contribution of the different life cycle phases to the carbon footprint of a functional unit (1 hL) of Peroni pale lager packed in 66- or 33-cL glass bottles (GB), the latter
being assembled either loose or in cluster (C), 33-cL Al cans (ALC), or 30-L stainless steel kegs (SSK).
Life cycle phases Carbon footprint for different Packaging formats
[kg CO
2e
hL
1
] [%] [kg CO
2e
hL
1
] [%] [kg CO
2e
hL
1
] [%] [kg CO
2e
hL
1
] [%] [kg CO
2e
hL
1
] [%]
Final product primary packaging 66-cL GB 33-cL GB 33-cL GBC 33-cL ALC 30-L SSK
Raw materials &processing aids (RPM) 16.88 24 16.88 21 16.88 20 16.88 21 16.88 46
Brewing processing (BRP) 6.26 9 6.26 8 6.26 7 6.26 8 6.26 17
Packaging materials (PM) 33.33 48 42.19 54 48.34 56 47.55 58 1.86 5
Packaging (PP) 2.15 3 2.14 3 2.14 2 2.07 3 2.15 6
Transportation (TR) 9.71 14 10.67 14 12.37 14 8.09 10 9.26 25
Waste disposal (WD) 0.58 1 0.58 1 0.58 1 0.57 1 0.61 2
Beer production excluding byproducts credits 68.91 100 78.71 100 86.57 100 81.42 100 37.02 100
Byproduct credits (BPC) 12.16 12.16 12.16 12.16 12.16
Beer production including byproducts credits 56.76 66.55 74.41 69.26 24.86
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 13
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

contribution of packaging materials was minimum in the case of
kegs (~1.86 kg CO
2e
hL
1
), in virtue of their expected life of 72 uses,
and maximum in the case of the 33-cL glass bottle cluster packs
(~48.3 kg CO
2e
hL
1
). The latter was slightly greater than that of 33-
cL Al cans (~47.6 kg CO
2e
hL
1
). In the case of beer packed in 33-cL
aluminum cans, the contribution of packaging materials to the
overall CF was about the double of that of raw materials, processing
aids and brewing, and not comparable to the aforementioned
items, as estimated by Pasqualino et al. (2011). The contribution of
transportation varied from as low as ~8.1 kg CO
2e
hL
1
to as high as
~12.4 kg CO
2e
hL
1
in the case of Al cans or 3-packs, respectively. As
shown in Table 9, utilization of beer co-products (i.e., spent grains,
and surplus yeast) as feed definitively reduced the overall GHG
emissions by ~12.1 kg CO
2e
hL
1
.
By accounting for the sales volumes of the different formats
examined, it was also assessed an overall CF value of ~59.2 kg CO
2e
per hL of lager beer allocated to any Birra Peroni distribution center
in any of the formats accounted for.
3.1. Sensitivity analysis
Owing to the numerous assumptions, the above calculated
carbon footprints cannot in principle be directly compared to those
of other lager beers, the latter differing one from another not only
for production methods and recipes, but also for the use of un-
known emission factors. To improve the scientific value of this CF
assessment exercise, it was decided to resort to transparent data,
such as those shown in Tables 1e8, and to study how the uncer-
tainty in the output of the LCA model of CF, defined by Eq. (1), can
be apportioned to different sources of uncertainty in its input
variables, especially in the emission factors EF
i
of any activity
parameter.
Sensitivity may be measured by monitoring changes in the
dependent variable of choice (i.e., CF in this specific case) by
differentiating CF with respect to the generic i-th independent
variable (EF
i
) while keeping all the other emission factors (EF
j
)
constant for j si. Alternatively, it would be possible to change one-
emission factor-at-a-time (EF
i
), while keeping all the others (EF
jsi
)
at their baseline (nominal) values, to observe what effect this would
produce on CF. In this way, it is possible to compare the results,
having been all effects computed with reference to the same central
point in space, and to identify the model inputs having no or limited
effect on CF. Despite its simplicity, such approach does not fully
explore the input space, since it does not account for the simulta-
neous variation of the input variables. According to Saltelli et al.
(2006), this approach is illicit and unjustified, unless the model
under study is proved to be linear, as the simple linear model
described by Eq. (1).
Therefore, to provide perspectives on the key drivers to the CF of
the pale lager under study, the sensitivity of the overall CF of 1 hL of
pale lager packed in all the formats examined was assessed by
changing the emission factor (EF
i
) of a given activity by ±50% with
respect to the default condition (Table 1). More specifically, the
following activities were taken into account: (i) barley or corn
grown locally or globally using (ii) an organic or a conventional
agriculture method; (iii) glass bottles or aluminum cans with
different recycled contents; (iv) electricity generated by more or
less renewable sources; (v) thermal energy from methane deriving
from different combination of fossil and biogenic sources; (vi)
differently combined modes of final product transportation by road
and railway. The main results of such a sensitivity analysis are
shown in Fig. 7.
The range of variation for CF appeared to be smaller than ±3%
within a ±50% range of variation for the distance traveled by barley,
differently-recycled aluminum cans, electricity generated by
differently combined fossil and renewable sources, and corn of
different agriculture methods. The variation of CF was of the order
of ±4% provided that the emission factors for the means of trans-
port used to dispatch the final product to the distribution centers,
as well as for methane of more or less fossil origin, exhibited a ±50%
variation with respect to the default case. CF was more sensitive to
changes in the emission factors for glass bottles and barley. In
particular, if they were reduced by 50%, CF accordingly exhibited
about a 20 or 10% reduction with respect to the basic case,
respectively.
More specifically, as shown in Table 10, the use of Italy-grown
organic barley instead of conventional one (their emission factors
being equal to 0.545 or 1.345 kg CO
2e
kg
1
, as listed in Table 1)
would lessen CF by about 11%, while the use of barley, organically
or conventionally cultivated at an average distance from the malt-
house of 1500 km, would lower or increase CF by about 6orþ9%,
respectively. The use of electricity produced only by fossil fuels or
photovoltaic sources (their corresponding emission factors
reducing from 0.545 to 0.055 kg CO
2e
kWh
1
) affected the overall
CF of the pale lager beer under study by approximately þ2or5%
with respect to the corresponding default value, respectively.
Finally, the change in the final product transportation mode from
road to railway (their relative emission factors reducing from 0.2 to
0.0393 kg CO
2e
Mg
1
km
1
) would lessen CF by about 6%.
Thus, glass bottle production and barley cultivation resulted to
be the controlling life cycle phases of the pale lager examined.
4. Discussion of results
Table 11 summarizes the main material, water, and energy
consumption yields, as well as waste production ones, relative to
the operation of the Birra Peroni Srl brewery, as compared to the
typical and average (within brackets) ones for some European
breweries (Donoghue et al., 2012; IFC, 2007; Olajire, 2012; Sturm
et al., 2013; UNEP, 1996).
According to Olajire (2012), a well-run brewery would use from
8 to 12 kWh electricity, ~5 hL water, and ~150 MJ fuel energy per hL
of beer produced. In the circumstances, the Birra Peroni brewery
-24
-18
-12
-6
0
6
12
18
24
-50 -25 0 25 50
Variatio n EF
i
[%]
Variation CF
[%]
Fig. 7. Effect of the percentage variation of the emission factor (EF
i
) for malted barley
(-), barley production site (:), maize grits (
▵
), glass bottles (B), aluminum cans
(▫), electric (C) and thermal (A) energy, or means of transport of final product (>)
on the percentage variation of the carbon footprint of a functional unit (1 hL) of Peroni
lager packed in all the formats examined with respect to the basic case.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e1814
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

may be classified as a brewery with low consumption figures, also
because the process effluent is firstly submitted to anaerobic
digestion, the recovered biogas being used for process heating. The
registered entity of ~10 MJ hL
1
represented the (13e16) % of the
overall brewery thermal energy needs. Water consumption was
near to 4 L per L of beer produced, this being less that the figure
issued by Miller S.A.B. (4.56 L L
1
) and InBev (4.32 L L
1
)(Miller
S.A.B., 2011). Actually, water consumption for modern breweries
varies from 3.5 to 10 L L
1
of beer produced, but in old micro-
breweries it can be as high as 19 L L
1
(Sturm et al., 2013). The
recovery of spent yeast as a feeding supplement appeared to be less
than that usually registered, probably because of its re-use as
production yeast for at least 4 fermentation cycles. The greatly
reduced specific PVPP consumption yield was attributed to the
brewery choice of using the regenerable PVPP type. Even, the hop
pellet consumption yield was quite low, but this was related to the
recipe formulation used by Birra Peroni Srl.
As concerning the carbon footprint of beer, it depends on how
much of the life cycle is included. According to Saxe (2010), the
value for the total life cycle varies from 80 to 150 kg CO
2e
per hL of
beer; but, excluding the user phase, it lessens to as low as 40 kg
CO
2e
hL
1
. Recently, Lalonde et al. (2013) provide an Environmental
Product Declaration for a few ales packaged in 12-oz glass bottles.
In particular, the estimated CF values ranged from 48 to 116 or from
144 to 175 kg CO
2e
hL
1
provided that the use phase was excluded
or included. By referring to a Chilean small-sized brewery (1500 hL/
year), the CF of a lager beer packed in 33-cL glass bottles from
cradle to retail was equal to 178 ±49 kg CO
2e
hL
1
(Mu~
noz et al.,
2012). On the contrary, Berners-Lee (2010) estimated a CF of 53
or 88 kg CO
2e
per hL of beer produced locally or imported from
abroad, if consumed in a pub. These data are quite lower than those
assessed for Tuborg
®
beer (EPD, 2011a), that amounted to 106 or
149 kg CO
2e
per hL of beer barreled in plastic or steel kegs, and to
190 k g CO
2e
per hL of beer kept in glass bottles. By referring to the
different GHG burden of the main commercial Birra Peroni formats
(Table 9), the allocation of 1 hL of beer packed in 66-or 33-cL glass
bottles (the latter assembled loosely in cardboards or in cluster
packs), 33-cL cans or 30-L steel kegs resulted in the emission of ~57,
67, 74, 69 or 25 kg CO
2e
, respectively. Such CF values were smaller
than the aforementioned ones (EPD, 2011a), probably because of
the greater production scale and shorter distribution chain of Birra
Peroni brewery, as well as for the slighter glass bottles used (190 vs.
200 g, respectively).
According to the CF values estimated here, the overall impact of
beer consumption in Italy, equaling 29.2 L per capita in 2013
(Assobirra, 2013), would represent from 0.1 to 0.3 % of the overall
Italian direct GHG emissions (458.2 Tg CO
2e
), including net GHG
emissions adsorbed from land use, land-use change and forestry, in
2011 (ISPRA, 2013).
The sensitivity analysis on CF pointed out the primary effects of
glass bottle production and then barley cultivation in agreement
with previous studies by Cordella et al. (2008), Koroneos et al.
(2005), and Talve (2001). In particular, the packaging materials
represented from 48 to 58% of the GHG burden in the case of glass
bottles and aluminum cans, respectively; while the use of stainless
steel kegs, owing to their high reusing coefficient, limited its impact
to as little as the 5% of CF. On the contrary, the high mass of the
empty keg (i.e., 9.6 kg) enhanced the contribution of its transport to
the 25% of CF, this being instead limited to 14% or 10% in the case of
glass bottles or Al cans.
The environmental impact of different reuse percentages for
glass beer bottles was assessed by Mata and Costa (2001). The
advantages of the use of returnable bottles over that of non-
returnable ones increase with the number of cycles performed by
the returnable bottles. In the case of a 50% reuse (i.e. the same
number of returnable and non-returnable bottles), the contribution
Table 10
Effect of different parameters on the carbon footprint (CF) of a functional unit (1 hL) of Peroni lager packed in 66- or 33-cL glass bottles (GB), the latter being assembled either
loose or in cluster (C), 33-cL Al cans (ALC), or 30-L stainless steel kegs (SSK).
Parameter CF
66-cL GB
[kg CO
2e
hL
1
]
33-cL GB
[kg CO
2e
hL
1
]
33-cL GBC
[kg CO
2e
hL
1
]
33-cL ALC
[kg CO
2e
hL
1
]
30-L SSK
[kg CO
2e
hL
1
]
All formats
[kg CO
2e
hL
1
]
Italy-grown barley 56.76 66.55 74.41 69.26 24.86 59.23
Low impact barley grown in Italy 50.32 60.12 67.98 62.83 18.43 52.80
Low impact barley grown abroad 53.30 63.09 70.95 65.80 21.40 55.77
High impact barley grown abroad 61.91 71.71 79.57 74.42 30.02 64.39
Electric energy from fossil fuels 57.98 67.96 75.81 70.74 26.13 60.52
Photovoltaic electric energy 54.04 63.43 71.29 65.96 22.05 56.38
Rail transportation 53.21 62.58 69.20 66.24 21.52 55.51
Table 11
Specific consumption yields of raw materials, processing aids, thermal and electric energy, detergents, and water, as well as generation of byproducts and methane, relative to
the Birra Peroni brewery (Rome, Italy) and main European breweries. The data within brackets refer to the average values.
Specific consumption yield Birra Peroni Srl European breweries UdM Ref.
Malted barley 10.8 15e18 kg hL
1
UNEP (1996)
Corn grits 4.7 ekg hL
1
This work
Hop pellets 91.6 260 g hL
1
Assobirra (2013)
Diatomaceous earth 112 80e570 (255) g hL
1
IFC (2007); UNEP (1996)
PVPP 0.1 20e40 g hL
1
Gopal and Rehmanji (2000)
Caustic soda (30% w/w) 0.6 0.39e1.07 (0.7) kg hL
1
UNEP (1996)
Carbon dioxide 619e1320 830e3060 (1830) g hL
1
UNEP (1996)
Thermal energy 64e78 150e350 (110) MJ hL
1
Olajire (2012); Sturm et al. (2013); UNEP (1996)
Biogas generated 10 3.0e3.3 MJ hL
1
Donoghue et al. (2012); IFC (2007)
Electric energy 8.3e9.8 8e20 (12.7) kWh hL
1
IFC (2007); Olajire (2012); Sturm et al. (2013); UNEP (1996)
Water 3.4e4.03 5e20 (4.9) hL hL
1
Olajire (2012); Sturm et al. (2013); UNEP (1996)
Spent grains 17.4 14-19 (17) kg hL
1
IFC (2007); UNEP (1996)
Surplus yeast 1.4 2-4 (3) kg hL
1
IFC (2007); UNEP (1996)
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 15
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

of returnable bottles to global warming, acidification, photochem-
ical ozone creation, critical air and water volume, human toxicity,
energy and raw-material consumption was found to be smaller
than that of the non-returnable bottles after the second reuse. On
the contrary, the contribution of returnable bottles to eutrophica-
tion, ozone depletion, solid waste, water and auxiliary material
consumption was larger even after several reuses (Mata and Costa,
2001). Thus, the optimal reuse percentage should be identified by
accounting not only for the environmental, but also for the eco-
nomic, technological and social implications of the different alter-
native distributions of beer in returnable or non-returnable bottles.
The use of the novel PET bottles enriched with nanoclays,
manufactured by Nanocor
®
(Lan, 2007), would not only extend the
beer shelf-life up to 30 weeks thanks to their high barriers to CO
2
and O
2
migration (Senturk et al., 2013; Smith, 2008), but it would
also reduce either the primary packaging mass from 185-290 g,
typical of a generic glass bottle, to ~30 g, or the packaging material
emission factor from ~9 kg CO
2e
kg
1
for aluminum cans to 3e4kg
CO
2e
kg
1
. The now commercially available polymereclay nano-
composite (PCNC) bottles have become popular with beverage
manufacturers, such as Miller Brewing Company (Senturk et al.,
2013). Moreover, in the case of Tuborg
®
beer the use of plastic
drums (having a unitary mass of 290 g; and a capacity of 20 L)
resulted in about 70% less GHG emissions (EPD, 2011a).
The possibility of resorting to a novel enzymatic technique to
replace conventional malting in beer production, its climate ben-
efits amounting to ~8.4 g CO
2e
per 33-cL can of beer (Kløverpris
et al., 2009), would save ~3.6% of the carbon footprint of the Per-
oni lager packaged in Al-cans, this equaling to about one third of the
savings achievable by resorting to local organic barley.
It would also possible to reduce by ~70% the global warming
potential for the current industrial beer DE-filtration, regenerable
PVPP stabilization and pasteurization procedures by resorting to a
novel combined pale lager precentrifugation, PVPP stabilization,
cartridge filtration and final ceramic tubular crossflow micro-
filtration procedure, as reported by Cimini and Moresi (2015).
Finally, there is a growing interest towards the social sustain-
ability of beverage consumption (Ali et al., 2010). According to the
product category rules for beer from malt (PCR, 2013), the use
phase scenario should be included in the life-cycle analysis of the
product. For instance, Watson (2008) assessed that the highest
environmental impact of beer life cycle was due to the use phase,
this being mainly affected by energy consumption for refrigeration
and dispensing, consumer displacement, as well as disposal of solid
wastes and wastewaters generated by the consumer itself. In
particular, consumer displacement, mainly attributed to car use,
seemed to be the hottest spot, being strongly linked to fossil fuel
combustion, both as resource consumption and combustion.
Effective action to reduce environmental burden should be
considered at consumer level and a few suggestions, such drinking
draught beer instead of bottled one, or reducing car use to reach
dispensing location, were given by Normand et al. (2012) and
Watson (2008). In fact, by comparing the different GHG burden of
the main commercial Peroni pale formats, consuming 33 cL of beer
from a glass bottle, a can, or a keg in a pub would involve the
emission of 246, 229 or 82 g CO
2e
, respectively. Such CF values were
by far smaller than that (589 ±161 g C O
2e
) estimated by Mu~
noz
et al. (2012) for a small-scale brewery. Altogether, such data still
suggest that consumers might choose a more responsible con-
sumption of draught beer in a local pub. Furthermore, draught beer
might be dispensed from beer pipelines (linking the pub to the
storage tanks of a brewery in close proximity) rather than from
steel or plastic kegs. The distribution of the latter severely affects
local traffic, especially in historic sites, such as Bruges in Belgium
(AFP, 2014), or during beer festivals, such as the Oktoberfest in
Munich in Germany (Becker, 2014). Unfortunately, the present-day
major consumption of beer is by far from glass bottles. Indeed, the
sales for Peroni lager barreled in steel kegs were as little as the 3.2%
of the overall ones (Table 7).
5. Conclusions
By referring to fully transparent primary and secondary data,
the estimated carbon footprint (CF) of Peroni pale lager significantly
varied with the package used. In particular, it was equal to ~57, 67,
74, 6 9 or 2 5 kg C O
2e
per hL of lager beer packed in 66-cL glass
bottles, 33-cL glass bottles assembled in cardboards or cluster
packs, 33-cL aluminum cans or 30-L stainless steel kegs, respec-
tively. Such a difference in CF was due to the different contribution
of packaging materials and transportation. The former was mini-
mum in the case of kegs (~1.86 kg CO
2e
hL
1
) and maximum in the
case of the 33-cL glass bottle cluster packs (~48.3 kg CO
2e
hL
1
),
while the latter ranged from as low as ~8.1 kg CO
2e
hL
1
in the case
of aluminum cans to as high as ~12.4 kg CO
2e
hL
1
for three 33-cL
bottle packs.
The CF values for the packages examined here were consider-
ably lower than those recently reported in the literature, probably
because of the larger production scale and shorter distribution
chain of Birra Peroni brewery. Also, the CO
2
credits from the
anaerobic digestion of wastewaters and utilization of beer co-
products as feed partly offset the higher energy-related emissions
in the product chain.
The one-factor-a-time sensitivity analysis revealed that two
promising strategies might be applied to reduce the overall GHG
emissions. Firstly, the replacement of glass bottles and steel kegs
with plastic bottles and drums; and, secondly, use of organic barley
grown locally were, in descending order, the options yielding the
greatest reduction in the carbon footprint of pale lager. Both these
strategies might be generally applied, being not specifically related
to the brewery production scale examined here.
Contrary to the quite numerous environmental assessment ex-
ercises openly available, the choice of resorting to wholly trans-
parent data allows the present CF model to be reproduced by any
researcher, this being one of the main principles of the scientific
method. Thus, despite further work is needed to collect primary
data for barley and corn agriculture, consumption of equipment
lubricant and forklift diesel oil, as well as post-consumer waste
management, this CF model relying on transparent data may serve
as an example for similar studies in other countries, as well as it
may generate other CF estimates as soon as better quality data are
available. Also the effect of the beer production scale on the carbon
footprint should be quantified, especially because of the current
market growth of the microbrewery phenomenon in Europe and
North America.
Acknowledgments
The authors thank Dr. Luigi Serino and Dr. Giovanni Battista
Morlino of Birra Peroni Srl (Rome, Italy)for their continuous support
on data gathering.
This research was supported by the Italian Ministry of Instruc-
tion, University and Research, special grant PRIN 2010e2011 eprot.
2010ST3AMX_003. It was partly presented as a poster at the 11th
International Trends in Brewing Symposium, Ghent (Belgium),
13e17 April 2014.
List of symbols
AA Tap water
AC Compressed air
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e1816
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

ALC 33-cL aluminum can
AP Well water
APR Process water
AT Articulated truck
B33 33-cL glass bottles
B66 66-cL glass bottles
BCI Primary-packaged Peroni lager
BCII Secondary-packaged Peroni lager
BCIII Tertiary-packaged Peroni lager
BP Chill haze-free Peroni lager ready-to-be packaged
BPC Byproduct credits
BRP Brewing processing
BS Beer sales
C Multipack
CA Cartons
CF Carbon footprint of a functional unit (kg CO
2e
hL
1
), as
defined by Eq. (1)
CL Cardboard multipack
CO Can open closures
CO
2
Carbon dioxide
COC Carton or tray adhesive
COE Label adhesive
COL Neck labels
CSK Plastic keg ball locker
DE Diatomaceous earth
E Body labels
EBC Bar code labels
EE Electric energy
EF
i
Emission factor for the generic i-th activity
EO Keg holographic label
EP Pallet label
F 30-L stainless steel keg
FP Pallet wrap stretch film
FV Tray wrap stretch film
GB Glass bottle
GHG Greenhouse gas
GWP Global Warming Potential
HRT Heavy rigid truck
ID Ink and diluent
i, j dummy indexes
LCA life-cycle assessment
ME related to the Messina province of Italy
NP Pallet tape
ODP Ozone depletion potential
P Wooden pallet
PCNC Polymereclay nanocomposite
PM Packaging materials
PP Packaging
PR related to the Parma province of Italy
PVPP Polyvinylpyrrolidone
Q Thermal energy
RPM Raw materials and processing aids
RT Lightemedium rigid truck
SAC Carbon dioxide waste
SB Beer waste
SB33 33-cL glass bottle waste
SB66 66-cL glass bottle waste
SC Can waste
SCA Carton waste
SCO Can open closure waste
SCOL Neck label waste
SCSK Keg ball locker waste
SE Body label waste
SEBC Bar code label waste
SEO Keg holographic label waste
SEP Pallet label waste
SF Keg waste
SFP Pallet wrap film waste
SFV Tray wrap film waste
SNP Pallet tape waste
SSK Stainless steel keg
STC Crown closure waste
STP Standard temperature (273.15 K) and pressure (1 bar)
SV Tray waste
TC Crown closure
TR Transportation
UCY Utility consumption yield
V Cardboard tray.
WD Waste disposal.
J
i
Entity of the i-th activity, expressed in kg, J or Mg km.
References
ADEME, 2007. Emission Factors Guide. Version 5.0. Emission Factors Calculation and
Bibliographical Sources Used, p. 23.
AFP, 2014. Bruges to Build Beer Pipeline to Stop Traffic Ruining Locals' Lives. The
Telegraph. Available at: http://www.telegraph.co.uk/finance/newsbysector/
retailandconsumer/11117320/Bruges-to-build-beer-pipeline-to-stop-traffic-
ruining-locals-lives.html (accessed 14.01.15.).
Ali, S., Caum, A., Hall, N., Haney, A., Lepre, J., Zimmerma, Z., 2010. Sustainability
Report: Comparison between New Belgium and Budweiser Beer. University of
Vermont. Available at: http://www.uvm.edu/~shali/Beer.pdf (accessed
14.01.15.).
Assobirra (Associazione degli Industriali della Birra e del Malto), 2013. Annual
Report 2013. Assobirra, Rome, Italy. Available at: http://www.assobirra.it/press/
wp-content/ar2013-assobirra_03-09.pdf (accessed 14.01.15.).
Becker, A., 2014. Siemens Technology Controls the Beer Pipeline in the Hacker
Festival Tent. Siemens AG, München, Germany. Available at: http://www.
siemens.com/press/pool/de/feature/2014/corporate/2014-09-oktoberfest/fact-
sheet-bierpipeline-en.pdf (accessed 14.01.15.).
Berners-Lee, M., 2010. How Bad are Bananas? the Carbon Footprint of Everything.
Profile Books Ltd, London.
BIER, June 2012. Research on the Carbon Footprint of Beer. Beverage Industry
Environmental Roundtable. Available at: http://media.wix.com/ugd/49d7a0_
70726e8dc94c456caf8a10771fc31625.pdf (accessed 14.01.15.).
Birra Peroni, 2012. 2011e2012 Sustainibility Report (Rapporto di sostenibilit
a2011-
2012). Available at: http://www.birraperoni.it/pdf/rapporto_sostenibilita.pdf
(accessed 14.01.15.).
BSI, 2008a. Publicly Available Specification (PAS 2050) for the Assessment of the Life
Cycle Greenhouse Gas Emission of Goods and Services. British Standards
Institution, London.
BSI, 2008b. Guide to PAS: How to Assess the Carbon Footprint of Goods and Ser-
vices. British Standards Institution, London.
Chicago Manufacturing Center, 2009. Product Carbon Footprint of Goose Island
Brewery 312 Beer Keg. Available at: http://www.glproject.com/filebin/pdf/312_
Keg_Footprint_Report.pdf (accessed 31.07.14.).
Cimini, A., Moresi, M., 2015. Novel cold sterilization and stabilization process
applied to a pale lager. J. Food Eng. 145, 1e9. http://dx.doi.org/10.1016/
j.jfoodeng.2014.08.00.
Climate Conservancy, 2008. The Carbon Footprint of Fat Tire
®
Amber Ale. Available
at: http://www.newbelgium.com/Files/the-carbon-footprint-of-fat-tire-amber-
ale-2008-public-dist-rfs.pdf (accessed 14.01.15.).
Coca Cola, 2010. What's the Carbon Footprint of a Coca Cola?. Available at: http://
www.coca-cola.co.uk/environment/what-s-the-carbon-footprint-of-a-coca-
cola.html (accessed 14.01.15.).
Cordella, M., Tugnoli, A., Spadoni, G., Santarelli, F., Zangrando, T., 2008. LCA of an
Italian lager beer. Int. J. Life Cycle Assess. 13 (2), 133e139.
Donoghue, C., Jackson, G., Koop, J.H., Heuven, A.J.M., 2012. The Environmental
Performance of the European Brewing Sector. Report no. 3101010DR02. The
Brewers of Europe, Brussels, B.. Available at: http://www.brewersofeurope.org/
uploads/mycms-files/documents/archives/publications/2012/envi_report_
2012_web.pdf (accessed 14.01.15.).
EPD, 2010. EPD
®
Carlsberg
®
and Tuborg
®
Beer. Reg. no S-EP-00264. Available at:
http://www.environdec.com (accessed 14.01.15.).
EPD, 2011a. EPD
®
Tuborg
®
Beer. Reg. no S-P-00311. Available at: http://www.
environdec.com (accessed 14.01.15.).
EPD, 2011b. EPD
®
BAP Bock Chiara®and BAP Bock Rossa
®
Beer. Reg. no. S-P-00314.
Available at: http://www.environdec.com (accessed 14.01.15.).
Eurammon, 2011. Ammonia ea Natural Refrigerant. Eurammon Information No. 2.
Available at: http://www.eurammon.com/sites/default/files/attachments/
eurammon02en_0_0.pdf (accessed 14.01.15.).
Eßlinger, H.M., 2009. Handbook of Brewing and Beer Processes eTechnology e
Markets, first ed. Wiley-VCH Verlag GmbH, Weinheim, Germany.
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e18 17
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063

Field to Market, 2012. Corn: Environmental Results. Available at: https://www.
fieldtomarket.org/report/national-2/FINAL_Fact_Sheet-Env_Results_Corn_
081913.pdf (accessed 14.01.15.).
Frees, N., Pedersen Weidema, B., 1998. Life Cycle Assessment of Packaging Systems
for Beer and Soft Drinks. Energy and Transport Scenarios. Environmental
project n

406. Danish Environmental Protection Agency, Miljøstyrelsen,
København, Denmark.
Goedkoop, M., Spriensma, R., 2001. The Eco-indicator 99. A Damage Oriented
Method for Life Cycle Assessment. Methodology Annex, third ed. PR
e Consul-
tants B.V., Amersfoort (NL). Available at: http://teclim.ufba.br/jsf/indicadores/
holan%20ecoindicator%2099.pdf. (accessed 14.01.15.).
Gopal, C., Rehmanji, M., 2000. PVPP ethe Route to Effective Beer Stabilization.
Brewers' Guardian. May, 1e6.
Gresser, A., 2010. Il manuale del birraio pratico. Teoria e pratica della preparazione
del malto e della fabbricazione della birra. Fachverlag Hans Carl GmbH, Nürn-
berg, Germany, pp. 76e78.
Hospido, A., Moreira, M.T., Feijoo, G., 2005. Environmental analysis of beer pro-
duction. Int. J. Agric. Resour. Gov. Ecol. 4, 152e162.
IFC, 2007. Environmental, Health, and Safety Guidelines for Breweries. Available at:
http://www.ifc.org/wps/wcm/connect/a1b1ce8048855d0e8dc4df6a6515bb18/
finalþ-þbreweries.pdf?mod¼ajperes (accessed 14.01.15.).
International Aluminium Institute, 2013. Carbon Footprint Guidance Document.
Position Paper. Available at: http://www.world-aluminium.org/media/filer_
public/2013/01/15/fl0000169.pdf (accessed 14.01.15.).
IPCC, 2007. Climate Change 2007. IPCC 4th Assessment Report. The physical science
basis. Available at: http://www.ipcc.ch/ipccreports/ar4-wg1.htm (accessed
14.01.15.).
ISO, 2006. Environmental Management - Life Cycle Assessment - Principle and
Framework. International Organisation for Standardisation (ISO), Geneve, CH.
ISPRA (Istituto Superiore per la Protezione e la Ricerca Ambientale), 2012. Ther-
moelectric Production and CO
2
Emissions. Renewable Sources and Plants Sub-
mitted to Emission Trading Scheme (Produzione termoelettrica ed emissioni di
CO
2
. Fonti rinnovabili e impianti soggetti a ETS). Rapporto n. 135/2011. Available
at: http://www.isprambiente.gov.it/it/pubblicazioni/rapporti/produzione-
termoelettrica-ed-emissioni-di-co2 (accessed 14.01.15.).
ISPRA, 2013. Italian Greehouse Gas Inventory 1990-2011: National Inventory Report
2013. ISPRA Rapporti 177, Rome, Italy. Available at: http://www.isprambiente.
gov.it/it/pubblicazioni/rapporti/italian-greenhouse-gas-inventory-1990-2011.-
national-inventory-report-2013 (accessed 14.01.15.).
Kanokkantapong, V., Kiatkittipong, W., Panyapinyopol, B., Wongsuchoto, P.,
Pavasant, P., 2009. Used lubricating oil management options based on life cycle
thinking. Resour. Conserv. Recycl. 53, 294e299.
Kløverpris, J.H., Elvig, N., Nielsen, P.H., Nielsen, A.M., Ratzel, O., Karl, A., 2009.
Comparative Life Cycle Assessment of Malt-based Beer and 100% Barley Beer.
Available at: http://www.novozymes.com/en/sustainability/Published-LCA-
studies/Documents/Comparative%20LCA%20of%20malt%20based%20and%
20barley%20based%20beer%20-%20Full%20report.pdf (accessed 14.01.15.).
Koroneos, C., Roumbas, G., Gabari, Z., Papagiannidou, E., Moussiopoulos, N., 2005. Life
cycle assessment of beer production in Greece. J. Clean. Prod. 13 (4), 433e439.
Lalonde, S., Nicholson, A., Schenck, R., 2013. Life Cycle Assessment of Beer in Sup-
port of an Environmental Product Declaration. Institute for Environmental
Research and Education (IERE). Available at: http://iere.org/wp-content/
uploads/2013/10/IERE_Beer_LCA_Final.pdf (accessed 14.01.15.).
Lan, T., 2007. Nanocomposite materials for packaging applications. In: Annual
Technical Conference SPE, Cincinnati (Ohio, USA, May, 6-10, 2007. Available at:
http://www.nanocor.com/tech_papers/Antec-Nanocor-Tie%20Lan-5-07.pdf
(accessed 14.01.15.).
Ma, B.L., Liang, B.C., Biswas, D.K., Morrison, M.J., McLaughlin, N.B., 2012. The carbon
footprint of maize production as affected by nitrogen fertilizer and maize-
legume rotations. Nutr. Cycl. Agroecosyst. 94, 15e31.
Madden, R., Darby, H., 2012. Hops Harvest Moisture Determination. University of
Vermont. Available at: http://www.uvm.edu/extension/cropsoil/wp-content/
uploads/Hop_harvest_fact_sheet.pdf (accessed 14.01.15.).
Manfredi, S., Tonini, D., Christensen, T.H., Scharff, H., 2009. Landfilling of waste:
accounting of greenhouse gases and global warming contributions. Waste
Manag. Res. 27, 825e836.
Mata, T.M., Costa, C.A.V., 2001. Life cycle assessment of different reuse percentages
for glass beer bottles. Int. J. Life Cycle Assess. 6 (5), 307e319.
Mejía, D., 2003. Maize post-harvest Operations. AGST/FAO. Available at: http://
www.fao.org/fileadmin/user_upload/inpho/docs/Post_Harvest_
Compendium_-_MAIZE.pdf (accessed 14.01.15.).
Miller, S.A.B., 2011. Water Futures: Addressing Shared Water Challenges through
Collective Action. Available at: http://www.water-futures.org/fileadmin/user_
upload/PDF/2011_Water_Futures_Report.pdf (accessed 14.01.15.).
Moresi, M., May/June 2014. Assessment of the life cycle greenhouse gas emissions
in the food industry. Agro Food Ind. Hi-Tech 25 (3), 53e62.
Moresi, M., Paone, G., 2012. Carbon footprint of Claudia natural sparkling mineral
water (Carbon Footprint dell'acqua minerale naturale effervescente naturale
“Claudia”). Ind. Bevande 41 (238), 9e23.
Mu~
noz, E., Riquelme, C., Cardenas, J.P., 2012. Carbon footprint of beer eanalysis of a
small scale processing plant in Chile. In: Proceedings of the 2
nd
LCA Conference,
November, 6-7, 2012, Lille, France.
NGA, 2013. National Greenhouse Accounts Factors - July 2013. Available at: http://
www.environment.gov.au/climate-change/greenhouse-gas-measurement/
publications/national-greenhouse-accounts-factors-july-2013 (accessed
14.01.15.).
Normand, J., Battista, K., Bobier, J., Schritt, C., Antony, S., 2012. Life Cycle Assessment
of Beer. Available at: https://prezi.com/idfmkerh9vrt/life-cycle-assessment-of-
beer/ (accessed 14.01.2015).
Olajire, A.A., 2012. The brewing industry and environmental challenges. J. Clean.
Prod. xxx, 1e21 (in press).
Pasqualino, J., Meneses, M., Castells, F., 2011. The carbon footprint and energy
consumption of beverage packaging selection and disposal. J. Food Eng. 103,
357e365.
PCR, 2013. UN CPC 2431 Beer Made from Malt. Version 1.01, 2013-07-16. The In-
ternational EPD®System.
Platto Feeding Company, 2014. Beer Spent Grains (Trebbie di birra). Available at:
http://www.platto.it/?p¼116 (accessed 14.01.15.).
Saltelli, A., Ratto, M., Tarantola, S., Campolongo, F., 2006. Sensitivity analysis prac-
tices: strategies for model-based inference. Reliab. Eng. Syst. Saf. 91,1109e1125 .
Saxe, H., 2010. LCA-based Comparison of the Climate Footprint of Beer vs. Wine &
Spirits. Institute of Food and Resource Economics, Report no. 207, Copenhagen.
Available at: http://curis.ku.dk/ws/files/44663043/FOI_report_207.pdf
(accessed 14.01.15.).
Senturk, A., Yalcın, B., Otles, S., 2013. Nanotechnology as a food perspective.
J. Nanomater. Mol. Nanotechnol. 2, 6e8. http://dx.doi.org/10.4172/2324-8777.
100 0125 .
Smith, A., 2008. One Small Step. Available at: http://www.packaging-gateway.com/
features/feature43794/ (accessed 14.01.15.).
Sturm, B., Hugenschmidt, S., Joyce, S., Hofacker, W., Roskilly, A.P., 2013. Opportu-
nities and barriers for efficient energy use in a medium-sized brewery. Appl.
Therm. Eng. 53, 397e404.
Takamoto, Y., Mitani, Y., Takashio, M., Itoi, K., Muroyama, K., 2004. Life cycle in-
ventory analysis of a beer production process. Master Brew. Assoc. Am. 41 (4),
363e365.
Talve, S., 2001. Life cycle assessment of a basic lager beer. Int. J. Life Cycle Assess. 6
(5), 293e298.
Tukker, A., Huppes, G., Guin
ee, J., Heijungs, R., de Koning, A., van Oers, L., Suh, S.,
Geerken, T., Van Holderbeke, M., Jansen, B., Nielsen, P., 2006. Environmental
Impact of Products. (EIPRO). Analysis of the Life Cycle Environmental Impacts
Related to the Final Consumption of the EU-25. Technical Report EUR 22284 EN.
European Communities. Available at: http://ec.europa.eu/environment/ipp/pdf/
eipro_report.pdf (accessed 14.01.15.).
UNEP,1996. Environmental Management in the Brewing Industry. Technical Report
Series No. 33. United Nations Environment Programme eIndustry and Envi-
ronment, Paris.
Watson, J., 2008. Draught Beer Beats Bottled in Life Cycle Assessment. Treehugger.
Available at: http://www.treehugger.com/clean-technology/draught-beer-
beats-bottled-in-life-cycle-assessment.html (accessed 14.01.15.).
A. Cimini, M. Moresi / Journal of Cleaner Production xxx (2015) 1e1818
Please cite this article in press as: Cimini, A., Moresi, M., Carbon footprint of a pale lager packed in different formats: assessment and sensitivity
analysis based on transparent data, Journal of Cleaner Production (2015), http://dx.doi.org/10.1016/j.jclepro.2015.06.063