Life cycle assessment of orange peel waste management

Resources, Conservation & Recycling 127 (2017) 148–158

Contents lists available at ScienceDirect

Resources, Conservation & Recycling journal homepage: www.elsevier.com/locate/resconrec

Full length article

Life cycle assessment of orange peel waste management a,⁎

a

a

Viviana Negro , Bernardo Ruggeri , Debora Fino , Davide Tonini a b

MARK

b

Department of Applied Science and Technology, Politecnico di Torino, Corso Duca degli Abruzzi 24, Torino 10129, Italy Department of Environmental Engineering, Technical University of Denmark, 2800 Kongens Lyngby, Denmark

A R T I C L E I N F O

A B S T R A C T

Keywords: Incineration Pyrolysis Anaerobic digestion Animal feeding Limonene

The management of orange peel waste constitutes an economic and environmental problem in regions in which there are important citrus processing industries, as is the case of southern Italy. Traditional handling techniques are either not economically attractive (e.g. composting and animal feeding) or discouraged by European policy (landfilling). As an alternative to these technologies, others aimed at recovering energy and resources are currently receiving increasing attention. The consequential life cycle assessment adopted in this work compares the environmental performance of ten orange peel waste management scenarios. These include mono-treatment scenarios (pyrolysis, incineration, and anaerobic mono-digestion) and co-treatment (four anaerobic co-digestion strategies with animal manure and seaweed) ones aimed at energy/resource recovery, which were compared with three traditional non-energy focused handling techniques (landfilling, composting and animal feeding). Overall, the co-digestion scenarios appear to be the best, in terms of global warming and resource depletion mitigation. However, they also suffer from a drawback, that is, a potential eutrophication impact, due to nitrate leaching following on-land digestate use. Orange peel waste use for animal feeding, while appearing interesting from an environmental perspective (for example to reduce meal imports), presents practical challenges as far as the nutritional aspects and costs are concerned, and these eventually hinder its market potential. A preliminary cost flow analysis has concluded that anaerobic digestion strategies are economically preferable to the other alternatives.

1. Introduction In the European Union, orange production is concentrated in the Mediterranean area, with more than 6 million tonnes gathered each year in Spain, Italy, Greece and Portugal (USDA, 2013). About 30% of this production occurs in Italy, with a corresponding generation of a voluminous waste stream (about 0.6 million tonnes of orange waste) (Ferrari et al., 2016). Orange waste constitutes approximately 50–60% w/w (wet weight) of the processed fruit (Wilkins et al., 2007), and it is 60–65% w/w composed of peels, 30–35% w/w of internal tissue and the remaining share of seeds (Crawshaw, 2003). Currently, traditional solutions for orange peel waste (OPW) management (landfilling, composting, pectin extraction, animal feeding) are not economically attractive, since they present many drawbacks (Yoo et al., 2011). For example, as far as animal feeding is concerned, the high energy demand for the dehydration process, its bitterness and its low nutritional value currently discourage the use of citrus waste as an animal feed. Composting is economically costly, and the compost produced is often not of interest on the local market. The landfilling of organic waste is discouraged and should be minimized according to the

⁎

requirements of the EU landfilling directive (EC, 1999). On the other hand, citrus waste may be valorised through energy-focused treatments aiming at optimizing the recovery of energy and resources from this food-industry residual biomass, as suggested in the European resource and bio-economy strategies (de Besi and McCormick, 2015; de Man and Friege, 2016). These energy-focused treatments encompass both biological and thermochemical technologies. If biological processes are of concern, two alternatives are then applicable: extraction and removal of D-limonene from the OPW prior to the subsequent anaerobic monodigestion process (Negro et al., 2016a) or, alternatively, an anaerobic co-digestion treatment in order to dilute the concentration of D-limonene. Forgács et al. (2012) proved that D-limonene has an inhibitory effect on anaerobic digestion: at concentrations of 400 μL L−1, D-limonene affects mesophilic anaerobic digestion, while the thermophilic process shows inhibition in the range between 450 and 900 μL L−1. When thermochemical processes are of concern, the high moisture content of OPW often discourages their implementation (Ruiz and Flotats, 2016). In this context, a pre-hydration step is needed, prior to the thermochemical processes, as indicated in previous studies (Miranda et al., 2009; Siles et al., 2016; Volpe et al., 2015) that

Corresponding author. E-mail address: [email protected] (V. Negro).

http://dx.doi.org/10.1016/j.resconrec.2017.08.014 Received 12 January 2017; Received in revised form 22 June 2017; Accepted 16 August 2017 0921-3449/ © 2017 Elsevier B.V. All rights reserved.

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

relevant for this study. For example, ozone depletion was not considered, as this is mainly associated with the release of CFCs (not a relevant issue in this context), photochemical ozone formation, as this is mainly related to urban smog (not a relevant issue in this context), and metal depletion as metal use/recovery is not an issue in the studied system. The corresponding characterization methods were based on the recommendations of Hauschild et al. (2013): global warming (GW) was quantified according to IPCC 2007 (Forster et al., 2007), acidification (AC) was calculated as the accumulated exceedance, that is, according to Seppälä et al. (2006), marine nitrogen-eutrophication (EP(N)) was quantified in line with the EUTREND method (ReCiPe, 2008), toxicity to humans in relation to carcinogenic substances (HTc) and ecotoxicity to freshwater (ET) were evaluated according to the ECOtox model (Rosenbaum et al., 2008), and abiotic depletion of fossil resources (AD fossil) was quantified according to CML 2002 (Guinée et al., 2002). The environmental impacts were quantified for a 100-year time horizon. Furthermore, the environmental impacts due to capital goods were not included in this assessment because of a lack of data. The assessment was conducted with the EASETECH LCA-model, which was specifically developed for the modelling of waste and energy technologies (Clavreul et al., 2014).

investigated the use of OPW as a feedstock for combustion and pyrolysis processes. While a number of studies have focused on the conversion of OPW to biofuels through experimental tests (e.g., El-Shimi et al., 1992; Forster-Carneiro et al., 2013; Negro et al., 2016b), to the best of the Authors’ knowledge only one study has so far addressed citrus waste management from an environmental perspective (Pourbafrani et al., 2013). However, in that study, the authors only considered two biorefinery scenarios that were aimed at the production of bioethanol, biogas and D-limonene. Thermal treatments, co-digestion strategies and traditional disposal techniques were not evaluated. Furthermore, no cost analysis was included. In an attempt to provide deeper insight into OPW management, the aim of the present study has been to quantify the environmental impacts and economic costs that arise from different OPW management strategies and, on the basis of the results, to provide local authorities and decision-makers with recommendations for the optimal management of this biomass waste. To this aim, we applied a consequential life cycle assessment methodology to assess the environmental impacts of ten OPW management scenarios, using southern Italy as a case study. The investigated scenarios included: mono-treatments (incineration, pyrolysis and anaerobic mono-digestion), co-treatments (four different anaerobic co-digestion scenarios with animal manure and seaweed) and traditional handling scenarios (landfilling, composting and animal feeding), which were here used as a reference for comparison purposes. Additionally, a cost flow analysis was performed to estimate the preliminary costs associated with the ten investigated scenarios.

2.2. Orange peel waste (OPW) management scenarios Three biomass substrates were considered in this assessment: OPW, animal manure and seaweed. The latter two are needed as co-substrates for co-digestion strategies in order to dilute the concentration of toxicants and prevent process inhibition. The related chemical composition is reported in Table S1 (Supporting Information). Ten OPW management scenarios were assessed: three were mono-treatments, four were co-treatments (with the use of the co-substrates) and three traditional non-energy focused techniques. Overall, ten OPW management scenarios were assessed (Fig. 1): (i) pyrolysis with tar upgrading for biofuel production (PYR), (ii) incineration with electricity production (INC), (iii) extraction of D-limonene and anaerobic mono-digestion with biogas combustion in a stationary engine for electricity production (EXT + AD), (iv) anaerobic co-digestion of OPW and manure with biogas combustion in a stationary engine for electricity production (CD1), (v) anaerobic co-digestion of OPW and manure with biogas upgrading to biomethane for use in vehicles (CD1 + UP), (vi) anaerobic co-digestion of OPW, manure and seaweed with biogas combustion in a stationary engine for electricity production (CD2), (vii) anaerobic codigestion of OPW, manure and seaweed with biogas upgrading to biomethane for use in vehicles (CD2 + UP), (viii) conventional landfilling with flaring (LANDF), (ix) direct composting with bio-filter (COMP) and (x) animal feeding (FEED).

2. Methodology 2.1. Scope and functional unit The environmental impacts of the OPW management scenarios were quantified using a consequential life cycle assessment (LCA). Consequential LCA is a useful tool to assess the environmental performance of alternative scenarios and to identify critical environmental consequences associated with management strategies (Finnveden et al., 2009; Weidema et al., 2009). In this case study, the changes that could be induced on the energy and feed markets through the management of OPW (e.g. due to the production of electricity, bio-methane, or animal feeds) were expected to be “small enough” (infinitesimal) not to change the overall market trends, in this way justifying the application of a consequential approach (Ekvall et al., 2016). For example, if the whole amount of OPW (0.6 Mt per year, i.e. 0.14 Mt dry matter) was used as an animal feed, this would still represent less than 1.5% of the overall energy-feed demand for Italy (more than 10 Mt dry matter from corn and wheat; USDA, 2013). The LCA was performed according to the principles outlined in the ISO standards (ISO 14040-44, 2006), using system expansion to handle multi-functional processes as this technology fulfils the waste management service but also recovers energy, resources and products. The functional unit of the assessment was the management of 1 t of OPW (wet weight). Conforming with typical waste management LCAs, we used a “zero burden” approach, i.e. the activities related to the generation of the OPW were not taken into account, as they were the same for all of the investigated waste management scenarios. A middle-term temporal scope (2015–2030) was considered for the choice of waste treatment technologies (efficiencies and emissions) and background information (e.g. displaced technologies and products, transport distances, legislative context), while the south of Italy was focused on as the geographic scope of the analysis. The environmental impact categories considered in this assessment were selected according to the recommendations for relevant categories to be addressed in biomass/bioenergy LCA from Broeren et al. (2017). On this basis, we addressed: global warming, acidification, marine nitrogen-eutrophication, toxicity to humans, toxicity to ecosystems and abiotic resource depletion (the latter mainly concerns fossil fuel depletion). Other categories were disregarded as they were not considered

2.3. System boundaries In line with the normal practice in consequential LCA, the energy carriers and products generated along with the management of the OPW were assumed to substitute the corresponding energy carriers/ products produced through conventional market technologies, i.e. the system boundary was expanded to account for the benefits of avoiding the production and supply of these products. These conventional market technologies/products, in consequential LCA, should be identified as “marginal technologies/products”, i.e. those that are able to react to changes in demand (Weidema et al., 2009; Weidema, 2003). As far as the Italian market is concerned, these were identified as: a natural gas power plant for electricity generation (Turconi et al., 2011) and gasoline (with the related supply and production) as transport fuel. The N, P and K nutrients applied on-field with the digestate (residual organic substrate after anaerobic digestion) and compost were assumed to substitute calcium ammonium nitrate, diammonium phosphate and potassium chloride, respectively, on the basis of the NPK content of the digestate (and compost), in agreement with the common practices in 149

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Fig. 1. Overview of the orange peel waste (OPW) management.

combustion in internal combustion engines, biogas upgrading through membranes followed by further compression until the requirements for the gas grid (40 bar), combustion of bio-methane in a conventional compressed natural gas car and digestate dewatering by means of centrifuges. A detailed description of these processes may be found in the SI (Section 2), where the material and energy flows are also provided. As far as pyrolysis is concerned, the feedstock, with a moisture content of less than 15%, was sent to a pyrolysis reactor for the recovery of tar (49.9% w/w), char (28.1% w/w) and syngas (22% w/w), with a heating rate of 12 °C min−1 and a pyrolysis temperature of 740 °C (Negro et al., 2016b). The thus obtained tar was then upgraded to premium fuels (gasoline and diesel) through a catalytic hydrocracking process, with a product yield of 13.2% (Wang et al., 2015). Char and syngas were considered to partially cover the internal consumption, as they were burnt in a combustion chamber with a thermal efficiency of 85%. Additional heat was supplied, by means of natural gas, in order to support the overall process. The recovered premium fuels were considered to displace traditional gasoline and diesel on an energy basis in conventional cars. With respect to incineration, a thermal plant was modelled according to the best available techniques for medium-term production, in terms of efficiency and emissions (Energistyrelsen, 2012). After a pre-dehydration step to reduce the moisture content (to less than 30%), the OPW was combusted to generate electricity, with an efficiency of 25%, calculated on the basis of the wet LHV. Heat was assumed to be partially recovered and used internally for the de-hydration step, and an overall heat recovery efficiency of 37% was obtained. The internal electricity consumption was equal to 86 kW h t−1. No treatment of the wastewater, bottom ashes or fly ashes was included. Anaerobic mono-digestion and co-digestion were modelled as thermophilic digestion (55 °C), with electricity consumption estimated to be equal to 18 kW h t−1 (on a wet weight basis), according to the typical consumptions for homogenous industrial biomass material (Tonini et al., 2013), while the heat demand was estimated to be 512 kW h t−1 (on a dry basis) to heat the reactor from 8 to 55 °C (Alvarado-Morales et al., 2013). The methane yield was assumed to be 75% of the Bio-Methane Potential (BMP) for anaerobic mono-digestion, or equal to the practical BMP adopted in anaerobic co-digestion trials.

consequential LCA studies (e.g. Boldrin et al., 2009; Tonini et al., 2016b). The extracted D-limonene was assumed to displace the production and use of hexane, according to the approach of Shanab et al. (2013), on a double mass basis. Heat production was assumed to be used internally to cover the internal plant consumption (where necessary), as district heating networks are not available in the region under investigation. When OPW was used for feeding, the displacement of conventional energy- and protein-feed was assumed, in line with the energy and protein contents (calculated as total crude proteins) of the OPW. The marginal energy-feed was assumed to be maize, while the marginal protein-feed was assumed to be soymeal, on the basis of previous investigations (Tonini et al., 2016b). Since displacing soymeal also affects soy oil, this being a co-product, it was necessary to consider an induced production of palm oil to compensate for the soy oil displaced production, thus conforming to the well-known “soybean meal loop” dealt with in Dalgaard et al. (2008). It should be noted that, consistently with the system expansion approach used for any other product/energy carriers generated in the OPW scenarios, a conventional feed should be substituted considering how much energy and proteins are actually available in the managed OPW, i.e. on the basis of the functional unit. This follows a common approach in consequential LCA (e.g. Dalgaard et al., 2008; Styles et al., 2015; Tonini et al., 2016a,b). When animal manure and seaweed were used as co-substrates for anaerobic co-digestion, the system boundary of the scenario was expanded to account for the avoided management of these substrates: in the case of manure, this was assumed to be conventional storage and on-land use (without any treatment), while in the case of seaweed, this was assumed to be natural on-shore decay.

2.4. Technology and process inventory The inventory data was built upon previous published experimental studies of the Authors, unpublished experimental data of the Authors, the Ecoinvent database (Ecoinvent Center, 2015), estimations, literature and personal communications with experts in the field. The main technologies involved in the investigated scenarios were pyrolysis and tar upgrading, incineration, anaerobic digestion- and codigestion, landfilling and composting. In addition to these, other minor processes were included in the scenarios: pelleting, dehydration, biogas 150

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Table 1 Material and energy flows referring to the functional unit (1 t OPW). PYR: pyrolysis with tar upgrading. INC: incineration. EXT + AD extraction of D-limonene and anaerobic monodigestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: co-digestion of OPW, manure and seaweed. CD1 + UP: codigestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with bio-filtering. FEED: animal feeding. Rounded values. Scenario

Feed (kg) in

PYR INC EXT + AD

OPW: 1000 OPW: 1000 OPW: 1000

CD1

OPW: 1000 MN: 2088

CD1 + UP

OPW: 1000 MN: 2088

CD2

OPW: 1000 MN: 1630 SW: 434

CD2 + UP

OPW: 1000 MN: 1630 SW: 434

LANDF COMP FEED

OPW: 1000 OPW: 1000 OPW: 1000

Heat (MJ)*

Electricity (kW h) Out Hydrogen: 5.6 N: 2,7 K: 2.7 P: 0.3 Digestate: 136 Limonene: 6.6 N: 7.2 K: 15 P: 2.4 Digestate: 386 N: 7.2 K: 15 P: 2.4 Digestate: 386 N: 8.3 K: 14.4 P: 2 Digestate: 541 N: 8.3 K: 14.4 P: 2 Digestate: 541 Compost: 111 Animal feed: 254

In

Out

26 62 26

– 730 172

2158 1148 661

79

284

857

79

–

79

421

79

8 5 64

In

Fuel (MJ) out

in

Out

– – 39

621 – –

857

120

–

857c

0

120

3558

1197

1197

120

–

1197c

0

120

4914

b

2032 1148 661

a

84 42 1641c

* No district heating networks were assumed in this assessment. All the heat actually recovered was assumed to be used internally either for pre-treatments, in the case of thermochemical processes, or for heating the digestion reactor, in the case of anaerobic digestion processes. Any heating surplus (the difference between output and input) was not otherwise utilized. a Heat provided by the combustion of char and tar used to cover partially internal consumptions. b Heat partially provided by an external source (natural gas). c Heat totally provided by an external source (natural gas).

and soybean production on the basis of the energy and protein contents. Table 1 summarizes the material and energy inputs/outputs that were specifically evaluated for each scenario.

The BMP for OPW and the practical BMP for co-digestion trials (OPW with manure for CD1 and CD1 + UP, and OPW with manure and seaweed for CD2 and CD2 + UP) were experimentally determined (see Table 1 and Section 2.2 of the Supporting Information). The methane leakages from the digesters were assumed to be 2% of the produced CH4, in accordance with Evangelisti et al. (2014). The produced biogas was assumed to be combusted in an internal combustion engine (stationary engine) with an electricity efficiency of 32% of the LHV of the methane input (Evangelisti et al., 2014); heat was assumed to be utilized internally to heat the fermenter, and overall recovery efficiencies of 34%, 26% and 25% on the LHV of the methane input were obtained for the EXT + AD, CD1, and CD2 scenarios, respectively. The digestate obtained from anaerobic digestion was considered to replace the marginal N, P, and K fertilizers (production and application), after a dewatering process with an efficiency of 80% and an electricity demand of 7 kW h t−1 (wet basis) (Tonini et al., 2013). When anaerobic monodigestion was modelled, L-limonene was assumed to be extracted (prior to digestion) through a steam extraction process, with a recovery yield of 94% (Pourbafrani et al., 2013), and used as a hexane substitute. The landfilling technology was modelled considering systems for the collection and treatment of leachate and for gas collection and flaring, in line with the modelling of Manfredi and Christensen (2009), while the composting process was modelled according to the state-of-the-art covered tunnel plant located in Treviso (Italy), which is equipped with bio-filters to clean the exhaust gas (Boldrin et al., 2011). In the feeding scenario, the feedstock was assumed to be dried and pelleted prior to use as a feed, with an estimated demand of 1.6 MJ kg−1 on a wet basis for heating and 1.2 kW h kg−1 on a dry weight basis for electricity. As previously mentioned, the pelleted OPW was assumed to displace maize

2.5. Sensitivity analyses The nature of the uncertainties investigated in this study concerns: 1) parameter uncertainty (when the input data have a low level of confidence) and 2) scenario uncertainty, which is related to the assumptions on which the scenarios are built (e.g. system boundaries and choice of marginal technologies) (Clavreul et al., 2012). As far as the parameter uncertainty is concerned, the tier approach described in Clavreul et al. (2012) was followed to evaluate the uncertainty of the GW results: the first step was to select the parameters that have the most effect on the GW results. This can be achieved by varying the parameters one at a time over their technical feasibility range and then estimating the uncertainty due to all the previously selected inputs. When the effect of the parameters on GW had been evaluated, the sensitivity ratio was calculated and the parameters with the highest sensitivity ratio (above 0.5) were selected to establish the uncertainty propagation. The uncertainty propagation was then performed through a Monte Carlo analysis (number of simulations equal to 1000). For additional details, reference can be made to SI. The scenario uncertainty was evaluated considering two fundamental assumptions regarding the choice of the marginal technologies: (S1) wind energy, as a marginal technology for electricity production instead of natural gas power plants, and (S2) natural gas as a marginal transport fuel instead of petrol. Sensitivity scenario S1 was considered because an increment in wind energy supply has been recorded for the Italian electricity 151

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Fig. 2. LCA results for global warming (GW) (over a 100 year horizon) (expressed as kg CO2-eq t−1), terrestrial acidification (AC) (expressed as kg SO2-eq t−1) and marine N-eutrophication (EP(N)) (expressed as kg N-eq t−1). All the scenarios represent a temporal scope of 20 years. I) monotreatments, II) co-treatments and III) non-recovered energy treatments. PYR: pyrolysis with tar upgrading. INC: incineration. EXT + AD extraction of D-limonene and anaerobic mono-digestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: co-digestion of OPW, manure and seaweed. CD1 + UP: co-digestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with biofiltering. FEED: animal feeding.

highlighted in previous studies (Panepinto et al., 2014).

production market over the 2010–2014 years, together with a simultaneous decrement in natural gas derived electricity (IEA, 2014). Sensitivity scenario S2 was considered because an increment in natural gas consumption has been witnessed on the Italian transport fuel market in recent years (around 32300 TJ for 2010 instead of 48800 in 2014) at the expense of petroleum-derived fuels (IEA, 2014). These statistics highlight that the future marginal technologies on the Italian market may be wind power (instead of natural gas-based electricity) and natural gas (instead of gasoline).

3. Results Figs. 2 and 3 show the results of the selected environmental impact categories (GW, AC, EP(N), HTc, ET and AD fossil) for each scenario. The breakdown of the impact contribution includes the following processes: chemical substitution, avoided management of co-substrates, feed substitution, mineral fertilizer substitution, on-land use, energy substitution, processing, and landfilling/composting. The net value, referring to 1 t OPW (wet weight), is the sum of the negative contributions, that is, those corresponding to the avoided emissions (savings), and positive contributions, that is, those corresponding to the induced emissions (burdens). When the net value is negative, the corresponding scenario allows environmental benefits to be obtained for the chosen impact category (Reference can be made to Section 5 in SI for additional details).

2.6. Preliminary cost flow analysis A preliminary cost flow analysis was performed considering the available market prices for products and technologies. As far as prices are concerned, the situation in Italy in the year 2015 was considered. However, it should be recalled that prices are by nature subjected to important fluctuations, and this analysis can therefore only be considered as a preliminary analysis. Capital, maintenance and operation costs (including chemicals and labour) were included (see Section 4 of the Supporting Information) as were revenues from the sales of electricity and bioproducts. Taxes, subsidies and externalities were not included. The electricity, heat and natural gas for transportation fuel prices were €0.028 kW h−1, €0.12 MJ−1, and €0.98 kg−1, respectively (Assogasmetano, 2016; RAV, 2015). Average prices were assumed for gasoline and diesel, that is, €0.43 and €0.40 L−1, respectively (EUROSTAT, 2016). The difference in the capital goods cost, due to the change in the car fleet, and the cost for bio-methane distribution were neglected. The cost of D-limonene was estimated to be €1 kg−1 (after Pourbafrani et al., 2013). The cost of spreading digestate or compost on land was equal to that avoided, because of the substitution with the corresponding mineral fertilizers, and thus no economic benefits were obtained from the revenues. The sale of pelletized citrus waste for animal feeds (with a moisture content of less than 10%) was calculated as €0.18 kg−1. The economic assessment was performed for all the investigated scenarios: However, it should be recalled that the pyrolysis technology data are affected by a high degree of uncertainty, due to a lack of data on full-scale plants that combine pyrolysis and gasification, as already

3.1. Global warming Among the various mono-treatments that were considered, incineration was found to be the best performing scenario (INC; −203 kg CO2-eq t−1 OPW), and this was followed by anaerobic digestion coupled with D-limonene extraction (EXT + AD; −137 kg CO2-eq t−1 OPW), and then pyrolysis (PYR; −5.5 kg CO2-eq t−1 OPW). When incineration (INC) is compared with anaerobic mono-digestion (EXT +AD), it can be observed that the contribution from energy recovery and the substitution of conventional fossil sources lead to higher GHG savings (around −300 CO2-eq t−1 OPW vs. −200 CO2-eq t−1 OPW, respectively for INC and EXT + AD). Moreover, the processing impacts were lower for incineration (+95 CO2-eq t−1 OPW), mainly because the anaerobic digestion scenarios involve fugitive CH4 emissions, due to leakages in the fermentation reactor and due to the non-combusted CH4 in the stationary engine, which together account for 40% of the total emissions in the “Processing” phase. It is important to note that the anaerobic mono-digestion scenario (EXT + AD) achieved GW savings not only for the production of biofuels (about −200 CO2-eq t−1 OPW), which act as a substitute for 152

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Fig. 3. LCA results for human toxicity in relation to carcinogenic substances (HTc) (expressed as CTU t−1), ecotoxicity to freshwater (ET) (expressed as CTU t−1) and abiotic depletion of fossil resources (AD fossil) (expressed as MJ t−1). All the scenarios represent a temporal scope of 20 years. I) mono-treatments, II) co-treatments and III) non-recovered energy treatments. PYR: pyrolysis with tar upgrading. INC: incineration. EXT + AD extraction of D-limonene and anaerobic mono-digestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: co-digestion of OPW, manure and seaweed. CD1 + UP: co-digestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with bio-filtering. FEED: animal feeding.

disposal techniques (361 kg CO2-eq t−1 OPW), and this was followed by composting (57 kg CO2-eq t−1 OPW). Landfilling and composting are not permitted in Italy because of the produced CH4 and N2O emissions; moreover, there is no energy recovery. The obtained savings were only represented by carbon sequestration, although these were very limited, due to the fact that OPW mainly consists of easily degradable carbon, and by the substitution of mineral fertilizers in the case of compost. The use of OPW for animal feeding showed the best GW performance of all the traditional handling techniques. This was a result of the GHG savings obtained due to the avoidance of the production of conventional crops (maize and soybean) and the associated land-use change impacts (–473 kg CO2eq t−1 OPW). While these GHG benefits are theoretically achievable, assuming the energy and proteins of OPW are used efficiently as animal feeds, in practice there are important challenges, related to costs and feed quality, that hinder the use of this biomass for animal feeding.

natural gas derived-electricity, but also for the on-land application of the residual digestate, as it avoided the need for the production and onland application of mineral fertilizers, and functioned as a carbon storage for the residual non-degraded carbon. In addition, savings in the mono-digestion scenario (EXT + AD) were also achieved for the chemical substitution (D-limonene in place of hexane), and were calculated to be −44 CO2-eq t−1 OPW. The pyrolysis scenario showed comparatively lower GHG savings (PYR; −5.5 kg CO2-eq t−1 OPW), because of the limited benefits from diesel and gasoline recovery (-69 CO2-eq t−1 OPW), due to the low efficiency of tar conversion to biofuels: only about 50% of the input dry matter in the OPW was converted into tar, and only 13% was converted into petrol and diesel. The co-digestion strategies showed larger savings than those of anaerobic mono-digestion, mainly because of the benefits associated with the use of co-substrates. These benefits consisted of an additional energy production and in avoiding the (otherwise) traditional management processes. As far as manure is concerned, this would consist of the storage and on-land spreading of untreated manure (Bacenetti et al., 2016). Instead, for seaweed, the benefits would consist of a natural on shore degradation. The largest savings for the four co-digestion scenarios (CD1, CD1 + UP, CD2 and CD2 +UP) were achieved when biogas was upgraded to bio-methane, which was injected onto the gas grid and used in place of gasoline (i.e., −222 CO2-eq t−1 OPW and −260 CO2-eq t−1 OPW for scenarios CD1 and CD1 + UP, respectively). This was a consequence of the larger GHG savings obtained when biogas was used to substitute gasoline instead of natural gas-based electricity, due to the higher GHG emission factor of gasoline (ca. 90 g CO2 MJ−1) than that of natural gas (ca. 57 g CO2 MJ−1). Landfilling was the worst of the traditional non-energy focused

3.2. Acidification The anaerobic digestion scenarios with a biogas combustion unit (in particular anaerobic mono-digestion, i.e. EXT + AD and co-digestion with animal manure and seaweed, i.e. CD2) along with landfilling and composting showed net impacts on the acidification category. These were mainly due to the NOx and NH3 emissions that resulted from the processing. As far as the anaerobic digestion scenarios are concerned, NOx emissions occur as a result of the combustion of biogas in a stationary engine, while NH3 emissions occur for landfilling and composting, due to the degradation of proteins. The induced N-emissions for the anaerobic EXT + AD and CD2 digestion scenarios, were nor counterbalanced during combustion in the stationary engine (+0.5 kg SO2-eq t−1 OPW and +1.1 kg SO2-eq t−1 OPW, respectively) by the 153

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

lead to an increase in manure production by the animals, as the animals would be fed just the same with the alternative conventional feed. 3.4. Human toxicity and ecotoxicity Animal feeding has been found to be the best scenario in the Human toxicity and Ecotoxicity categories (Fig. 3), mainly due to the remarkable savings induced by the avoided use of pesticides, herbicides and mineral fertilizers in the production of soybean and maize feeds (i.e. the substituted crops in this analysis). In this context, it should be recalled that this study only considers the end-of-pipe management of OPW, while none of the upstream uses of chemicals for the production of oranges are part of the system boundary. This means that little (or no) toxicity impacts are expected from the OPW as such (see the chemical composition provided in Table S1 in SI). It is necessary to note that OPW is a specific by-product of the food industry, and the lack of heavy metals and organic pollutions as well as no risks from HTc or ET were expected. Compared to the animal feeding scenario, the anaerobic mono- and co-digestion scenarios showed no savings (on the ET category) or relatively smaller savings (on HTc), mainly because of the substitution of fossil fuel-based energy production.

Fig. 4. Parameter uncertainty for global warming (GW). The uncertanties were obtained by varying the selected sensitive parameters by 20% in all the scenarios (except for the animal feeding scenario where the iLUC impact was varied by 50%). I) mono-treatments, II) co-treatments and III) non-recovered energy treatments. PYR: pyrolysis with tar upgrading. INC: incineration. EXT + AD extraction of D-limonene and anaerobic digestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: co-digestion of OPW, manure and seaweed. CD1 + UP: co-digestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with bio-filtering. FEED: animal feeding.

3.5. Resource depletion All the scenarios involving energy recovery showed significant savings in this impact category (Fig. 3), because of the displacement of fossil sources. The CD2 co-digestion scenario performed best, because of the energy recovery from both OPW and the manure substrates. Conversely, the conventional management techniques, which did not involve any energy recovery (feeding, composting and landfilling), did not show any environmental savings. In this respect, it should be pointed out that animal feeding, while contributing to the potential savings in all the other environmental categories, was here one of the worst scenarios due to the fact that no-energy recovery was obtained. If fossil energy saving is a priority, then the use of OPW for animal feeding is not a worthwhile alternative, and neither are composting or landfilling.

avoided N-emissions from the displaced natural gas-based electricity (−0.3 kg SO2-eq t−1 OPW and −0.7 kg SO2-eq t−1 OPW, respectively), and there were no N-savings whatsoever in the case of landfilling and composting, as energy recovery was not applied. Higher savings on acidification were observed for the anaerobic digestion scenarios in which upgrading of the biogas to bio-methane for use in the transport sector (i.e. CD1 + UP and CD2 + UP) was considered, because of the better substitution factor when gasoline was displaced, in terms of NOx and SOx. The incineration scenario also contributed with net environmental benefits on AC (−0.3 kg SO2-eq t−1 OPW), thanks to the avoided production of natural gas-based electricity and to the low NOx and SOx emissions at the stack, which are strictly regulated by national and EU legislation, and in particular by Directive 2000/76/EC concerning waste incineration (EC, 2000).

3.6. Sensitivity analysis and uncertainty analysis Sensitivity ratio (SR) values have been quantified for several parameters identified within the scenarios (for more details, see SI, Section 3), with the purpose of selecting the most sensitive ones. These parameters, with an SR close or higher than 1, resulted to be: i) the methane yield at the digester, ii) the electricity recovery efficiency from the combustion of biogas in the gas engine, iii) the electricity recovery efficiency for the incineration technology, iv) changes in land use for the production of conventional crop tar, and v) pyrolysis product (gasoline, diesel and hydrogen) yields in the pyrolysis scenario, and the electricity and heat demand for the tar upgrading process. The propagation of the uncertainties on the scenarios (Fig. 4) showed that the scenario affected by the largest uncertainty is animal feeding (FEED; −266 ± 174 kg CO2eq t−1OPW), because of the significant uncertainty associated with the iLUC impact, here assumed to vary ± 50% around the mean on the basis of previous estimates (Tonini et al., 2012). In the pyrolysis scenario, the SR values were significantly high (around 3.5 for product yield, i.e. tar, gasoline, diesel and biohydrogen, and around −2 for electricity and heat demand). When a Monte Carlo analysis was implemented, the values varied by as much as 62.5% of the mean value, thus demonstrating that this scenario is affected by high uncertainties, and further research should be conducted to obtain a better understanding of the impact of pyrolysis technologies, for which very few LCA studies have been performed compared with, for example, incineration (see the review by Astrup et al., 2015). The

3.3. Marine eutrophication (EP(N)) As far as marine EP(N) is concerned, only the scenarios involving anaerobic digestion and further on-land applications of the residual digestate showed environmental burdens. These were the consequence of nitrate leaching subsequent to the on-land application of the digestate. Such a result is in line with a myriad of previous LCA studies (e.g. among others: Alvarado-Morales et al., 2013; Blengini et al., 2011; Fantin et al., 2015; Hamelin et al., 2014, 2011; Styles et al., 2016, 2016; Tonini et al., 2016b). When digestate was spread on land, only a low fraction was emitted to the air, while the remaining part underwent leaching. Though varying according to the season, precipitation, the plant N up-taking efficiency and the type of soil, the potential N leaching, when liquid digestate was applied on-field, was in fact much higher than its counterpart, i.e. a mineral N-fertilizer, as recently demonstrated by Yoshida et al. (2016). The animal feeding scenario (FEED) highlighted important savings due to the fact that i) it saves on the use of mineral fertilizers for crop production and ii) it does not involve any (additional) on-land handling of residuals, e.g. digestate or compost. In fact, it should be noted, that using OPW as a feed does not

154

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Table 2 Total costs and revenues for each scenario with reference to 1 t of OPW (€ t−1). Positive values represent economic burdens, while negative values represent revenues. Taxes and subsidies have not been included. Values rounded. PYR: pyrolysis with tar upgrading. INC: incineration. EXT + AD extraction of D-limonene and anaerobic mono-digestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: co-digestion of OPW, manure and seaweed. CD1 + UP: co-digestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with bio-filtering. FEED: animal feeding.

Fig. 5. Scenario uncertainty for wind power, as marginal for electricity, and natural gas as marginal for transportation fuel. The baseline values are also reported (stack columns) for comparison purposes. PYR: pyrolysis with tar upgrading. I) mono-treatments, II) cotreatments and III) non-recovered energy treatments. INC: incineration. EXT + AD extraction of D-limonene and anaerobic mono-digestion. CD1: co-digestion of OPW and manure. CD1 + UP: co-digestion of OPW and manure, and biogas upgrading. CD2: codigestion of OPW, manure and seaweed. CD1 + UP: co-digestion of OPW, manure and seaweed, and biogas upgrading. LANDF: conventional landfilling with flaring. COMP: direct composting with bio-filtering. FEED: animal feeding.

Scenario

Total cost (€ t−1)

Revenues (€ t−1)

Net (€ t−1)

PYR INC EXT + DA CD1 CD1 + UP CD2 CD2 + UP LANDF COMP FEED

105 115 102 130 125 170 144 100 40 199

−5 −20 −91 −111 −85 −116 −121 – – −46

100 95 11 18 45 14 50 100 40 153

context have shown that incineration without heat recovery for district heating is not economically viable. The high capital and operative costs (€115 t−1 OPW) were not compensated for by the revenues from the sale of electricity (€20 t−1 OPW). Anaerobic mono-digestion (EXT + AD), coupled with a stationary engine, has a lower total cost (capital and operative cost), but the economic return from revenues (electricity (€–5 t−1) and D-limonene (€–7 t−1)) did not appear so important (Table 2). However, the net value, considering costs and revenues, was significantly lower (€11 t−1) than the incineration plant (€95 t−1). When biogas was upgraded to transportation fuel (e.g. in CD1 + UP), the income was significantly higher (€–85 t−1) than when it was converted to electricity (e.g. €–8 t−1 in CD1). However, because of the necessity of providing heat for the process, the scenarios involving biogas upgrading resulted in higher operative cost, which increased the total cost. The most attractive scenarios appeared to be limonene extraction, followed by anaerobic mono-digestion and codigestion with electricity production (EXT + DA, CD1, and CD2). Landfilling and composting were not economically viable. Among the non-energy focused treatments, the use of OPW as animal feeds in substitution of conventional meals was the worst option from the economic point of view, due to the expenses necessary for de-hydration (€190 t−1) and pelleting (€2 t−1). Feeding therefore appears environmentally sustainable, but not economically feasible, under the assumption that OPW is fed in the form of pellets (to facilitate handling, i.e. storage and transport).

uncertainty on the remaining scenarios was not important enough to change the ranking of the scenario performances on GW, i.e. anaerobic co-digestion scenarios with biogas upgrading to biomethane remained favourable from a GW perspective (−343 ± 34 kg CO2eq t−1OPW). As far as the scenario uncertainty is concerned (Fig. 5), when natural gas was considered as the marginal transport fuel instead of gasoline (S1), a decrease in the total GW savings for the scenarios involving anaerobic (mono- and co-) digestion and biogas upgrading was observed (CD1 + UP and CD2 + UP). However, the benefits were still comparable with (or higher than) those obtained in the anaerobic codigestion scenarios in which the use of the produced biogas was considered for electricity generation (CD1, CD2). Considering wind electricity as the marginal technology for electricity production, instead of natural gas power plants, determines a drastic decrease in the environmental benefits of the scenarios aimed at electricity production from OPW (incineration and co-digestion: INC, CD1, and CD2). Conversely, the performance of the scenarios aimed at transport biofuel production from OPW (pyrolysis and co-digestion with upgrading: PYR, CD1 + UP, and CD1 + UP) was improved as a result of the decreased impact of processing. The same was found for the animal feeding scenario, because of the lower impacts of pelleting. This highlights that, in low-carbon electricity systems, producing transport biofuel, feeds and biomaterials is much more beneficial than producing electricity, from a GW mitigation perspective, as also illustrated in previous studies, e.g. Tonini et al. (2016a). Overall, the results of this study appear robust with respect to certain scenario uncertainties, as the co-digestion scenarios (with biogas upgrading to biomethane for use in the transport sector) were preferable to the others for both the case of petrol and of natural gas as displaced marginal fuels. In the latter case, they were comparable with the co-digestion scenarios in which electricity was produced via biogas combustion in gas engines, again indicating that anaerobic co-digestion is the most favourable treatment.

4. Discussion 4.1. Lessons learned from studies on orange peel waste management Both this and the study by Pourbafrani et al. (2013) have shown that important environmental savings can be achieved by recovering transport biofuels, nutrients and chemicals (e.g. D-limonene) from OPW via biological treatments. While the study by Pourbafrani et al. (2013) only considered two energy-focused scenarios for OPW management (a large biorefinery that produced ethanol, biomethane, digestate, D-limonene and a small biorefinery that produced D-limonene, biomethane, and digestate), this study has enlarged the number of investigated treatment scenarios by also including non-energy focused disposal techniques, such as landfilling, composting and animal feeding. Enlarging the number of investigated scenarios has highlighted that animal feeding may also be an environmentally sound path, when the benefits derived from avoiding the cultivation of crops and the related land-use change impacts are included. This is in line with other recent

3.7. Preliminary cost flow analysis The preliminary cost flow analysis results (Table 2) for the Italian 155

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

of D-limonene is challenging, because of its antimicrobial-activity. For this reason, co-digestion with manure (or algae, or other similar locally available wastes) should be promoted. Manure is a suitable substrate to dilute the toxic compounds and to provide nutrients for microbial growth and buffering capacity (Angelidaki and Ellegaard, 2003). Furthermore, manure is widely and easily available, and the anaerobic digestion of manure alone is not of interest, due to the low carbon and dry matter content. In this respect, manure and citrus waste management may be profitably combined to optimize economic and environmental benefits. Incentives in this direction are therefore recommended. An alternative to co-digestion is the recovery of Dlimonene, prior to digestion. This would not only benefit the energy production process, but also increase the revenues of plant operators, in view of the increasing consumers’ demand for natural products (ViudaMartos et al., 2008). Furthermore, innovative strategies for the storage and spreading of liquid digestate could be introduced in order to reduce the formation of ammonium and nitrous oxides emitted into the air (Ostfoldforskning, 2016). Thermal treatment is also a possible scenario. However, the incineration of such a wet feedstock requires a high amount of energy for the de-hydration step. The high operational costs (per unit of wet OPW) and high capital cost make incineration economically uninteresting in the absence of district heating networks. Pyrolysis and gasification may be of interest for annual rate usages of less than 100,000 t y−1, since incineration at a small-scale is not economically sustainable, as the system has been designed for larger waste streams (ATO, 2015). However, the results of this specific case study show that low upgrading yields and high moisture contents hinder their economic and environmental performance.

studies (Styles et al., 2015, 2016b; Tonini et al., 2016a,b; Tufvesson et al., 2013). However, although this substrate appears to be sustainable, on the basis of the results of this LCA, a number of practical challenges hinders its actual use as a feed. One problem is its bitter flavour, which makes this substrate not very attractive to ruminants (Feedipedia, 2016), thus hampering its marketability. The second is the cost of dehydration for fresh OPW (high energy demand calculated to be around 1.7 MJ kg−1 OPW); this process is in fact necessary since the pellets not only provide ease of handling (transport and storage), but are also preferred by livestock over fresh citrus fruit pulp or silage (Bampidis and Robinson, 2006; Göhl, 1978). In order to reduce energy expenditure, a partial, natural de-hydrating step could be performed on fresh OPW in Mediterranean areas (Tamburini and Zema, 2009). The third problem is related to the overall low nutritional value: dried OPW is characterized by a suitable energetic value for animal feeding, due to the easily digestible carbohydrate and fibre contents. However, the low protein content and the poor digestibility requires protein integration to compensate for this nutritional lack in the diet (Caparra et al., 2007) and, although remarkable environmental benefits could be achieved, the lack of proteins and the bitterness constitute a significant barrier to the actual marketability of OPW as an animal feed. Nonetheless, some hidden potential still exists regarding the use of OPW for feeding; this regards, for example, the fact that D-limonene is a natural terpenic antimicrobial compound: if present in the diet of ruminants, it can reduce Salmonella populations (Callaway et al., 2011). Furthermore, the transfer of terpenes from essential oils, such as D-limonene, to cows is currently under investigation to improve the flavouring properties of milk and extend the shelf-life of meat (Lejonklev et al., 2013). Unlike existing studies on citrus waste treatment (e.g. Martín et al., 2010; Pourbafrani et al., 2013; Volpe et al., 2015), this study has included the potential additional benefits associated with co-substrate utilization in the digestion treatment of OPW, and animal manure in particular. This aspect is of fundamental importance in Europe where manure is the most abundant biomass residue in many regions, clearly illustrated in Panoutsou et al. (2009), and its management still relies to a great extent on storage and on-land applications without any treatment, in this way giving rise to high environmental impacts (De Vries et al., 2012; Hamelin et al., 2014, 2011; Tonini et al., 2016a,b). In line with many recent studies on the subject, the results of this LCA indicate how co-digestion may not only be a technical requirement (to dilute inhibitors and control the process), but also an opportunity to digest the raw locally available manure and boost the overall environmental and economic profits of plant operators, who often look for manure cosubstrates to increase their biogas production per unit of reactor capacity (De Vries et al., 2012; Hamelin et al., 2014). In other words, while the anaerobic mono-digestion of OPW can reduce most of the environmental impacts associated with OPW management, the co-digestion of OPW with animal manure, algae or possibly with other similar wastes, should be encouraged as it favors process control, increases energy production and represents an opportunity for a better management of difficult substrates, especially manure (De Vries et al., 2012). Incineration and, in general, thermal treatments are also possible options. However, environmental and economic benefits are hampered by the high moisture content of the substrate and by the fact that the regions with large citrus waste productions are generally located in agricultural areas in which there is no infrastructure/market for district heating.

4.3. Margins for future research This study uses state-of-the-art knowledge and method to derive robust LCA results for the six impact categories considered. However, in the endeavor to provide authorities and decision-makers with additional information, other important aspects, which were not considered in this assessment, may be of interest for future research: i) addressing local environmental aspects, e.g. quality of soil after application of compost, land and water use, and ii) addressing social issues (i.e. job losses, and acceptance of the waste technology by the community). As far as the environmental perspective is concerned, the improved quality of Mediterranean soils, due to the application of compost on land (Doni et al., 2017), should be investigated in detail, especially in areas subject to soil erosion and degradation, considering the fraction of the spread carbon actually taken up by the crops. Such effects are not captured in this (or traditional) life cycle assessment studies. Additionally, the analysis of land and water use could also be of interest when a specific local context has to be investigated. However, it should be recalled that the displacement effects (e.g. crops displaced by the use of orange peel waste as a feed) could ultimately involve imported feeds, i.e. may be well beyond the local contexts investigated. Other local effects, which may also be of interest for a comprehensive assessment, are acoustic pollution and emission of odours, which have not been considered in this analysis due to lack of reliable data. 5. Conclusion Co-digestion with manure (or eventually seaweed) for biogas and nutrient production has appeared to be the best management option for orange peel waste, in terms of global warming and resource depletion mitigation effects. Among the alternatives considered for the final utilization of the obtained biogas, upgrading to biomethane for use in the transport sector appeared to be the best alternative. A drawback of all the anaerobic digestion strategies was the eutrophication impact due to nitrogen leaching as a result of the on-land application of the residual organic digestate. A preliminary cost analysis also highlighted

4.2. Policy recommendations In order to optimize resource recovery from OPW and move away from the end of pipe treatment, conventional handling techniques, such as landfilling and composting, should be avoided. Anaerobic digestion, targeting the production of high-value products such as biofuels, chemicals and nutrients, appears to be the most favourable technology to maximize environmental and economic benefits. However, the presence 156

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

produce bio-energy—a life cycle perspective. Bioresour. Technol. 125, 239–248. Doni, S., Macci, C., Longo, V., Souid, A., Garcia, C., Masciandaro, G., 2017. Innovative system for biochemical monitoring of degraded soils restoration. CATENA 152, 173–181. http://dx.doi.org/10.1016/j.catena.2017.01.016. EC, 1999. Landfill waste – Environment – European Commission [WWW Document]. URL http://ec.europa.eu/environment/waste/landfill_index.htm (Accessed 18 January 2017). EC, 2000. The Waste Incineration Directive – Environment – European Commission [WWW Document]. URL http://ec.europa.eu/environment/archives/air/stationary/ wid/legislation.htm (Accessed 18 January 2017). EUROSTAT, 2016. Consumer prices of petroleum products, end of second half 2015 (EUR per litre) YB16. png – Statistics Explained [WWW Document]. URL http://ec.europa. eu/eurostat/statistics-explained/index.php/File:Consumer_prices_of_petroleum_ products,_end_of_second_half_2015_(EUR_per_litre)_YB16.png (Accessed 18 January 2017). Ecoinvent Center, 2015. ecoinvent Version 3 [WWW Document]. URL http://www. ecoinvent.org/database/database.html (Accessed 18 January 2017). Ekvall, T., Azapagic, A., Finnveden, G., Rydberg, T., Weidema, B.P., Zamagni, A., 2016. Attributional and consequential LCA in the ILCD handbook. Int. J. Life Cycle Assess. 21, 293–296. http://dx.doi.org/10.1007/s11367-015-1026-0. El-Shimi, S.A., El-Housseini, M., Ali, B.E., El-Shinnawi, M.M., 1992. Biogas generation from food-processing wastes. Resour. Conserv. Recycl. 6, 315–327. http://dx.doi. org/10.1016/0921-3449(92)90055-7. Energistyrelsen, 2012. Technology Data For Energy Plants. Generation of Electricity and District Heating, Energy Storage and Energy Carrier Generation and Conversion. Copenhagen. Evangelisti, S., Lettieri, P., Borello, D., Clift, R., 2014. Life cycle assessment of energy from waste via anaerobic digestion: a UK case study. Waste Manag. 34, 226–237. http://dx.doi.org/10.1016/j.wasman.2013.09.013. Fantin, V., Giuliano, A., Manfredi, M., Ottaviano, G., Stefanova, M., Masoni, P., 2015. Environmental assessment of electricity generation from an Italian anaerobic digestion plant. Biomass Bioenergy 83, 422–435. http://dx.doi.org/10.1016/j.biombioe. 2015.10.015. Feedipedia, 2016. Feedipedia: An on-line encyclopedia of animal feeds|Feedipedia [WWW Document]. URL http://www.feedipedia.org/ (Accessed 18 January 2017). Ferrari, A., Morone, P., Tartiu, V., 2016. Tackling uncertainty through business plan analysis—a case study on citrus waste valorisation in the south of Italy. Agriculture 6, 5. http://dx.doi.org/10.3390/agriculture6010005. Finnveden, G., Hauschild, M.Z., Ekvall, T., Guinée, J., Heijungs, R., Hellweg, S., Koehler, A., Pennington, D., Suh, S., 2009. Recent developments in life cycle assessment. J. Environ. Manage. 91, 1–21. http://dx.doi.org/10.1016/j.jenvman.2009.06.018. Forgács, G., Pourbafrani, M., Niklasson, C., Taherzadeh, M.J., Hováth, I.S., 2012. Methane production from citrus wastes: process development and cost estimation. J. Chem. Technol. Biotechnol. 87, 250–255. http://dx.doi.org/10.1002/jctb.2707. Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D.W., Haywood, J., Lean, J., Lowe, D.C., Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M., Van, R., 2007. Changes in atmospheric constituents and in radiative forcing changes in atmospheric constituents and in radiative forcing. Chapter 2. Climate Change 2007. The Physical Science Basis. Forster-Carneiro, T., Berni, M.D., Dorileo, I.L., Rostagno, M.A., 2013. Biorefinery study of availability of agriculture residues and wastes for integrated biorefineries in Brazil. Resour. Conserv. Recycl. 77, 78–88. http://dx.doi.org/10.1016/j.resconrec.2013.05. 007. Göhl, B., 1978. Citrus by-products for animal feed [WWW Document]. URL http://www. fao.org/docrep/004/x6512e/X6512E08.htm (Accessed 10 January 2017). Guinée, J.B., Gorrée, M., Heijungs, R., Huppes, G., Kleijn, R., Koning, A., de Oers, L., van Wegener Sleeswijk, A., Suh, S., Udo de Haes, H.A., Bruijn, H., de Duin, R., van Huijbregts, M.A.J., 2002. Handbook on Life Cycle Assessment: Operational Guide to the ISO Standards. Kluwer Academic Publishers. Hamelin, L., Wesnæs, M., Wenzel, H., Petersen, B.M., 2011. Environmental consequences of future biogas technologies based on separated slurry. Environ. Sci. Technol. 45, 5869–5877. http://dx.doi.org/10.1021/es200273j. Hamelin, L., Naroznova, I., Wenzel, H., 2014. Environmental consequences of different carbon alternatives for increased manure-based biogas. Appl. Energy 114, 774–782. http://dx.doi.org/10.1016/j.apenergy.2013.09.033. Hauschild, M.Z., Goedkoop, M., Guinée, J., Heijungs, R., Huijbregts, M., Jolliet, O., Margni, M., De Schryver, A., Humbert, S., Laurent, A., Sala, S., Pant, R., 2013. Identifying best existing practice for characterization modeling in life cycle impact assessment. Int. J. Life Cycle Assess. 18, 683–697. http://dx.doi.org/10.1007/ s11367-012-0489-5. IEA, 2014. IEA – Report [WWW Document]. URL http://www.iea.org/statistics/ statisticssearch/report/?country=ITALY&product=NaturalGas&year=2010 (Accessed 5 June 2017). ISO 14040, 2006. ISO 14040:2006 – Environmental management – Life cycle assessment – Principles and framework [WWW Document]. URL http://www.iso.org/iso/ catalogue_detail?csnumber=37456 (Accessed 21 June 2017). ISO 14044, 2006. ISO 14044:2006 – Environmental management – Life cycle assessment – Requirements and guidelines [WWW Document]. URL http://www.iso.org/iso/ catalogue_detail?csnumber=38498 (Accessed 21 June 2017). Lejonklev, J., Løkke, M.M., Larsen, M.K., Mortensen, G., Petersen, M.A., Weisbjerg, M.R., 2013. Transfer of terpenes from essential oils into cow milk. J. Dairy Sci. 96, 4235–4241. http://dx.doi.org/10.3168/jds. Manfredi, S., Christensen, T.H., 2009. Environmental assessment of solid waste landfilling technologies by means of LCA-modeling. Waste Manag. 29, 32–43. http://dx.doi.org/ 10.1016/j.wasman.2008.02.021. Martín, M.A., Siles, J.A., Chica, A.F., Martín, A., 2010. Biomethanization of orange peel

anaerobic mono- and co-digestion scenarios as being economically preferable to the alternatives. Among these, animal feeding was also an environmentally appealing option, in particular because of its use as a substitute for conventional meals, but practical market challenges emerged due to the processing costs. The sensitivity analysis generally confirmed the robustness of the results, but also highlighted that further efforts are needed to collect better inventory data on pyrolysis technologies. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.resconrec.2017.08.014. References Alvarado-Morales, M., Boldrin, A., Karakashev, D.B., Holdt, S.L., Angelidaki, I., Astrup, T., 2013. Life cycle assessment of biofuel production from brown seaweed in Nordic conditions. Bioresour. Technol. 129, 92–99. http://dx.doi.org/10.1016/j.biortech. 2012.11.029. Angelidaki, I., Ellegaard, L., 2003. Codigestion of manure and organic wastes in centralized biogas plants: status and future trends. Appl. Biochem. Biotechnol. 109, 95–106. http://dx.doi.org/10.1385/ABAB:109:1-3:95. Assogasmetano, 2016. ASSOGASMETANO [WWW Document]. URL http://www. assogasmetano.it (Accessed 1 October 2016). Astrup, T.F., Tonini, D., Turconi, R., Boldrin, A., 2015. Life cycle assessment of thermal waste-to-energy technologies: review and recommendations. Waste Manag. 37, 104–115. http://dx.doi.org/10.1016/j.wasman.2014.06.011. ATO, 2015. Verifica di fattibilità per la formazione di una struttura societaria integrata del sistema di gestione dei rifiuti nell’ambito territoriale torinese. Relazione conclusiva del gruppo di lavoro [WWW Document]. URL http://www.atorifiutitorinese. it/cms/ente/atti-e-documenti/documentitc (Accessed 20 January 2017). Bacenetti, J., Bava, L., Zucali, M., Lovarelli, D., Sandrucci, A., Tamburini, A., Fiala, M., 2016. Anaerobic digestion and milking frequency as mitigation strategies of the environmental burden in the milk production system. Sci. Total Environ. 539, 450–459. http://dx.doi.org/10.1016/j.scitotenv.2015.09.015. Bampidis, V.A., Robinson, P.H., 2006. Citrus by-products as ruminant feeds: a review. Anim. Feed Sci. Technol. 128, 175–217. http://dx.doi.org/10.1016/j.anifeedsci. 2005.12.002. Blengini, G.A., Brizio, E., Cibrario, M., Genon, G., 2011. LCA of bioenergy chains in Piedmont (Italy): a case study to support public decision makers towards sustainability. Resour. Conserv. Recycl. 57, 36–47. http://dx.doi.org/10.1016/j.resconrec. 2011.10.003. Boldrin, A., Andersen, J.K., Møller, J., Christensen, T.H., Favoino, E., 2009. Composting and compost utilization: accounting of greenhouse gases and global warming contributions. Waste Manag. Res. 27, 800–812. http://dx.doi.org/10.1177/ 0734242x09345275. Boldrin, A., Neidel, T.L., Damgaard, A., Bhander, G.S., Møller, J., Christensen, T.H., 2011. Modelling of environmental impacts from biological treatment of organic municipal waste in EASEWASTE. Waste Manag. 31, 619–630. http://dx.doi.org/10.1016/j. wasman.2010.10.025. Broeren, M.L.M., Zijp, M.C., Waaijers-van der Loop, S.L., Heugens, E.H.W., Posthuma, L., Worrell, E., Shen, L., 2017. Environmental assessment of bio-based chemicals in early-stage development: a review of methods and indicators. Biofuels Bioprod. Biorefining. http://dx.doi.org/10.1002/bbb.1772. Callaway, T.R., Carroll, J.A., Arthington, J.D., Edrington, T.S., Anderson, R.C., Rossman, M.L., Carr, M.A., Genovese, K.J., Ricke, S.C., Crandall, P., Nisbet, D.J., 2011. Orange peel products can reduce Salmonella populations in Ruminants. FOODBORNE Pathog. Dis. 8, 1071–1075. http://dx.doi.org/10.1089/fpd.2011.0867. Caparra, P., Foti, F., Scerra, M., Sinatra, M.C., Scerra, V., 2007. Solar-dried citrus pulp as an alternative energy source in lamb diets: effects on growth and carcass and meat quality. Small Rumin. Res. 68, 303–311. http://dx.doi.org/10.1016/j.smallrumres. 2005.11.015. Clavreul, J., Guyonnet, D., Christensen, T.H., 2012. Quantifying uncertainty in LCAmodelling of waste management systems. Waste Manag. 32, 2482–2495. http://dx. doi.org/10.1016/j.wasman.2012.07.008. Clavreul, J., Baumeister, H., Christensen, T.H., Damgaard, A., 2014. An environmental assessment system for environmental technologies. Environ. Model. Softw. 60, 18–30. http://dx.doi.org/10.1016/j.envsoft.2014.06.007. Crawshaw, R., 2003. Co-product feeds: animal feeds from the food and drinks industries. J. Sci. Food Agric. 83, 362. http://dx.doi.org/10.1002/jsfa.1326. Dalgaard, R., Schmidt, J., Halberg, N., Christensen, P., Thrane, M., Pengue, W.A., 2008. LCA of soybean meal. Int. J. Life Cycle Assess. 13, 240–254. http://dx.doi.org/10. 1065/lca2007.06.342. de Besi, M., McCormick, K., 2015. Towards a bioeconomy in europe: national, regional and industrial strategies. Sustainability 7, 10461–10478. http://dx.doi.org/10.3390/ su70810461. de Man, R., Friege, H., 2016. Circular economy: european policy on shaky ground. Waste Manag. Res. 34, 93–95. http://dx.doi.org/10.1177/0734242x15626015. De Vries, J.W., Vinken, T.M.W.J., Hamelin, L., De Boer, I.J.M., 2012. Comparing environmental consequences of anaerobic mono-and co-digestion of pig manure to

157

Resources, Conservation & Recycling 127 (2017) 148–158

V. Negro et al.

Tonini, D., Hamelin, L., Wenzel, H., Astrup, T., 2012. Bioenergy production from perennial energy crops: a consequential LCA of 12 bioenergy scenarios including land use changes. Environ. Sci. Technol. 46, 13521–13530. http://dx.doi.org/10.1021/ es3024435. Tonini, D., Martinez-Sanchez, V., Astrup, T.F., 2013. Material resources, energy, and nutrient recovery from waste: are waste refineries the solution for the future? Env. Sci. Technol. 6 (47). http://dx.doi.org/10.1021/es400998y. Tonini, D., Hamelin, L., Alvarado-Morales, M., Astrup, T.F., 2016a. GHG emission factors for bioelectricity, biomethane, and bioethanol quantified for 24 biomass substrates with consequential life-cycle assessment. Bioresour. Technol. 208, 123–133. http:// dx.doi.org/10.1016/j.biortech.2016.02.052. Tonini, D., Hamelin, L., Astrup, T.F., 2016b. Environmental implications of the use of agro-industrial residues for biorefineries: application of a deterministic model for indirect land-use changes. Appl. Energy 8, 690–706. http://dx.doi.org/10.1111/ gcbb.12290. Tufvesson, L.M., Lantz, M., Börjesson, P., 2013. Environmental performance of biogas produced from industrial residues including competition with animal feed -life-cycle calculations according to different methodologies and standards. J. Clean. Prod. 53, 214–223. http://dx.doi.org/10.1016/j.jclepro.2013.04.005. Turconi, R., Butera, S., Boldrin, A., Grosso, M., Rigamonti, L., Astrup, T., 2011. Life cycle assessment of waste incineration in Denmark and Italy using two LCA models. Waste Manag. Res. 29, 78–90. http://dx.doi.org/10.1177/0734242x11417489. USDA, 2013. Gain Report [WWW Document]. URL http://gain.fas.usda.gov/ RecentGAINPublications/CitrusSemi-annual_Madrid_EU-27_6-12-2013.pdf (Accessed 27 October 2016). Viuda-Martos, M., Ruiz-Navajas, Y., Fernández-López, J., Pérez-Álvarez, J., 2008. Antifungal activity of lemon (Citrus lemon L.), mandarin (Citrus reticulata L.), grapefruit (Citrus paradisi L.) and orange (Citrus sinensis L.) essential oils. Food Control 19, 1130–1138. http://dx.doi.org/10.1016/j.foodcont.2007.12.003. Volpe, M., Panno, D., Volpe, R., Messineo, A., 2015. Upgrade of citrus waste as a biofuel via slow pyrolysis. J. Anal. Appl. Pyrol. 115, 66–76. http://dx.doi.org/10.1016/j. jaap.2015.06.015. Wang, H., Wang, L., Shahbazi, A., 2015. Life cycle assessment of fast pyrolysis of municipal solid waste in North Carolina of USA. J. Clean. Prod. 87, 511–519. http://dx. doi.org/10.1016/j.jclepro.2014.09.011. Weidema, B.P., Ekvall, T., Heijungs, R., 2009. Guidelines for application of deepened and broadened LCA Guidelines for application of deepened and broadened LCA Deliverable D18 of work package 5 of the CALCAS project [WWW Document]. URL http://www.leidenuniv.nl/cml/ssp/publications/calcas_report_d18.pdf (Accessed 27 October 2016). Weidema, B., 2003. Market information in life cycle assessment. Environmental Project No. 863. Danish Environmental Protection Agency [WWW Document]. URL http:// citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.197.5739&rep=rep1&type= pdf (Accessed 27 October 2016). Wilkins, M.R., Suryawati, L., Maness, N.O., Chrz, D., 2007. Ethanol production by Saccharomyces cerevisiae and Kluyveromyces marxianus in the presence of orangepeel oil. World J. Microbiol. Biotechnol. 23, 1161–1168. http://dx.doi.org/10.1007/ s11274-007-9346-2. Yoo, J.H., Lee, H.B., Choi, S.W., Kim, Y.B., Sumathy, B., Kim, E.K., 2011. Production of an antimicrobial compound by Bacillus subtilis LS 1–2 using a citrus-processing byproduct. Korean J. Chem. Eng. 28, 1400–1405. http://dx.doi.org/10.1007/s11814010-0516-9. Yoshida, H., Nielsen, M.P., Scheutz, C., Jensen, L.S., Bruun, S., Christensen, T.H., 2016. Long-term emission factors for land application of treated organic municipal waste. Environ. Model. Assess. Environ. Model. Assess 21, 111–124. http://dx.doi.org/10. 1007/s10666-015-9471-5.

waste. Bioresour. Technol. 101, 8993–8999. http://dx.doi.org/10.1016/j.biortech. 2010.06.133. Miranda, R., Bustos-Martinez, D., Blanco, C.S., Villarreal, M.H.G., Cantú, M.E.R., 2009. Pyrolysis of sweet orange (Citrus sinensis) dry peel. J. Anal. Appl. Pyrol. 86, 245–251. http://dx.doi.org/10.1016/j.jaap.2009.06.001. Negro, V., Mancini, G., Ruggeri, B., Fino, D., 2016a. Citrus waste as feedstock for biobased products recovery: review on limonene case study and energy valorization. Bioresour. Technol. 214, 806–815. http://dx.doi.org/10.1016/j.biortech.2016.05. 006. Negro, V., Ruggeri, B., Mancini, G., Fino, D., 2016b. Recovery of D-limonene through moderate temperature extraction and pyrolytic products from orange peels. J. Chem. Technol. Biotechnol. http://dx.doi.org/10.1002/jctb.5107. Ostfoldforskning, 2016. Biogassproduksjon fra m atavfall og møkk fra ku, gris og fjørfe. Panepinto, D., Senor, A., Genon, G., 2014. Environmental and Economic Analysis of the Turin Incineration Plant. WIT Press, pp. 479–490. http://dx.doi.org/10.2495/ WM140421. Panoutsou, C., Eleftheriadis, J., Nikolaou, A., 2009. Biomass supply in EU27 from 2010 to 2030. Energy Policy 37, 5675–5686. http://dx.doi.org/10.1016/j.enpol.2009.08. 032. Pourbafrani, M., McKechnie, J., Maclean, H.L., Saville, B.A., 2013. Life cycle greenhouse gas impacts of ethanol, biomethane and limonene production from citrus waste. Environ. Res. Lett. 8 (1). http://dx.doi.org/10.1088/1748-9326/8/1/015007. RAV, 2015. Relazione annuale sullo stato dei servizi e sull’attività svolta [WWW Document]. URL http://www.autorita.energia.it/allegati/relaz_ann/15/RAVolumeI_ 2015.pdf (Accessed 27 September 2016). ReCiPe, 2008. A Life Cycle Impact Assessment Method Which Comprises Harmonised Category Indicators at the Midpoint and the Endpoint Level. First Edition Report I: Characterisation [WWW Document]. Minist. Housing, Spat. Plan. Environ. Netherlands. URL http://www.lcia-recipe.net/ (Accessed 18 January 2017). Rosenbaum, R.K., Bachmann, T.M., Gold, L.S., Huijbregts, M.A.J., Jolliet, O., Juraske, R., Koehler, A., Larsen, H.F., MacLeod, M., Margni, M., McKone, T.E., Payet, J., Schuhmacher, M., van de Meent, D., Hauschild, M.Z., 2008. USEtox—the UNEPSETAC toxicity model: recommended characterisation factors for human toxicity and freshwater ecotoxicity in life cycle impact assessment. Int. J. Life Cycle Assess. 13, 532–546. http://dx.doi.org/10.1007/s11367-008-0038-4. Ruiz, B., Flotats, X., 2016. Effect of limonene on batch anaerobic digestion of citrus peel waste. Biochem. Eng. J. 109, 9–18. http://dx.doi.org/10.1016/j.bej.2015.12.011. Seppälä, J., Posch, M., Johansson, M., Hettelingh, J.P., 2006. Country-dependent characterisation factors for acidification and terrestrial eutrophication based on accumulated exceedance as an impact category indicator. Int. J. Life Cycle Assess. 11, 403–416. http://dx.doi.org/10.1065/lca2005.06.215. Shanab, K., Neudorfer, C., Spreitzer, H., 2013. Green solvents in organic synthesis: an overview. Curr. Org. Chem. 17 (11), 1179–1187. http://dx.doi.org/10.2174/ 1385272820666160209212804. Siles, J.A., Vargas, F., Gutiérrez, M.C., Chica, A.F., Martín, M.A., 2016. Integral valorisation of waste orange peel using combustion, biomethanisation and co-composting technologies. Bioresour. Technol. 211, 173–182. http://dx.doi.org/10.1016/j. biortech.2016.03.056. Styles, D., Gibbons, J., Williams, A.P., Stichnothe, H., Chadwick, D.R., Healey, J.R., 2015. Cattle feed or bioenergy? Consequential life cycle assessment of biogas feedstock options on dairy farms. GCB Bioenergy 7, 1034–1049. http://dx.doi.org/10.1111/ gcbb.12189. Styles, D., Dominguez, E.M., Chadwick, D., 2016. Environmental balance of the of the UK biogas sector: an evaluation by consequential life cycle assessment. Sci. Total Environ. 560–561, 241–253. http://dx.doi.org/10.1016/j.scitotenv.2016.03.236. Tamburini, V., Zema, D.A., 2009. Citrus – Trattato di agrumicoltura. In: Vacante, V. (Ed.), Il Sole 24 Ore-Edagricole, pp. 459–470.

158