An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems

Applied Energy xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems R. Santagata a, M. Ripa a,⇑, S. Ulgiati a,b a b

Parthenope University of Naples, Italy School of Environment, Beijing Normal University, China

h i g h l i g h t s  Animal by-products use for electricity generation is investigated as a case-study.  Different methodological approaches to deal with by-products are explored in LCA.  Adopting a holistic perspective is crucial to achieve a circular economy framework.

a r t i c l e

i n f o

Article history: Received 20 December 2015 Received in revised form 20 June 2016 Accepted 16 July 2016 Available online xxxx Keywords: Animal by-products Urban waste management LCA Electricity production Circular economy

a b s t r a c t The food processing industry continues to grow, generating large amount of organically rich waste flows per year: these processors face significant economic and environmental pressures for appropriate conversion and disposal of these waste flows. Solid waste disposal problems, mostly in highly urbanized environments, energy shortages (primarily oil) and/or high petroleum prices, as well as environmental issues such as the shrinking landfill capacity, can all be addressed by converting waste material into useful and saleable products. This paper brings to the attention a possible strategy in order to meet the general EU directives concerning the residues utilization and percentage contribution for the total energy consumption by 2020, by evaluating the use of animal by-products (category 3, as defined in the directive 2002/1774/EC) for energy purposes. Slaughterhouse waste represents an important potential source of renewable energy: on average, 40–50% of a live animal is waste, with a potential energy content close to diesel fuel. Treatment of animal waste from slaughterhouse and the subsequent conversion to electricity is investigated as a case study in the Campania Region (Italy): the animal waste undergoes a rendering process, to separate a protein-rich fraction useful for animal meal production and a fat-rich fraction, to be combusted in a diesel engine for power and heat generation (CHP). An environmental assessment of the entire process is performed by means of LCA, providing a quantitative understanding of the plant processing. The study aims to understand to what extent electricity production from animal fat is environmentally sound and if there are steps and/or components that require further attention. The environmental impacts of the electricity production from animal waste are investigated adopting different points of view and they are also compared to the impacts of Italian electricity production (mix of fossil fuels and renewables). The study confirms that waste recovery represents a triple win solution, dealing simultaneously with human security, pollution, and, last but not least, energy recovery. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Urbanization and human population have considerably grown, in the last decades, along with industrialization development. This caused an increase of environmental pollution, as a side effect [1]. ⇑ Corresponding author. E-mail address: [email protected] (M. Ripa).

The relationship between economic growth and environmental pollution has become the subject of passionate research over the latest decade. Since there is a large and growing world-wide consumption of fossil fuels, the amount of CO2 released to the atmosphere also increased [2]. In modern societies, the environmental pollution mainly relies on two key issues: (1) the depletion of fossil fuels and limited availability of other non-renewable resources; and (2) waste generation that is pushing to the limits the

http://dx.doi.org/10.1016/j.apenergy.2016.07.073 0306-2619/Ó 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

2

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

biosphere’s ability to dilute waste. Both could be considered as byproducts of the technological development of human society [3]. In natural systems, all material flows are circular and the very concept of waste does not apply: ‘waste’ products and flows from a process always become inputs to other processes. Instead, human dominated systems are typically unable to continuously re-use all waste flows, which puts increased pressure on the environment in terms of pollution as well as ever-increasing depletion of natural resources [4]. Particularly, the limited capacity of conventional oil production to meet growing demand and the limited amount of crude oil reserves, may have large impact on the evolution of human societies in the next years [5]. The increasing demand for fossil fuel gives rise to environmental concerns such as larger CO2 and greenhouse gas (GHG) emissions and global warming. The world energy consumption doubled between 1971 and 2001 and the world energy demand is expected to increase by 53% within the year 2030 [6]. This context invigorated the necessity to look for environmental friendly renewable energy sources as well as to increase the overall resource use efficiency, within a circular economy framework. In many countries energy policies have been implemented to decrease consumption by means of increased energy efficiency as well as to increase the share of renewable energy in the energy country mix, not only to respond to the international pressure for low carbon energy transition but also to intensify their energy security and economic affordability for industries and end users as critical issues to deal with [7]. Furthermore, waste generation applies pressure on both the environment and the human health, thus calling for improved waste management strategies to replace the present polluting methods. For instance, landfilling is one of the most commonly used waste disposal method, and accounts for approximately 67% of the total collected MSW worldwide [8] (31% in the only European Community [9]) with heavy environmental consequences due to leachate contamination of underground water as well as methane release to the atmosphere; incineration, most often considered as another mainstream technology, has faced a rapid development in recent years, in spite of the fact that toxic substances such as heavy metals and dioxin released during combustion may cause negative effects to the environment and human health [10,11], entailing heavy costs for management [12] and being a cause of degradation for the standard of living of populations in urbanized environments. As a result, scientists and concerned managers avidly seek alternatives to fossil fuels: focus on recovery of materials, resources and energy from waste has increased significantly in the endeavor of reducing fossil fuel consumptions and resources depletion [13]. In addition, in those regions where landfilling (instead of, for example, incineration) is the most common disposal method, the recovery of the organics (e.g. kitchen waste, tissues, etc.) becomes a necessary priority in order to minimize landfilling volume and comply with legislative targets [14]. The recycling of materials, and thus the minimization of waste to be disposed of, is a basic concept which must be implemented in order to meet the sustainable development goals in both industrialized and developing countries. It has been claimed that the carrying capacity of the planet has already been exceeded in several areas, including climate forcing [15]. Energy efficiency and clean energy have been recognized as key factors to minimize the cost and negative effect of climate change on the environment and society [16]. The energy sector is the largest contributor to GHG emissions [17], and for this reason, strategies to reduce the emissions from this sector are a key point of climate change mitigation strategies [18]. Waste contains fossil derived materials such as plastics. Moreover, it also contains biogenic materials such as paper, card and food waste. All of these fractions can be potentially converted into energy. The implementation of waste-to-energy (WTE) supply chains was suggested as a suitable method for energy production from waste, in order to

address two of the main waste management environmental issues (limited landfilling sites and leachate). The WTE supply chain, in its CHP (Combined Heat & Power) version, if properly managed provides a method for simultaneously addressing energy demand, waste management and GHG emissions within a circular economy perspective (CES) [19]. Although not exempt from criticisms when applied to the very complex composition of MSW (municipal solid waste), WTE-CHP is put forward as an alternative to waste landfilling or incineration without energy recovery [20]. The EU Directives on waste management prescribe prevention, reuse and recycling as the very first alternatives, indicating the energy recovery option only for smaller amounts for which the previous alternatives are not easily feasible or fail. This seems to be the case of fat fractions of slaughterhouse residues dealt with in this paper, after other uses have been explored. 1.1. Urban waste and its energy potential Most waste generation (in particular food waste) occurs in cities, where more than 50% of world population lives. Cities are increasingly concerned of both energy consumption and waste generation and their possibility to close the loop requires accurate energy/urban planning. Urban/local energy planning was developed later than urban planning, in order to suggest conceptual strategies for economic, environmental and energy-related regulations and policy implementation. Therefore, environmentally friendly city planning needs to facilitate the integration of ‘urban planning’ and ‘energy planning’ into a unified ‘urban/local energy planning’ system and to support adequate technology for each process [21]. Innovative waste refining processes provide potential solutions for energy and materials recovery. The generation of agro-industrial waste has been rising to such alarming levels that the public has become aware of the problems caused by inaccurate management. Nowadays the generation of waste biomass is so abundant and so centralized that there is insufficient capacity for its natural degradation, and various treatment techniques have to be applied. Animal slaughterhouse waste is also city related, in that the demand for meat-based diet is growing in cities and is not expected to decrease in the short run. Slaughterhouses represent one of the most important sectors of the meat industry [22]. Non-edible feedstocks, such as animal fat waste (AFW), have recently increased in popularity as alternatives to vegetable oils in the production of biodiesel [23–26]. Animal by-products are defined by European Directive 2002/1774/EC as entire bodies, or parts of animals or products of animals, not intended for human consumption. The Directive lists three different categories of animal by-products: Category 1 - all by-products likely to be the vehicles of infectious diseases-, Category 2 - all materials coming from manure, digestive tract contents and material collected when treating waste water-, Category 3 - by-products fit for human consumption, but not intended for it for commercial reasons [27]. Animal fats are primarily derived as by-products of meat animal processing facilities and of the rendering process. A large percentage of livestock live weight (an amount of about 48% by mass) consists of byproducts (i.e. fat and meal) [28] which show an energy content not far from diesel fuel (animal fat: 3.98E+04 J/g average, animal meal: 1.85E+04 J/g average) [29,30]. Ariyaratne et al. [30] also show how meat and bone meal can be used as renewable energy source in substitution of pulverized coal in rotary kiln burners (up to 40%, reducing annual CO2 emissions of these burners by 10%). Studies about AFW address the need to manage these residues in a way that is capable to simultaneously dispose of the waste material and also obtain benefits from co-products, both in terms of energy (biodiesel and bio-methane [6,20,22–26,29–40]) and added value industrial products [41–44]. However, the existing literature about AFW conversion lacks of three main

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

characteristics: (a) assessment of environmental costs and benefits, to complement the energy evaluation; (b) reliance on real case studies instead of feasibility scenarios, and (c) exploration of energy uses other than biofuels (e.g. electricity). Moreover, recovery patterns and technologies are able to manage the large amounts of organic slaughterhouse residues that are daily available and cannot be stored for a too long time to prevent degradation. This means that a solution must at the same time address the need for a multiplicity of products and the fact that these products cannot only be small amounts for niche users (such as cosmetics or pharmaceuticals). A slaughterhouse bio-refinery is still to be developed starting from real case processes. As a first step towards the implementation of an overall circular framework for food residues, we address in this study the three points above, by performing a life cycle assessment of a real case electricity production process, in order to start testing and monitoring benefits and costs as well as to tailor the assessment method to the characteristics and specificity of the slaughterhouse sector and its potential development into a bio-refinery pattern. Calculations include, within the adopted allocation procedure, also the correct assignment of inputs and burdens to potential co-products such as animal feed and other chemicals. The investigated electricity production is characterized by the needed flexibility and ease of transportation and use, that make it a basic product and the first step for the planned biorefinery and circular feedback to the urban system. In a previous study concerning the same industrial company, we have already investigated the environmental impacts and benefits of waste cooking oil recovery to biodiesel, within such circular perspective [45]. The present paper explores, by means of the LCA method, the environmental feasibility of processing slaughterhouse animal waste electricity as a case study in Campania Region (Italy), addressing simultaneously the European energy (Directive 2009/28/EC) and waste directives (Directive 2008/98/EC). The full rendering process and electricity generation process will be assessed under different perspectives, in order to understand how they affect the final results: animal by-products stream will be accounted alternatively as a waste stream and as a product stream, also comparing the electric energy obtained to the one coming from the Italian grid. Different approaches will be considered also to allocate burdens at the different products generated by the rendering process. Besides investigating on the feasibility and impacts of the AFW to electricity process, the aim of the present study is to point out to what extent perspectives and assumptions affect the final results.

2. Materials and methods The methodological framework used in this paper is the LCA as defined by ISO standards and ILCD Handbook guidelines [46–48]. Life Cycle Assessment is a methodological framework to assess the potential environmental impacts and resources used throughout a product’s life cycle, from raw material acquisition, via production and use phases, to waste management. All activities and processes result in environmental impacts due to consumption of resources, emissions of substances into the natural environment, and other environmental exchanges. In other words, LCA looks at the process relation with the environment as a source and as a sink, and provides indicators related to many different environmental impact categories, such as climate change, stratospheric ozone depletion, depletion of resources, toxicological effects, among others [49]. LCA is a relatively recent method that has rapidly grown to become a standard tool to investigate the environmental performance of a wide range of human-dominated processes.

3

2.1. Case study description The process investigated as case study in this paper is operated by Proteg S.P.A., a company located in the industrial area of the municipality of Caivano, in Campania region, Italy. Proteg also operates in the field of the waste vegetable oil refining for biodiesel production. The industrial plant is capable of producing average 5.1 MW of electric energy using animal by-products properly processed. The system under examination consists of three stages: (1) collection of AFWs, (2) rendering process in order to obtain animal meal and animal fat, and (3) AFW conversion into electricity through refining and combustion of animal fat in a marine-derived engine. In the first phase, raw materials - i.e. all the unusable parts of a carcass, including bones, internal organs and trimmings - are collected for processing from abattoirs and from butchers and food processing sites, using trucks belonging to the fleet of the investigated company. After delivering at the processing plant, these materials are sent to the rendering process: they are placed in storage tanks and then are transferred to a crusher where their size is reduced to fragments measuring up to 30 mm. The organic material is moved throughout the plant by screw conveyors, bucket elevators and pumps powered by electricity. The chopped material is cooked in a continuous steam cooker for a period of no less than 1 h, at temperatures over 150 °C. This process dries up the water and sterilizes the organic material. After this phase, two different fractions are obtained: a solid protein fraction (i.e. bone and meat) and a fluid fraction (i.e. un-purified animal fats). The protein fraction undergoes various processes of pressing, crushing, and then grinding by means of hammer-mills. The obtained meal is stored in special silos. The grinding plant includes also two bag-house filters, to facilitate the removal from the mill and the cooling of the milled product. The removed air is sent to a scrubber. The fluid fraction is pumped to a series of vertical and horizontal centrifuges, to reduce the percentage of impurities. Alternatively, fats can be sent to a horizontal centrifuge and to a desiccator, to remove water at temperatures around 70 °C, and then to a filter-press, to remove impurities. After purification, the fat is stored and warmed in special tanks using steam, if necessary. The animal meal and part of the fat (around 50% by mass) are sold to the market. The meal is used mainly for pet-food production and as organic fertilizer, while animal fat is used mainly for bio-diesel and cosmetics production. The remaining 50% of fat is used in the second phase of the process, as fuel for a naval-derived engine within a co-generation plant. The cogeneration plant is essentially formed by a MAN 12v 32/40 engine, connected to a cast iron flywheel and an alternator, all held together by a steel basement. The engine is fuelled with animal fat, lubricating oil and urea. Urea is used for the uptake of NOx emissions, through a chemical reaction known as ‘selective reduction’. A small amount of fossil diesel fuel is also used, but only during the off state, to clean the inside of the engine by fat residues, which may solidify and cause damage. The plant is operative 24 h/day, all days of the week. A small amount of the electric energy produced (between 10% and 20%) is used for the self-consumption of the plant, while the rest is delivered to the grid. The plant also includes a thermal oxidizer for odorous emissions abatement, which allows oxidizing ammonia, sulphur and volatile organic compounds (VOC), and a wastewater treatment plant, which processes an average of 3 m3/h of generated wastewater. A schematic diagram of the entire process is shown in Fig. 1. 2.2. Goal and scope definition LCA methodology, as a four stages tool (definition of goal and scope, inventory analysis, impact assessment and interpretation) for environmental management at global level, is standardized in

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

4

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

Fig. 1. Rendering process and electricity generation schematic diagram.

ISO documents 14040/2006 and 14044/2006, as well as in the ILCD Handbook [46–48]. The ILCD Handbook, a detailed technical guide to LCA redacted by European Commission on the base of ISO standards, confirms the importance and the role of LCA as a decisionsupporting tool in contexts ranging from product development to policy making. The ISO standards describe the goal and scope step as the phase in which the intended uses and users of the LCA results are identified and the overall context is framed (functional unit, quality of data, regional boundary, etc.). The main purpose of this paper is to analyze the generation of electric energy from AFW, coming from the rendering process of animal by-products, under an environmental perspective, comparing it to the electricity production by conventional routes. The most significant and sensitive steps in the system under study are identified in order to evaluate the most crucial ones, in terms of environmental impacts and energy requirements, and suggest potential improvements. Waste management systems are often seen as primarily targeted at getting rid of its inputs (the waste flows), and would therefore be categorized as multi-inputs systems, with the consequence that all its impacts should be allocated to these foreground inputs, regardless of whether any of them are converted into useful outputs (e.g. recycled materials or recovered energy) or not. While this fundamental categorization may be considered to still hold true, it should however be noted that modern

waste management systems are more and more being likened to production systems too, where equal importance is put on their role as active contributors to (secondary) material and energy production. According to this last consideration, the Functional Unit (FU) referred to in this study is 1 MW h of electric energy produced. The entire process can be divided into two sub-processes: (1) Rendering process of the organic material, yielding meal and fat fractions; (2) Generation of electric energy from combustion of animal fat (Fig. 2). All input flows and environmental burdens are allocated to the meal and fat exiting the first phase as coproducts, according to their mass (CASE 1); then, the fraction of inflows and outflows allocated to fat is assigned to the electricity generated in the second phase. A ‘zero-burden waste’ approach is assumed in this study, by not including the burdens associated to the upstream generation of treated waste. A ‘gate to gate’ approach is used, since the system boundary is considered coincident with the physical boundaries of the plant. In agreement with the ILCD Handbook, this study is centered in the proper accounting of different environmental impacts of the process without however accounting for large-scale consequences on the background system. The analyzed context can thus be identified as a micro-level decision support (so called situation A in ILCD) and an attributional LCI modeling framework is therefore applied. Since during the

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

5

Fig. 2. System boundaries including: transport, pre-treatment/cooking (included in the rendering process), purification and electricity production (cogeneration) of slaughterhouse residues (CASE 1 & CASE 3). The grey line refers to the feedback of electricity produced in the investigated case. Livestock and slaughtering steps in the dotted box represent upstream processes taken into account in the sensitivity analysis (CASE 2).

waste treatment processes more than a single valuable product is produced (e.g. fat and meal at the end of the rendering process), an allocation procedure, based on physical criterion, was applied [48]. Therefore, during the inventory analysis, a mass allocation was pursued to partition the input and output burdens between the different co-products. As it can be observed in Table 1, about 47% of the total mass of the co-products in output is animal fat, while 53% is meal from the rendering process. Finally, sensitivity and uncertainty assessments were performed to test the robustness of results (see Sensitivity and Uncertainty analyses). For sensitivity purpose, not only different allocation procedures were adopted (CASE 3, allocation according to the energy content of fat and meal products), but also different

Table 1 Mass and energy allocation between animal fat and animal meal produced by the rendering process. Input

Quantity

Unit

Slaughterhouse residues

1.80E +03

kg/ MW h

4.33E +02 4.97E +02 8.70E +02

kg/ MW h kg/ MW h kg/ MW h

Mass allocation (%)

Energy allocation (%)

47

65

53

35

Output Fat Meal Wastewater and other residues

methodological assumptions were assumed (CASE 2, no zeroburden approach as with CASE 1), in order to check how these changes affect the final results. Furthermore a Monte Carlo analysis was carried out to address the uncertainty related to data collection and processing.

2.3. Life cycle inventory An accurate inventory of all the input and output flows to and from the analyzed system, including those inputs that are traditionally only accounted for in terms of energy, is critical to the quality of the assessment [50]. During the inventory phase, local data were collected for each of the above-mentioned phases: all different materials (e.g. concrete, steel, glass), machinery, as well as the energy consumption for buildings construction, and plant operation. The construction and delivery of the major components of the power plant were also included. Table 2 presents a simplified inventory (LCI), organized according the two steps: (a) inventory of the rendering phase (Table 2a) and (b) inventory of electricity production (Table 2b). All the flows in Table 2 are referred to 1 MW h of electric energy produced (functional unit). Primary data, e.g. specific information about input flows to the process, recovered materials and emissions were made available by Proteg S.P.A. and are referred to a three-month period (September – November 2014). When direct measurements were not available, estimations were made by experts in both Company and research team, and their consistency was verified in literature.

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

6

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

2.4. Life cycle impact assessment

Table 2a Rendering process inventory.a Input

Unit

Amount/MW h

Animal by-products Electricity from animal fat feedback Underground water Methane Transportation

kg kW h kg m3 t ⁄ km

1.80E+03 1.41E+02 1.16E+03 1.17E+02 2.07E+02

Output

Unit

Amount/MW h

Animal fat Animal meal Particulate NOx SOx CO TOC VOC NH3

kg kg kg kg kg kg kg kg kg

4.33E+02 4.97E+02 7.16E 02 6.83E 01 4.68E 01 2.68E 01 1.16E 01 4.07E 01 9.72E 02

a The inventory only includes the main flows. Capital goods and machinery are not included for lack of space, but where accounted for in the results.

Table 2b Electricity generation process inventory.b Input

Unit

Amount/MW h

Animal fat Diesel fuel Lubricating oil Urea

kg kg kg kg

2.32E+02 2.91E 01 2.14E 01 1.33E+01

Output

Unit

Amount/MW h

Electricity from animal fat Hot water CO O2 NO2 Particulate

MW h m3 kg kg kg kg

1 7.72E+00 3.98E 01 5.17E+02 9.33E 01 2.39E 01

b The inventory only includes the main flows. Capital goods and machinery are not included for lack of space, but where accounted for in the results.

Background data over the supply chain of energy and materials were derived from the Ecoinvent v3.0 database. In particular, a comparative LCIA between electricity production from AFW and Italian electricity mix was carried out: key data for the quantification of inputs and outputs of the Italian production mix of medium voltage electricity were derived from Ecoinvent database. The percentage composition of the Italian electric mix is shown in Table 3 [51].

Table 3 Composition of the Italian electricity mix (including electricity imports) for the year 2008 [51]. Energy source

Supply mix (%)

Fossil Fuels

67.93

Hydro Nuclear Renewables other than hydro

13.84 0 3.88

Waste Other Imports

1.24 0.25 12.86

Total

100.00

Hard coal – 11.58% Industrial gases – 1.48% Petroleum products – 8.46% Natural Gas – 46.42%

Geothermal – 1.57% Solar – 0.05% Wind – 1.44% Wood – 0.74% Biogas – 0.08%

LCIA was performed by means of the LCA software SimaPro 8.0.5.13. The impact assessment was performed by means of one of the most recent and up-to-date LCA methods, the ReCiPe method [52]. In particular, ReCiPe Midpoint (H) v.1.12 (http:// www.lcia-recipe.net/) was chosen, considering that it includes both upstream categories (i.e. referred to depletion of natural resources, such as fossil, metal and water depletion categories) and downstream categories (i.e., referred to impacts generated on natural matrices, such as terrestrial, marine or freshwater acidification) [53]. Moreover, the ReCiPe Midpoint (H) method allowed to assess the environmental impacts in a large number of impact categories of interest in waste management (e.g. global warming, abiotic depletion, acidification, eutrophication). The ReCiPe method provides characterization factors to quantify the contribution of processes to each impact category and normalization factors to allow a comparison across categories (Europe ReCiPe Midpoint (H), 2000, revised 2010). Normalization is a life cycle impact assessment tool used to express characterized impact indicator data in a way that they can be compared among impact categories. This procedure normalizes the indicator results by dividing characterized values by a selected reference value for each category. There are numerous methods for selecting a reference value, including, for example, the total emissions or resource use in a given area, global, regional or local [54]. In this study, the following categories are explored: Global Warming Potential (GWP, in kg CO2 eq), Human Toxicity Potential (HTP, in kg 1,4-DB eq), Fossil Depletion Potential (FDP, in kg oil eq), Metal Depletion Potential (MDP, in kg Fe eq), Water Depletion Potential (WDP, in m3), Freshwater Eutrophication Potential (FEP, in kg P eq), Terrestrial Acidification Potential (TAP, in kg SO2 eq), Terrestrial Ecotoxicity Potential (TEP, kg 1,4-DB eq), Photochemical Oxidant Formation Potential (POFP, in kg NMVOC). 2.5. Sensitivity analysis A sensitivity analysis was also performed, in order to test the robustness of results when changing perspective and allocation procedure. Sensitivity analysis is a crucial step in the model design and result communication process. Through sensitivity analysis we gain essential insights on model behavior, on its structure and on its response to changes in the model inputs. In the presence of two or more intermediate or final products, several allocation criteria are allowed in LCA, according to the share of physical (mass, energy, exergy) flows associated to the products under examination, or according to the share of total economic value of the output flows, depending on the final goal of the assessment study. The importance of a correct allocation of input flows and environmental burdens is highlighted in the main LCA literature [48] as well as in the ISO standards. A good sensitivity analysis may improve the credibility of LCA, which is generally affected by different levels of uncertainty, i.e. the uncertainty associated with discrete changes, and for changes in system borders or allocation rules, evaluating if the contribution of the uncertain data to the outcome is small or negligible, and expressing the result as a probability quantifying the degree of uncertainty [55]. In order to quantify how results may change by adopting different perspectives, the sensitivity analysis is conducted shifting from a zero-burden, waste-oriented management to an output-oriented perspective. This shift is operated at the beginning of the investigated process, particularly at the input material level. According to this perspective, the slaughterhouse residues were considered as a co-product of the meat production process. The methodological distinction to which we refer to here in the case of waste management is rooted in whether a process is primarily seen as

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

7

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

targeted at respectively treating all its inputs, or instead at generating its intended outputs. Waste management systems invariably have more than one input and output. So, clearly, in a strict sense they are simultaneously multi-input and multi-output flows. However, the two perspectives are investigated in this study as separate options to provide a broader picture of the ‘services’ that can be achieved, and the dependence of calculated impacts upon the intended function of the process. Two different approaches are considered in this study (see Fig. 2). In the first assumption (CASE 1), the slaughterhouse material is considered as a ‘burden free’ waste stream, according to standard procedure for the life cycle of waste [56]. Afterwards, a sensitivity analysis is performed considering a second (CASE 2) and a third assumption (CASE 3). In CASE 2 the slaughterhouse material is accounted for as a coproduct of the meat production process, so the burdens related to the livestock farming phase are taken into account and allocated by mass to the meat and by-products considering a production of 52% of meat and 48% of by-products per kilogram of livestock [28]; in CASE 3 a ‘zero burden’ approach, like CASE 1, is used, but instead of the mass allocation, an energy allocation based on the energy content of animal fat and animal meal was performed to account the burdens related to the rendering process. The ILCD handbook recommends physical allocation be applied in LCA, but does not suggest any preference among the different physical allocation options. Since the slaughterhouse material in this paper is being considered as an energy substrate, an allocation according to energy content of co-products was also performed.

2.6. Uncertainty analysis Quantifying uncertainties is an important step in a process accounting, due to their potential effects on decision making. Indeed, when comparing design alternatives, apparent impact differences may be misleading if the uncertainty is large enough to overwhelm any relative differences between alternatives. Quantification of these uncertainties will support informed decision making. Data uncertainty is closely related to lack of knowledge about the ‘true value’ [57] which cannot be properly represented by a single static number. Parameter uncertainty can be represented by a probability distribution or as a range. In this study, a log-normal distribution is chosen to be a reasonable fit [58]: first, because it is asymmetric and always positive, whereas the normal distribution is symmetric and crosses the zero line; second, there is some evidence that the lognormal distribution occurs frequently in natural phenomena [59]; third, the simple mathematical properties of the log-normal distribution make it easier to work with it (e.g., multiply two independent distributions) and allow the use of the pedigree matrix [60]. The latter is a way to include qualitative uncertainty to the existing uncertainty distributions (Ecoinvent database includes only quantitative uncertainty values for parameters in many of its processes) by broadening them without changing their median values. In the pedigree matrix qualitative criteria are used to modify the basic uncertainty. The procedure is based on a matrix composed of five data quality indicators – reliability, completeness, temporal correlation, geographical correlation, and further technological correlation – each with a possible 1–5 score. A score of 1 means that the data is of high quality with regard to that particular indicator (e.g. ‘data from area under study’ for the indicator geographical correlation); a score of 5 means the data quality for that indicator is low (e.g. ‘non-qualified estimate’ for the indicator reliability). Once the single parameter uncertainties have been determined using this approach, they can be propagated using techniques such as Monte Carlo simulation, within the SimaPro 8.0.5.13 software. This propagation consists

of systematically varying input parameters to determine how sensitive the outputs are to each input. The Monte Carlo analysis was carried out with a confidence interval of 95% and a sufficiently large number of runs (1000). 3. Results LCA characterized impacts of CASE 1 electricity production from AFW are shown in Table 4. In order to gain an understanding of the suitability of such electricity generation process, a comparison with the impacts of the Italian electric mix was also accomplished, with reference to the production of 1 MW h of electricity. In each impact category, the total impact associated with electricity production from AFW is much lower than those associated with the Italian mix (MIX ITA), being the latter mainly derived from fossil fuels (with a large fraction of natural gas). In some impact categories – i.e. global warming, metal and water depletion - the impacts generated in CASE 1 process are around one order of magnitude smaller than MIX ITA. Fig. 3 shows the normalized impacts of the same processes. Normalization is an optional step in LCIA that is used to better understand the relative importance and magnitude of the impact category indicator results (without dimensional unit). The overall advantage of normalization is the possibility of making comparison across categories. According to the normalized impacts, the most affected categories are human toxicity (4.70E 2) and freshwater eutrophication (7.24E 2). Water depletion category is not detectable at all, due to the normalization factor equal to zero, and it is not shown in the Fig. 3. In order to highlight the contribution of each phase of the process, the environmental impacts of electricity production from AFW on ReCiPe Midpoint H categories were also explored step by step. Fig. 4 displays the percentage contribution of each process step to selected impact categories. Such breakdown of impacts confirms that the highest impacts in all categories come from the operation step, while construction (e.g. machinery and capital goods) plays a minor role, except for metal depletion (16% of total impact). The use of urea itself (for the control of NOx emissions in the co-generation plant) results to be the largest share of the global warming potential category (54%) and of the fossil depletion (70%), whereas only minor contributions come from the other input flows. Likewise, the contribution of urea is also high in the terrestrial acidification, freshwater eutrophication, human toxicity, metal and water depletion categories. Beyond urea, the second main contribution to environmental burdens comes from the use of methane (for the generation of steam) in global warming, human toxicity, freshwater eutrophication and water depletion, ranging from 28% to 50%. Local emissions provide a major contribution to terrestrial acidification and photochemical oxidant formation, with values of 62% and 84%, respectively.

Table 4 ReCiPe Midpoint (H) characterized impacts calculated for the generation of 1 MW h of electric energy for CASE 1 and MIX ITA. Impact category

Unit

CASE 1

MIX ITA

GWP HTP FEP TAP TEP FDP MDP POFP WDP

kg CO2 eq kg 1,4-DB eq kg P eq kg SO2 eq kg 1,4-DB eq kg oil eq kg Fe eq kg NMVOC m3

8.47E+01 2.96E+01 1.73E 02 1.27E+00 1.26E 02 2.53E+01 6.47E+00 1.35E+00 1.15E+02

6.08E+02 9.13E+01 1.02E 01 2.24E+00 8.44E 02 1.88E+02 1.59E+01 1.42E+00 3.12E+03

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

8

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

3.00E-01

Normalized Impacts

2.50E-01 2.00E-01 1.50E-01 1.00E-01 5.00E-02 0.00E+00

GWP

HTP

FEP

TAP

TEP

FDP

MDP

POFP

CASE 1

7.56E-03

4.70E-02

7.24E-02

3.70E-02

1.52E-03

1.63E-02

9.05E-03

2.37E-02

MIX ITA

5.43E-02

1.45E-01

2.47E-01

6.53E-02

1.02E-02

1.21E-01

2.22E-02

2.50E-02

Fig. 3. ReCiPe Midpoint H normalized impacts for CASE 1 and MIX ITA. Units on vertical axis are not shown, since values are unit-less ratios of actual burdens to reference burdens for standardization.

Characterized Impacts

100% 80% 60% 40% 20% 0%

GWP

HTP

FEP

TAP

TEP

FDP

MDP

POFP

Urea

54%

56%

43%

26%

63%

70%

48%

8%

WDP 42%

Methane

33%

34%

45%

8%

28%

13%

28%

4%

50%

Transportaon

11%

4%

4%

2%

7%

12%

7%

3%

4%

Machinery

2%

7%

7%

1%

2%

2%

16%

1%

4%

Diesel+Oil

0%

0%

0%

0%

0%

3%

0%

0%

1%

Local Emissions

0%

0%

0%

62%

0%

0%

0%

84%

0%

Fig. 4. Percent contributions to characterized impacts from different phases of CASE 1.

3.1. Sensitivity analysis After accounting for the impacts related to the ‘zero burden’ approach in CASE 1, two additional assumptions have been considered. CASE 2 and CASE 3 assumptions show different allocation approaches in different key-points of the investigated process. While in CASE 1 the slaughterhouse material at the beginning of the rendering process is treated like a ‘zero-burden’ waste stream, in that it enters the process without any associated impact, in CASE 2 the slaughterhouse material is considered like a material stream, being associated to the production burden emerging from the livestock phase. The burden per kilogram of livestock has been allocated according to the mass of co-products, namely 52% to the resulting meat, and 48% to the residues delivered to the electricity production process. In CASE 3, the burdens related to the rendering sub-process are allocated between animal fat and animal meal according to the fraction of energy contained in these two streams, according to Table 1. The accounted percentages of energy associated to fat and meal are respectively 65% and 35% of the total energy content of these co-products. Table 5 compares the

Table 5 ReCiPe Midpoint (H) characterized impacts calculated for the generation of 1 MW h of electric energy for CASE 1, CASE 2 and CASE 3. Impact category

Unit

CASE 1

CASE 2

CASE 3

GWP HTP FEP TAP TEP FDP MDP POFP WDP

kg CO2 eq kg 1,4-DB eq kg P eq kg SO2 eq kg 1,4-DB eq kg oil eq kg Fe eq kg NMVOC m3

8.47E+01 2.96E+01 1.73E 02 1.27E+00 1.26E 02 2.53E+01 6.47E+00 1.35E+00 1.15E+02

6.15E+03 4.42E+02 1.96E+00 1.49E+02 4.90E+00 2.88E+02 1.10E+02 1.01E+01 2.15E+03

9.98E+01 3.42E+01 2.09E 02 1.43E+00 1.44E 02 2.80E+01 7.60E+00 1.46E+00 1.44E+02

characterized results associated with the 3 cases considered. While CASE 3 shows only slightly larger impacts than CASE 1, CASE 2 shows much greater impacts in every category. For the categories HTP, FDP, POFP and WDP, the impact generated by CASE 2 are one order of magnitude greater than CASE 1, while for GWP, FEP, TAP, TEP and MDP the impacts generated are about two orders of magnitude greater.

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

9

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

Fig. 5 clarifies why this worse performance of CASE 2 occurs. Figure shows the percentage contribution of each process to the characterized impacts in CASE 2 assumption. In every impact category, the contribution is almost entirely attributable to the livestock phase, which provides a contribution greater than 90% in almost all categories, with exception of POFP. In detail, the major contributions come the feeding related processes ranging from 3% (TAP) to 88% (TEP); the shed and crops processes, ranging from 1% (TEP), to 61% (MDP). Transport has a lower impact contribution, with a peak of 8% in FDP. Local emissions show a notable contribution of 70% in GWP, 75% in FEP, 94% in TAP and 29% in POFP. Impacts contributions coming from the investigated process phase are almost always negligible. A non-negligible contribution from urea is observed in HTP (4%), FDP (6%), MDP (3%), GWP (1%) and WDP (1%), while methane contributes to HTP (2%), FDP (1%), MDP (2%), POFP (1) and WDP (2%). Local emissions from the investigated process show a valuable contributions (10%) to POFP. 3.2. Uncertainty analysis Input flows for the various scenarios analyzed so far have been treated as single mean values, but they may present some variability due to model assumptions, lack of data, inaccurate measurement, etc. Uncertainties of these parameters translate into uncertainty in the outcome of an LCA. The accuracy of foreground data collection and background database update are key factors in determining the uncertainty level and affect the final values of impact categories. In order to assess the uncertainty caused by variability in input and output data, a Monte Carlo analysis has been performed. Prior to performing the Monte Carlo analysis, a pedigree matrix was created. Results, for the selected impact categories, are summarized in Table 6. Table 6 confirms the reliability of the results achieved in this paper, even with the presence of a few large values of r and cv in some impact categories (i.e. HTP and TEP). A further discussion on this issue can be found in Appendix A. 4. Discussion

Characterized Impacts

The results presented in this paper show how the perspective used to analyze a chosen system affect the evaluation of the process performance. From a methodological point of view, it is of paramount important to make a decision about the way waste

Table 6 Results of Monte Carlo applied to the production of 1 MW h of electric energy in CASE 1. Impact category

Unit

Mean

Median

r

cv (%)

SEM

GWP HTP

kg CO2 eq kg 1,4DB eq kg P eq kg SO2 eq kg 1,4DB eq kg oil eq kg Fe eq kg NMVOC m3

84.51 88.02

83.38 81.68

8.89 1377.24

10.52 1564.68

0.28 43.55

0.02 1.27 0.01

0.01 1.26 0.01

0.01 0.07 0.04

53.67 5.35 251.48

2.86E 04 2.15E 03 1.12E 03

25.27 6.37 1.35 114.25

24.71 5.72 1.34 111.23

3.95 2.69 0.03 16.91

15.62 42.27 2.58 14.80

0.12 0.09 1.10E 03 0.53

FEP TAP TEP FDP MDP POFP WDP

r = Standard deviation. cv = Coefficient of variation. SEM = Standard error of mean.

materials should be considered in a process evaluation [4]. Although LCA is a ‘‘cradle to grave” approach, it clearly appears that waste and residues are a special category of flows that deserve a different consideration than primary input flows and nonrecyclable emissions. While designing an evaluation method [4] suitable for processes dealing with waste processing for resource recovery, it should not be disregarded that the investigated process is not a theoretical scenario, but instead a real industrial plant designed and managed within a circular economy oriented company. The process is already active and generates electric energy that is entered in the Italian grid, heat and other by-products.

4.1. Which impacts should be taken into account? In a ‘zero burden’ perspective (CASE 1 and CASE 3) the input material is accounted for as a waste-stream that does not carry any associated impact. The process is not seen as purposefully oriented to generate a ‘fuel’ (i.e. the residues) to be used for electricity generation purpose, but instead electricity is seen as an additional advantage gained when dealing with treatment and disposal of waste organic material. The impacts associated to livestock farming are all attributed to the main outputs, meat and dairy products, as it is perfectly understandable and reasonable. In fact, the reason supporting livestock farming is not electricity production (for

100% 80% 60% 40% 20% 0%

GWP

HTP

FEP

TAP

TEP

FDP

MDP

POFP

Feed (Livestock)

16%

33%

17%

3%

88%

46%

30%

33%

WDP 41%

Shed+Crops (Livestock)

9%

54%

7%

2%

1%

37%

61%

22%

45%

Local Emissions + Others (Livestock)

70%

1%

75%

94%

11%

0%

2%

29%

7%

Transportaon (Livestock)

3%

6%

0%

0%

0%

8%

3%

5%

2%

Methane

0%

2%

0%

0%

0%

1%

2%

1%

3%

Urea

1%

4%

0%

0%

0%

6%

3%

1%

2%

Local Emissions + Others (Electricity Producon)

0%

1%

0%

0%

0%

2%

1%

10%

1%

Fig. 5. Contributions to characterized impacts from different phases of CASE 2 (contributions less than 1% from the livestock phase are included in ‘Local emissions + Others (Livestock)’, while contributions less than 1% from the electricity production phase are labeled as ‘Local emissions + Others (Electricity production)’).

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

10

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

which other more efficient patterns exist) but instead food production. In so doing, by freeing waste materials from undue burdens (as with CASE 2, instead), the actual process of converting waste into electricity becomes a profitable conversion activity to yield a small amount of electricity at lower impacts (CASE 1 and CASE 3) than mainstream electricity production patterns (MIX ITA). Being livestock farming so largely impacting on final electricity production, results clearly suggest that the entire supply chain ‘livestock farming-meat and slaughterhouse residues treatment-electricity production’ is not a linear process that should be considered as a whole. Focus, and allocation procedure, must be placed primarily on the main product, meat, and only in second place can coproducts or by-products be considered and their impacts accounted for only on the basis of treatment and disposal costs to be treated as a separate process (‘zero-burden perspective’). By not charging production-chain additional impacts to the electric production, the use of electricity generated from slaughterhouse waste can be considered as less impacting than fossil based electricity, in so providing a motivation and an advantage to the process itself of energy recovery. Not only organic waste treatment for energy prevents additional difficult disposal of such animal residues, but its conversion to electricity increases the energy efficiency of the entire network of related activities (less grid electricity is needed), in so decreasing by a proportional and nonnegligible extent, the environmental impacts of electricity generation from fossil fuels at larger regional scale. 4.2. Allocation choices Allocation procedures therefore play a major role in the assessment of process performance. Firstly, it is not irrelevant to raise a question about the real motivation a process is carried out. In the case of meat production chain, the intended products are meat, milk and additional co-products (e.g. leather) and the final goal is to generate an income out of this production. When it comes to deciding which process product is responsible of impacts, it might be fully understandable to indicate that the main responsible could be the product that generates the larger income (therefore being the most important motivation for the process itself to occur). This money-based allocation is also included in the ILCD (2010) options, but the variability of market prices affecting income cannot be disregarded, thus making money-based allocation a very unstable and uncertain way to link environmental burdens to process products and by-products. Allocation according to mass of products and co-products is also very ‘weak’ in that if assumes mass as the main characteristics of products, no matter their nutritional content or other kinds of value. Mass allocation assumes all product flows have the same value. An energy-based allocation, instead, seems to capture the intended goal of the entire process to a larger extent, considering that both food products and residues are used for a broader energy purpose. This is, based on results, the most reasonable choice in the present study (CASE 3). 4.3. Crucial inflows and steps While is LCA certainly very useful to understand the impacts associated to a final or intermediate product, in so supporting comparison of products and policy/trade choices, it should not be neglected its important contribution to the analysis of a process structure and functioning, in all its steps, in order to identify performance drops and bottlenecks. In particular, step-by-step LCA allows to identify which input material or energy flows contribute the highest impacts, thus suggesting where to invest the biggest efforts for selective reduction of environmental burdens. This is certainly true also in the investigated process for conversion of slaughterhouse residues to energy. As seen in Section 4 (Fig. 4),

urea, used for the selective reduction of NOx in the electricity generation phase, is the largest contributor to almost all impact categories except TAP and POFP; methane, used for steam generation in the rendering process phase, contributes between 28% and 50% to GWP, HTP, FEP, TEP, MDP and WDP; transportation of the byproducts collected to the plant site contributes 11% to GWP and 12% to FDP; local emissions mainly from organic and fossil substrates combustion processes, contribute around 60% and 80% to TAP and POFP respectively; and finally machinery (both for transport and plant) contributes 16% to MDP. Considering that the normalized impacts show important CASE 1 contributions only in HTP, FEP and TAP categories, we can conclude that urea, methane and local emissions are the most important flows to take care of, while other flows are relatively negligible. Impact contributions under the assumptions of CASE 2 are quite different, since they include the burdens related to the livestock phase, that are highly dominating at the scale of the expanded livestock/food chain/residues recovery boundary. Shed and agricultural processes, feed and local emissions in livestock farming are the largest contributing factors to essentially all impact categories, with more or less similar proportions; instead, local emissions alone contributes between 70% and 94% to GWP and TAP, with minor contributions from other factors; feed is finally the largely dominating factor in the TEP category, with a percentage around 88%. These results confirm previous studies about agricultural production systems, that show the largest input flows (and associated impacts) to a conventional farm be represented by diesel, electricity, fertilizers and purchased feedstuff [61]. If CASE 2 is looked at from the point of view of the entire agroindustrial process with all its products, co-products and byproducts, the emerging picture is that the recovery of slaughterhouse waste is only the last step of a much larger system (farming & livestock + CASE 1), where recovery of electricity and feedstock provides very small impacts compared to the entire chain. In order to increase the overall sustainability of the system, focus should also be placed on intensive livestock farming that is a consequence of a too protein-rich diet in our urban systems. Within such a picture, recovery of residues can be considered as an increase of efficiency in resource use at the very end of the production chain, with possibility of feedback loops to the earlier stages of the process. Table 5 shows that, in order to produce 1 MW h of electric energy, 288 kg of oil equivalent are required (FDP) within CASE 2 approach, translating into a low 30% conversion efficiency. This is not comparable, of course, with the average efficiency of the Italian electricity mix (45%, according to Table 4), but still represents a non-negligible potential energy recovery for the system of CASE 2 within a circular perspective. If a like calculation is performed with CASE 3 (FDP = 28 kg oil equiv per MW h generated, Table 5), i.e. with reference only to the energy needed for the recovery and conversion of residues in a zero burden approach, the electricity generated is about 3 times the actual fossil energy investment in the process, amplified to about 6.6 times if the comparison is drawn with the fossil energy saved from the Italian mix considered in Table 4. Considering the results obtained in the present study, the investigated processes could further be improved, under an environmental point of view, by increasing energy use efficiency, by reducing the use of fossil fuels (i.e. methane and diesel), and, when possible, by switching to bio-fuels [36]. Bio-methane, potentially produced in the livestock phase, may replace methane used for steam-generation in the rendering process, while bio-diesel, produced from collection and refinement of waste cooking oil in the same investigated Company [45], may be used in diesel engines (as such or blended with diesel) [32], to substitute the diesel fuel used for truck transportation of slaughterhouse residues as well as to clean the marine-engine in the Company’s co-generation

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

plant. Electric vehicles (or, at least, hybrid vehicles) might also be used in the process, fueled by a larger share of the electricity produced, before delivering it to the Italian grid. 4.4. A circular economy framework This study was performed keeping in mind an idea of circularity [62]. Waste prevention, efficiency increase, resource exchange, reuse and recycle across scales, as well as eco-design of processes and products for easy optimization of resource use, are all concepts and tools that contribute to get out of the old paradigm ‘take, make and dispose’ towards a more environmentally sound production and consumption system. This is expected to make better use of the available resources, to allow increased wellbeing at no or low environmental costs, i.e. the implementation of a system in which products are created in a perspective of use and reuse, disassembling and recycling, gaining resources from ‘waste’ materials, minimizing the extraction of new raw materials [63]. Obviously, animals are not bred to produce electricity. The results of this study should not be intended as the proposal for an electricity production process, but instead as a way of increasing the sustainability of the larger scale process through material and energy loops, reusing the by-products generated by the meat industry, retrieving the residual value of something destined otherwise to be managed as ‘waste’, and finally providing a link between the production phase (primarily rural) and the consumers (primarily urban). Studies prove how meat consumption rises with the developing of urbanization, especially in developing countries (from 14 kg/capita to 25 kg/capita between 1983 and 1997, expected to reach 36 kg/capita in 2020) [64]. The world average meat consumption in 2015 is 43.4 kg/capita [65], equivalent to roughly 40 kg/capita of byproducts produced. It is therefore very clear that an appropriate use of by-products is urgent and mandatory. Energy and feedstock recovery are promising options, although other circular pathways (e.g. extraction of chemicals) should not be excluded, to generate feedback loops at larger scale [41]. In this work, the focus is set on exploiting the still valuable content of byproducts, by means of a practical case study. Indeed, the investigated waste refinery uses rendering treatment to produce two outputs from the incoming waste: fat as a fuel for electricity and a solid fraction (meal) for the production of chemical building blocks, a more environmentally sustainable and economically profitable alternative than the corresponding fossil-based chemistry. When resources become scarce, this behavior translates into a triple advantage merging together waste reduction, prevention of resources’ depletion and renewable energy production, all falling under the common umbrella of circular economy framework. The development of Circular Economy is part of the China Governmental policies [62] as a holistic top-down approach. The European Commission issued a Circular Economy package only in 2014, recently replaced by an ‘‘ambitious new Circular Economy Package to stimulate Europe’s transition towards a circular economy” [66]. The new CE package defines ‘circular economies’ as something in which ‘‘the value of products and materials is maintained for as long as possible; waste and resource use are minimized, and resources are kept within the economy when a product has reached the end of its life, to be used again and again to create further value” [67]. In the new plan, the European Commission has recognized the importance of ‘‘closing the loop” of product life-cycles through increased recycling and re-use, fostering energy savings and reducing GHG emissions. The Action Plan will include, among other actions, funding of over €6 billion; actions in order to halve food waste by 2030; development of quality standards for secondary raw materials and revised legislative proposals on waste, in order to achieve, among other goals, a 65% recycling of MSW and a 75% recycling of packaging waste by 2030; reduce landfill to maximum

11

of 10% of all waste by 2030 and ban of landfilling of separately collected waste. 5. Conclusions The life cycle perspective adopted in this study allowed a comprehensive assessment of the environmental impacts and benefits of the electricity production from AFW, exploring constraints and potentialities of the investigated system. In this paper, the impacts related to the real scale production of electricity from animal fats, obtained from a rendering process of animal by-products, was shown. The presented results prove that the ‘zero-burden’ approach to waste disposal is the most reasonable framework for dealing with waste treatment and conversion to useful output flows of energy and matter, and that, in such perspective, the electricity obtained is more environmentally sound than the average grid electricity mix. The LCA method allowed to identify the most crucial steps and inflows to the process, for potential improvement. The investigated process shows to be capable to process animal residues and to separate the protein fraction destined to animal meal and chemicals, from a residual animal fat fraction destined to electricity generation, within a bio-refinery perspective company. The assessment of costs and benefits via LCA shows that recovery of electricity and matter is beneficial from both environmental and energy points of view and suggests further steps towards increased circularity. Results are based on a real case plant, which makes them much stronger and reliable than just a feasibility scenario. A circular economy and technology framework is advocated in the study, where resource use is optimized over the entire production chain in order to make the best out of limited resources. Such efficiency increase is foreseen within a new paradigm for sustainable production and consumption, where lifestyles for resource optimization (e.g. diet, in the investigated case) are also an important aspect of the expected environmental improvement, to complement technological options for energy and material recovery. Acknowledgements Maddalena Ripa and Sergio Ulgiati gratefully acknowledge the financial support received from the European Union Project EUFORIE - European Futures for Energy Efficiency, funded under EU Horizon 2020 programme, call identifier H2020-EE-2014-2-RIA and topic EE-12-2014, Socio-economic research on energy efficiency. The authors also acknowledge the support of Proteg S.P.A. in collecting data. Sergio Ulgiati acknowledges the contract by the School of Environment, Beijing Normal University, within the framework of the National ‘‘One Thousand Foreign Experts Plan”. Appendix A Uncertainty analysis is not a mandatory component of LCA method. According to the ILCD Handbook, the construction and analysis of the systems involves potential sources of uncertainty. Three main sources of uncertainty can be identified: stochastic uncertainty, choice uncertainty, lack of knowledge. In this study we mainly focus on the former. The stochastic uncertainty can be assessed in two fundamentally different ways – analytical solution or simulation. Monte-Carlo Simulation is an especially suitable method to quantify stochastic uncertainty in LCA. However, the outcome of a stochastic uncertainty calculation also shows a high degree of uncertainty, in that it does not capture systematic uncertainty and gaps in both modeling and data [48], and therefore should not be over-interpreted. The results of the Monte-Carlo

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

12

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

Table A1 Monte Carlo analysis results of all the ReCiPe impact categories calculated for CASE 1. # 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Impact category

Unit 2

Agricultural land occupation Climate change Fossil depletion Freshwater ecotoxicity Freshwater eutrophication Human toxicity Ionising radiation Marine ecotoxicity Marine eutrophication Metal depletion Natural land transformation Ozone depletion Particulate matter formation Photochemical oxidant formation Terrestrial acidification Terrestrial ecotoxicity Urban land occupation Water depletion

m a kg CO2 eq kg oil eq kg 1,4-DB eq kg P eq kg 1,4-DB eq kBq U235 eq kg 1,4-DB eq kg N eq kg Fe eq m2 kg CFC-11 eq kg PM10 eq kg NMVOC kg SO2 eq kg 1,4-DB eq m2 a m3

Mean

Median

r

cv (%)

SEM

3.94 84.51 25.27 1.00 0.02 88.02 6.75 0.97 0.06 6.37 0.02 3.83E 06 0.43 1.35 1.27 0.01 0.99 114.25

3.72 83.38 24.71 0.96 0.01 81.68 5.39 0.93 0.06 5.72 0.02 3.55E 06 0.43 1.34 1.26 0.01 0.92 111.23

1.32 8.89 3.95 5.13 0.01 1377.24 5.06 4.15 2.63E 03 2.69 0.01 1.52E 06 0.02 0.03 0.07 0.04 0.38 16.91

33.58 10.52 15.62 511.49 53.67 1564.68 74.97 428.57 4.08 42.27 58.78 39.66 4.58 2.58 5.35 251.48 38.38 14.80

0.04 0.28 0.12 0.16 2.86E 43.55 0.16 0.13 8.30E 0.09 3.52E 4.80E 6.26E 1.10E 2.15E 1.12E 1.21E 0.53

04

05 04 08 04 03 03 03 02

r = Standard deviation. cv = Coefficient of variation. SEM = Standard error of mean.

7, 10, 18). These categories suggest the process to be source of some concern and the need to manage the process chain carefully for impact reduction. (4) Only one impact category showing medium impact value but extremely high standard deviation r (category 6). This behavior is, indeed, the only one that should create a real concern to the process managers, although the high standard deviation is shared with all the other ecotoxicity related categories (categories 4, 6, 8, 16) as it can be easily inferred looking at the coefficient of variation cv.

analysis presented in this paper (Table 6) fall perfectly within this stochastic uncertainty situation. In fact, Table A1 summarizes Monte-Carlo results for all the impact categories calculated by the ReCiPe method, with the possibility to observe four kinds of uncertainty situation: (1) Impact categories presenting both relatively small impacts (mean/median) and standard deviations, r, smaller than the impact (categories 1, 5, 9, 11, 12, 13, 14, 15, 17). These results suggest that the process does not contribute to any significant extent to life cycle environmental loading. (2) Impact categories showing relatively small impacts (mean/ median) but r values higher than the impacts (categories 4, 8, 16). These cases also suggest very small impact potential and should not worry stakeholders and policy makers, even assuming the impact to reach the largest possible value. (3) Impact categories with medium impact value (mean/ median), compared with literature (Table A2) and standard deviations r smaller than the impact itself (categories 2, 3,

Table A2 shows the results of a Monte-Carlo simulation for the generation of 1 MW h of electricity from the Italian grid as well as from a biogas powered cogeneration plant, according to the Ecoinvent database. In both cases, large values of standard deviation for ecotoxicity related impact categories are observed. In particular, Human Toxicity shows much larger values of r, thus calling for additional studies about the assessment method for this category, to decrease the global uncertainty related to its impact calculation.

Table A2 Monte-Carlo simulation of the generation of 1 MW h of electricity from the Italian grid and from a biogas powered cogeneration plant. Impact category

Agricultural land occupation Climate change Fossil depletion Freshwater ecotoxicity Freshwater eutrophication Human toxicity Ionising radiation Marine ecotoxicity Marine eutrophication Metal depletion Natural land transformation Ozone depletion Particulate matter formation Photochemical oxidant formation Terrestrial acidification Terrestrial ecotoxicity Urban land occupation Water depletion

Unit

m2 a kg CO2 eq kg oil eq kg 1,4-DB eq kg P eq kg 1,4-DB eq kBq U235 eq kg 1,4-DB eq kg N eq kg Fe eq m2 kg CFC-11 eq kg PM10 eq kg NMVOC kg SO2 eq kg 1,4-DB eq m2 a m3

Mix Ita

Biogas plant

Mean

r

Mean

r

10.32 611.23 187.75 4.82 0.10 712.97 93.99 4.32 0.08 15.97 0.10 0.00 0.73 1.43 2.26 0.10 1.88 3126.95

1.84 62.28 33.20 92.66 0.06 25003.70 109.93 75.02 0.01 3.92 0.04 0.00 0.08 0.25 0.25 0.64 0.76 276.32

746.12 291.90 58.84 790.57 0.09 214433.64 31.44 641.42 2.24 15.10 0.03 0.00 0.79 1.29 3.20 6.34 3.83 639.33

105.93 1109.72 386.12 19291.66 0.61 5206812.02 220.24 15619.98 1.31 124.67 0.17 0.00 3.03 4.66 10.07 133.46 9.86 2928.02

r = Standard deviation. Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx

References [1] Conca K, Dabelko G. Green planet blues: critical perspectives on global environmental politics. Westview Press; 2014. [2] Sharma SS. Determinants of carbon dioxide emissions: empirical evidence from 69 countries. Appl Energy 2011;88:376–82. http://dx.doi.org/10.1016/j. apenergy.2010.07.022. [3] Brown MT, Ulgiati S. Emergy evaluations and environmental loading of electricity production systems. J Clean Prod 2002;10:321–34. http://dx.doi. org/10.1016/S0959-6526(01)00043-9. [4] Bala Gala A, Raugei M, Ripa M, Ulgiati S. Dealing with waste products and flows in life cycle assessment and emergy accounting: methodological overview and synergies. Ecol Modell 2015. http://dx.doi.org/10.1016/j. ecolmodel.2015.03.00. [5] Owen NA, Inderwildi OR, King DA. The status of conventional world oil reserves—hype or cause for concern? Energy Policy 2010;38:4743–9. http://dx. doi.org/10.1016/j.enpol.2010.02.026. [6] Talebian-Kiakalaieh A, Amin NAS, Mazaheri H. A review on novel processes of biodiesel production from waste cooking oil. Appl Energy 2013;104:683–710. http://dx.doi.org/10.1016/j.apenergy.2012.11.061. [7] Hoggett R, Bolton R, Candelise C, Kern F, Mitchell C, Yan J. Supply chains and energy security in a low carbon transition. Appl Energy 2014;123:292–5. http://dx.doi.org/10.1016/j.apenergy.2014.02.008. [8] United Nations Statistics Division. Municipal waste treatmentAvailable from: 2011 [accessed December 7, 2015]. [9] Eurostat. Municipal waste generation and treatment, by type of treatment methodAvailable from: 2015 [accessed December 7, 2015]. [10] Palmiotto M, Fattore E, Paiano V, Celeste G, Colombo A, Davoli E. Influence of a municipal solid waste landfill in the surrounding environment: toxicological risk and odor nuisance effects. Environ Int 2014;68:16–24. http://dx.doi.org/ 10.1016/j.envint.2014.03.004. [11] Dong J, Chi Y, Zou D, Fu C, Huang Q, Ni M. Energy–environment–economy assessment of waste management systems from a life cycle perspective: model development and case study. Appl Energy 2014;114:400–8. http://dx. doi.org/10.1016/j.apenergy.2013.09.037. [12] Martinez-Sanchez V, Kromann MA, Astrup TF. Life cycle costing of waste management systems: overview, calculation principles and case studies. Waste Manage 2015;36:343–55. http://dx.doi.org/10.1016/j.wasman.2014.10. 033. [13] Giampietro M, Martin JR, Ulgiati S. Can we break the addiction to fossil energy? Energy 2012;37:2–4. http://dx.doi.org/10.1016/j.energy.2011.11.001. [14] EC. Directive 2008/98/EC on waste and repealing certain Directives; 2008. [15] Rockström J, Steffen W, Noone K, Persson A, Chapin FS, Lambin EF, et al. A safe operating space for humanity. Nature 2009;461:472–5. http://dx.doi.org/ 10.1038/461472a. [16] EU. Review of the EU sustainable development strategy (EU SDS); 2006. [17] Eurostat. Greenhouse gas emission statistics – statistics explainedAvailable from: 2012 [accessed October 15, 2015]. [18] Evangelisti S, Lettieri P, Clift R, Borello D. Distributed generation by energy from waste technology: a life cycle perspective. Process Safe Environ Prot 2015;93:161–72. http://dx.doi.org/10.1016/j.psep.2014.03.008. [19] Pan S-Y, Du MA, Huang I-T, Liu I-H, Chang E-E, Chiang P-C. Strategies on implementation of waste-to-energy (WTE) supply chain for circular economy system: a review. J Clean Prod 2015:1–13. http://dx.doi.org/10.1016/j. jclepro.2015.06.124. [20] Tan ST, Hashim H, Lim JS, Ho WS, Lee CT, Yan J. Energy and emissions benefits of renewable energy derived from municipal solid waste: analysis of a low carbon scenario in Malaysia. Appl Energy 2014;136:797–804. http://dx.doi. org/10.1016/j.apenergy.2014.06.003. [21] Yeo I-A, Yoon S-H, Yee J-J. Development of an Environment and energy Geographical Information System (E-GIS) construction model to support environmentally friendly urban planning. Appl Energy 2013;104:723–39. http://dx.doi.org/10.1016/j.apenergy.2012.11.053. [22] Marcos A, Al-Kassir A, Mohamad AA, Cuadros F, López-Rodríguez F. Combustible gas production (methane) and biodegradation of solid and liquid mixtures of meat industry wastes. Appl Energy 2010;87:1729–35. http://dx.doi.org/10.1016/j.apenergy.2009.09.037. [23] Chakraborty R, Gupta AK, Chowdhury R. Conversion of slaughterhouse and poultry farm animal fats and wastes to biodiesel: parametric sensitivity and fuel quality assessment. Renew Sustain Energy Rev 2014;29:120–34. http:// dx.doi.org/10.1016/j.rser.2013.08.082. [24] Adewale P, Dumont M-J, Ngadi M. Recent trends of biodiesel production from animal fat wastes and associated production techniques. Renew Sustain Energy Rev 2015;45:574–88. http://dx.doi.org/10.1016/j.rser.2015.02.039. [25] Alptekin E, Canakci M, Ozsezen AN, Turkcan A, Sanli H. Using waste animal fat based biodiesels–bioethanol–diesel fuel blends in a DI diesel engine. Fuel 2015;157:245–54. http://dx.doi.org/10.1016/j.fuel.2015.04.067. [26] Behçet R, Oktay H, Çakmak A, Aydin H. Comparison of exhaust emissions of biodiesel–diesel fuel blends produced from animal fats. Renew Sustain Energy Rev 2015;46:157–65. http://dx.doi.org/10.1016/j.rser.2015.02.015.

13

[27] EC. REGULATION (EC) No 1774/2002 laying down health rules concerning animal by-products not intended for human consumption; 2002. [28] Haines RJ. Farm to fork – a strategy for meat safety in Ontario – report of the meat regulatory and inspection review; 2004. [29] Kumar MS, Kerihuel A, Bellettre J, Tazerout M. A comparative study of different methods of using animal fat as a fuel in a compression ignition engine. J Eng Gas Turbines Power 2006;128. http://dx.doi.org/10.1115/1.218027. [30] Ariyaratne WKH, Melaaen MC, Eine K, Tokheim LA. Meat and bone meal as a renewable energy source in cement kilns: investigation of optimum feeding rate. In: Int conf renew energies power qual. Las Palmas de Gran Canaria: European Association for the Development of Renewable Energies, Environment and Power Quality; 2010. [31] Marvuglia A, Benetto E, Rege S, Jury C. Modelling approaches for consequential life-cycle assessment (C-LCA) of bioenergy: critical review and proposed framework for biogas production. Renew Sustain Energy Rev 2013;25:768–81. http://dx.doi.org/10.1016/j.rser.2013.04.031. [32] Varuvel EG, Mrad N, Tazerout M, Aloui F. Assessment of liquid fuel (bio-oil) production from waste fish fat and utilization in diesel engine. Appl Energy 2012;100:249–57. http://dx.doi.org/10.1016/j.apenergy.2012.05.035. [33] Rubio Rodríguez MA, De Ruyck J, Díaz PR, Verma VK, Bram S. An LCA based indicator for evaluation of alternative energy routes. Appl Energy 2011;88:630–5. http://dx.doi.org/10.1016/j.apenergy.2010.08.01. [34] Ebner J, Babbitt C, Winer M, Hilton B, Williamson A. Life cycle greenhouse gas (GHG) impacts of a novel process for converting food waste to ethanol and coproducts. Appl Energy 2014;130:86–93. http://dx.doi.org/10.1016/j. apenergy.2014.04.099. [35] Keller H, Rettenmaier N, Reinhardt GA. Integrated life cycle sustainability assessment – a practical approach applied to biorefineries. Appl Energy 2015;154:1–10. http://dx.doi.org/10.1016/j.apenergy.2015.01.095. [36] Pöschl M, Ward S, Owende P. Evaluation of energy efficiency of various biogas production and utilization pathways. Appl Energy 2010;87:3305–21. http:// dx.doi.org/10.1016/j.apenergy.2010.05.011. [37] Montefrio MJ, Xinwen T, Obbard JP. Recovery and pre-treatment of fats, oil and grease from grease interceptors for biodiesel production. Appl Energy 2010;87:3155–61. http://dx.doi.org/10.1016/j.apenergy.2010.04.011. [38] Browne JD, Murphy JD. Assessment of the resource associated with biomethane from food waste. Appl Energy 2013;104:170–7. http://dx.doi. org/10.1016/j.apenergy.2012.11.017. [39] Cao L, Wang J, Liu C, Chen Y, Liu K, Han S. Ethylene vinyl acetate copolymer: a bio-based cold flow improver for waste cooking oil derived biodiesel blends. Appl Energy 2014;132:163–7. http://dx.doi.org/10.1016/j.apenergy.2014.06.085. [40] Noshadi I, Kanjilal B, Du S, Bollas GM, Suib SL, Provatas A, et al. Catalyzed production of biodiesel and bio-chemicals from brown grease using Ionic Liquid functionalized ordered mesoporous polymer. Appl Energy 2014;129:112–22. http://dx.doi.org/10.1016/j.apenergy.2014.04.090. [41] Shahzad K. Comparative normative analysis of sustainability measurement methods. Austria: Graz University of Technology; 2014. [42] Riedel SL, Jahns S, Koenig S, Bock MCE, Brigham CJ, Bader J, et al. Polyhydroxyalkanoates production with Ralstonia eutropha from low quality waste animal fats. J Biotechnol 2015;214:119–27. http://dx.doi.org/10.1016/j. jbiotec.2015.09.002. [43] Kettl K-H, Shahzad K, Eder M, Narodoslawsky M. Ecological footprint comparison of biobased PHA production from animal residues. Chem Eng Trans 2012;29:439–44. http://dx.doi.org/10.3303/CET1229074. [44] Jayathilakan K, Sultana K, Radhakrishna K, Bawa AS. Utilization of byproducts and waste materials from meat, poultry and fish processing industries: a review. J Food Sci Technol 2012;49:278–93. http://dx.doi.org/10.1007/s13197011-0290-7. [45] Ripa M, Buonaurio C, Fiorentino G, Mellino S, Ulgiati S. Recycling waste cooking oil into biodiesel: a life cycle assessment. Int J Performability Eng 2014;10:347–56. [46] ISO 14040. Environmental management—life cycle assessment—principles and framework; 2006. [47] ISO 14044. Environmental management–life cycle assessment–requirements and guidelines; 2006. [48] JRC. International Reference Life Cycle Data System (ILCD) handbook – general guide for life cycle assessment – detailed guidance. Luxembourg: Publications Office of the European Union; 2010. http://dx.doi.org/10.2788/38479. [49] Pennington DW, Potting J, Finnveden G, Lindeijer E, Jolliet O, Rydberg T, et al. Life cycle assessment Part 2: Current impact assessment practice. Environ Int 2004;30:721–39. http://dx.doi.org/10.1016/j.envint.2003.12.009. [50] Ulgiati S, Raugei M, Bargigli S. Overcoming the inadequacy of single-criterion approaches to Life Cycle Assessment. Ecol Modell 2006;190:432–42. http://dx. doi.org/10.1016/j.ecolmodel.2005.03.022. [51] Itten R, Frischknecht R, Stucki M. Life cycle inventories of electricity mixes and grid. Uster; 2012. [52] Goedkoop M, Heijungs R, Huijbregts M, DeSchryver A, Struijs J, VanZelm R. ReCiPe 2008—a life cycle impact assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. Report I: characterisation. Den Haag; 2009. [53] Frischknecht R, Jungbluth N, Althaus HJ, Bauer C, Doka G, Dones R, et al. Implementation of life cycle impact assessment methods – ecoinvent report n 3. Dübendorf; 2007. [54] Sleeswijk AW, van Oers LFCM, Guinée JB, Struijs J, Huijbregts MAJ. Normalisation in product life cycle assessment: an LCA of the global and

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073

14

[55] [56]

[57] [58]

[59]

[60]

R. Santagata et al. / Applied Energy xxx (2016) xxx–xxx European economic systems in the year 2000. Sci Total Environ 2008;390:227–40. http://dx.doi.org/10.1016/j.scitotenv.2007.09.040. Steen B. On uncertainty and sensitivity of LCA-based priority setting. J Clean Prod 1997;5:255–62. http://dx.doi.org/10.1016/S0959-6526(97)00039-5. Ekvall T, Assefa G, Björklund A, Eriksson O, Finnveden G. What life-cycle assessment does and does not do in assessments of waste management. Waste Manage 2007;27:989–96. http://dx.doi.org/10.1016/j.wasman.2007.02.015. Björklund AE. Survey of approaches to improve reliability in lca. Int J Life Cycle Assess 2002;7:64–72. http://dx.doi.org/10.1007/BF02978849. Hofstetter P. Perspectives in life cycle impact assessment – a structured approach to combine models of the technosphere, ecosphere and valuesphere. Boston (MA): Springer US; 1998. http://dx.doi.org/10.1007/9781-4615-5127-0. Limpert E, Stahel WA, Abbt M. Log-normal distributions across the sciences: keys and clues. Bioscience 2001;51:341. http://dx.doi.org/10.1641/0006-3568 (2001)051[0341:LNDATS]2.0.CO;2. Weidema BP, Wesnæs MS. Data quality management for life cycle inventories – an example of using data quality indicators. J Clean Prod 1996;4:167–74. http://dx.doi.org/10.1016/S0959-6526(96)00043-1.

[61] Jaklicˇ T, Juvancˇicˇ L, Kavcˇicˇ S, Debeljak M. Complementarity of socio-economic and emergy evaluation of agricultural production systems: the case of Slovenian dairy sector. Ecol Econ 2014;107:469–81. http://dx.doi.org/ 10.1016/j.ecolecon.2014.09.024. [62] Ghisellini P, Cialani C, Ulgiati S. A review on circular economy: the expected transition to a balanced interplay of environmental and economic systems. J Clean Prod 2015. http://dx.doi.org/10.1016/j.jclepro.2015.09.007. [63] Ellen MacArthur Foundation. Towards the Circular Economy, vol. 1 – An economic and business rationale for an accelerated transition; 2012. [64] Delgado CL. Rising consumption of meat and milk in developing countries has created a new food revolution. J Nutr 2003;133:3907S–10S. [65] FAO. Food Outlook – Biannual report on global food markets; 2015. [66] European Commission. Press release – closing the loop: Commission adopts ambitious new Circular Economy Package to boost competitiveness, create jobs and generate sustainable growthAvailable from: 2015. [67] European Commission. Circular economy package : questions & answersAvailable from: 2015.

Please cite this article in press as: Santagata R et al. An environmental assessment of electricity production from slaughterhouse residues. Linking urban, industrial and waste management systems. Appl Energy (2016), http://dx.doi.org/10.1016/j.apenergy.2016.07.073