Environmental implications of biohydrogen based energy production from steam reforming of alcoholic waste

Environmental implications of biohydrogen based energy production from steam reforming of alcoholic waste Antonio Cortés, Gumersindo Feijoo, Antonio Chica, Javier Francisco Da Costa-Serra, María Teresa Moreira Accepted Mansucript How to cite: Cortés, A., Feijoo, G., Chica, A., Da Costa-Serra, J., & Moreira, M. (2019). Environmental implications of biohydrogen based energy production from steam reforming of alcoholic waste. Industrial Crops And Products, 138, 111465. doi: 10.1016/j.indcrop.2019.111465 Copyright information: © 2019 Elsevier Ltd. This manuscript version is made available under the CC-BY-NC-ND 4.0 license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Environmental implications of biohydrogen based energy production from steam reforming 1 of alcoholic waste 2 Antonio Cortésa, Gumersindo Feijooa, Antonio Chicab, Javier Francisco Da Costa-Serrab and 3 María Teresa Moreiraa* 4 a Department of Chemical Engineering, School of Engineering, Universidade de Santiago de 5 Compostela, Rúa Lope Gómez de Marzoa, s/n, 15782 Santiago de Compostela (Spain) 6 b Institute of Chemical Technology, Universitat Politècnica de València-Consejo Superior de 7 Investigaciones Científicas, Avd. de los Naranjos s/n, 46022 València (Spain) 8 * Corresponding author 9 E-mail: [email protected] 10 11 Abstract 12 Nowadays, there is an increasing demand for energy in the world. With an energy system still 13 based on fossil fuels, a paradigm shifts towards clean energy production based on available 14 renewable resources is necessary. Hydrogen is a high-quality energy carrier that can be used 15 with great efficiency and is expected to acquire a great importance in the next generation of 16 fuels. This study aims to analyze the potential environmental impacts associated with the 17 steam reforming of alcoholic waste from distilleries to produce clean electricity by using the 18 Life Cycle Assessment methodology. The main findings from this study reported that the global 19 environmental profile is better than other alternatives more common as sanitary landfill or 20 incineration. In terms of some impact categories as Abiotic and Ozone Depletion, Acidification 21 and Eutrophication, steam reforming of alcoholic waste performed better profiles than other 22 processes that produce hydrogen from diverse feedstocks. 23 Keywords: Alcoholic waste; Environmental profile; LCA; SOFC; Biohydrogen 24 2 Abbreviations 25 LCA Life Cycle Assessment SS Subsystem CC Climate change OD Ozone depletion TA Terrestrial acidification FE Freshwater eutrophication ME Marine eutrophication HT Human toxicity POF Photochemical oxidant formation PMF Particulate matter formation TET Terrestrial ecotoxicity FET Freshwater ecotoxicity MET Marine ecotoxicity FD Fossil depletion WW Wastewater WGS Water Gas Shift PSA Pressure Swing Adsorption ISO International Organization for Standardization 26 1. Introduction 27 Currently, global energy production is based on the use of fossil fuels such as coal, oil and 28 natural gas (Rossetti et al., 2015a) and accounts for approximately 65% of global GHG 29 emissions (Uusitalo et al., 2017). Dependence on the use of fossil fuels as an energy resource 30 has caused environmental problems of global impact, such as air pollution in terms of emission 31 of pollutants and particles, as well as the depletion of natural resources, among others (Hajjaji 32 et al., 2016; Reyes-Valle et al., 2015), which leads to adverse consequences for society in terms 33 of human health and damage to the ecosystem (Valente et al., 2019). So much so that the 34 2030 Agenda and the 17 Sustainable Development Goals (SDGs) set by the United Nations 35 3 includes ensuring access to affordable, reliable and sustainable energy for all. This objective 36 aims at guaranteeing universal access to energy service, substantially increasing the share of 37 renewable energy in the global energy mix and doubling the rate of improvement in energy 38 efficiency. This is why the paradigm shift towards clean energy production must be based on 39 available renewable resources (Da Costa-Serra and Chica, 2018). 40 In recent years, numerous alternatives to the use of traditional fossil fuels have been 41 proposed, such as the production of biofuels, bioalcohols, hydrogen or any type of renewable 42 energy (Balat, 2011). In particular, biomass is one of the renewable energy sources that has 43 experienced strong growth in recent years, due to its global availability and diversity (Spiridon 44 et al., 2016; Tian et al., 2018). Biofuels derived from biomass offer a number of advantages 45 over their oil-based counterparts according to Demirbas (2008): they can be considered carbon 46 neutral after-combustion by fixing carbon during biomass growth, close to a carbon-neutral 47 balance, so that they contribute to achieving sustainability goals. For this reason, numerous 48 initiatives have been developed in the development of conversion technologies based on 49 resources derived from biomass (Unrean et al., 2018). 50 Focusing on the different types of fuels, hydrogen is a high quality energy carrier that can be 51 used with high efficiency (Frolov et al., 2013) and is expected to acquire great importance in 52 next generation fuels (Alipour-Moghadam et al., 2014). This fact, together with declining fossil 53 fuel reserves, steadily rising prices and increasing pollution make hydrogen a very attractive 54 product for meeting global energy demand (Khaodee et al., 2011). 55 However, the environmental profile of hydrogen-based energy systems is as "clean" or "dirty" 56 depending on the scheme of conversion (Rabenstein and Hacker, 2008). The traditional 57 schemes producing H2 from natural gas are a major source of CO2, with emissions of 58 approximately 10-12 kg of CO2 per kg of H2 (Spath and Mann, 2001). Traditional plants produce 59 hydrogen by catalytic steam reforming of natural gas, which is a mature technology and is the 60 4 pathway by which most hydrogen is produced today. Because of this, reducing CO2 emissions 61 associated with hydrogen production would result in a considerable reduction of pollution 62 (Salkuyeh et al., 2018). 63 In this sense, fuel cells technology and the use of hydrogen are proposed as one of the most 64 promising environmental solutions in relation to the reduction of global emissions (Díaz 65 Alvarado and Gracia, 2010). Fuel cells are devices that electrochemically convert chemical 66 energy from fuels into electricity (Morales et al., 2010). Among the different types of fuel cells, 67 the Solid Oxide Fuel Cell (SOFC) is the most efficient, due to its high operating temperatures 68 and the fact that it is not poisoned with CO (Hernández and Kafarov, 2009). When this type of 69 battery is used, an efficiency around 50% can be obtained (Strazza et al., 2015); in addition, an 70 efficiency of 70% can be achieved if cogeneration system is used (Strazza et al., 2010). 71 Hydrogen production from renewable sources such as poplar (Susmozas et al., 2016) or willow 72 wood (González-García et al., 2012), sugar cane (Halleux et al., 2008), sweet potato (Costa et 73 al., 2018), sorghum (Aguilar-Sánchez et al., 2018) or sugar beet (Luo et al., 2009) have been 74 investigated as the first actions to achieve a significant reduction of environmental impacts 75 (Salkuyeh et al., 2018). Hydrogen can be obtained from different feedstocks through steam 76 reforming (Braga et al., 2016; López et al., 2019; Zheng et al., 2019), autothermal reforming 77 (Khila et al., 2017; Spallina et al., 2018; Xue et al., 2017) and aqueous phase reforming 78 (Coronado et al., 2018; Esteve-Adell et al., 2017; García et al., 2018), among them, steam 79 reforming is the most common, as almost 90% of H2 is produced by natural gas reforming. It 80 also has the highest conversion efficiency, around 70% (Haryanto et al., 2005). 81 Steam reforming of natural gas is the most popular method for producing commercial 82 hydrogen that currently covers about 50% of global hydrogen demand (Anzelmo et al., 2018) 83 and is sometimes referred to as steam methane reforming (SMR). Steam reforming is an 84 endothermic process based on the reaction of gas with steam at high temperature and 85 5 moderate pressure. In this way, the chemical reaction taking place leads to hydrogen and 86 carbon dioxide (Reaction 1): 87 CH3CH2OH + H2O  2CO2 + 6H2 ΔHr= 174 kJ mol-1 (1) 88 However, depending on the reaction mixture and operating conditions in the reactor, another 89 route can be followed, producing undesirable products (Ni et al., 2007), such as carbon 90 monoxide (Reaction 2), methane (Reaction 3) or ethylene (Reaction 4): 91 CH3CH2OH + H2O  2CO + 4H2 ΔHr= 256 kJ mol-1 (2) 92 CH3CH2OH  CO + CH4 + H2 ΔHr= 50 kJ mol-1 (3) 93 CH3CH2OH  C2H4 + H2O ΔHr= 46 kJ mol-1 (4) 94 Once the process is complete, the output stream must undergo purification treatment to avoid 95 the presence of by-products such as methane and carbon monoxide. The removal of CO is an 96 important step because it normally poisons the catalyst in fuel cells, that is why CO is removed 97 first by the Water Gas Shift (WGS) reaction (Reaction 5). WGS is an exothermic and reversible 98 reaction usually used in industry to produce high purity hydrogen (Alamolhoda et al., 2019). 99 Normally, 90% of the CO outflowing from the steam reforming reactor can be converted to 100 CO2 (Rossetti et al., 2015b). 101 CO + H2O  H2 + CO2 ΔHr= -41 kJ mol-1 (5) 102 Following this stage, the Pressure Swing Adsorption (PSA) process separates hydrogen from 103 the rest of the components of the gas stream with 85% efficiency, obtaining H2 with 99% purity 104 (Susmozas et al., 2013), and whose energy content is usually higher than that of the natural 105 gas used for reforming. 106 The implementation of other alternatives of hydrogen production can be considered from 107 alternative raw materials, such as alcohols (Rossetti et al., 2015a). In addition to steam 108 reforming of ethanol, studies have been published on steam reforming of different types of 109 6 alcohol with the aim of producing hydrogen. Some of these alcohols are butanol (Kumar et al., 110 2018), propanol (Wang et al., 2015), methanol (Tian et al., 2017) or glycerol (Menezes et al., 111 2018) but, even so, the use of ethanol for this purpose offers the best opportunity to produce 112 hydrogen from renewable sources (Ramírez and Homs, 2008), especially if this ethanol is 113 derived as residue from other processes. Specifically, the alcoholic wastes from the wine 114 industry results an attractive raw material due to 65% of world wine production is managed by 115 European winegrowers mostly small and medium-sized wineries according to the Comité 116 Européen des Enterprises Vins (CEEV, 2016). Wine production generates large amounts of solid 117 and liquid wastes, with a serious impact on the environment when they are not adequately 118 treated. The liquid wastes are processed in distilleries to obtain purified alcohols, but in these 119 processes, alcoholic purges without commercial value containing impurities separated from 120 the good quality alcohols are generated. Thus, the process here analysed aims to raise 121 awareness of the potential of these by-products and their valorization activities as a 122 sustainable way to produce hydrogen. 123 The main objective of the study is to analyze the potential environmental impacts associated 124 with the steam reforming of alcoholic waste from distilleries. Quantifying the consumption of 125 material and energy resources during the life cycle makes it possible to estimate potential 126 changes and emissions to the environment. The main product of the process is hydrogen (H2), 127 along with a certain amount of carbon monoxide (CO), carbon dioxide (CO2), methane (CH4) 128 and ethylene (C2H4), which accounts for a proportion lower than 30%. This output stream is 129 used to produce energy in a 3 kW SOFC. 130 2. Materials and methods 131 2.1. Definition of goal and scope 132 The Life Cycle Analysis methodology has been considered as a fundamental tool in the analysis 133 of the environmental profile associated with the steam reforming of alcoholic waste from 134 7 distilleries in order to identify key environmental performance indicators. In distilleries, alcohol 135 can be extracted from some wines that cannot been marketed. During this distillation process, 136 an ethanol-rich fraction is obtained, but also a residual fraction that remains in the distiller's 137 tail, which is the residue used in this study. 138 Figure 1 presents the block diagram of the process, identifying the system boundaries, the 139 different subsystems considered and the main inputs and outputs of the system. 140 No infrastructure process was considered in the evaluation, since the environmental impacts 141 per process unit, from installation, construction, decommissioning, infrastructure, machinery, 142 etc., have been considered negligible during the lifetime of this type of facilities. This has been 143 a common practice in other life cycle assessment studies of biorefineries (Jeswani et al., 2015; 144 Karlsson et al., 2014). However, this study has taken into account the manufacture of the 145 catalyst and the SOFC phase, due to the fact that their useful life is clearly shorter than that of 146 large installations. 147 148 Figure 1. System boundaries of the reforming system for the valorization of the alcoholic 149 waste. Caption: T: Transport; R1: Reforming reactor; T1: Heat exchanger. 150 8 The foreground system includes the process units that are the direct object of this study. For 151 the purposes of the study, three subsystems (SS) have been considered, which are detailed 152 below: 153 Subsystem 1: Catalyst formulation. This subsystem considers all the materials necessary for the 154 manufacture of the catalyst used in the reforming reactor (Menor et al., 2017). The catalyst is 155 composed of a sepiolite base with Nickel (15% weight) and Lanthanum (1% weight). Its 156 considered useful life is 20 months, regenerating every 4 months. The transport of the catalyst 157 to the plant is also considered, taking as distance 100 km. 158 Subsystem 2: Steam reforming. This process includes all the inputs needed to perform the 159 steam reforming process. These inputs are mainly electricity, water and alcohol residues from 160 distilleries. The transport of alcoholic waste to the plant is not included, as this type of facility 161 is designed to be included in the distillery. The waste produced in this subsystem is the catalyst 162 spent at the end of its useful life and is considered 100 km as the average transport to the 163 landfill. 164 Subsystem 3: SOFC. This subsystem includes the net production of electricity in the SOFC using 165 the SS2 gas stream as feed. At the exit of this subsystem, CO2 and H2O emissions are derived 166 from the reactions taking place with CO and CH4, C2H4 and H2 inside the SOFC. The electricity 167 produced is fed into the grid. The SOFC works at a high temperature of around 600ºC and 168 produces a large amount of heat, as represented is Figures 1 and 2, this heat is redirected to 169 the system and used to heat the stream entering the reforming reactor. This subsystem 170 includes the SOFC manufacturing stage. Gas stream purifying processes are not included 171 because SOFC are not poisoned by the presence of CO (Hernández and Kafarov, 2009). This 172 type of device directly provides electricity from the chemical reaction taking place. The 173 electrodes of this type of battery are catalytic, so they are relatively stable and are not 174 consumed (Fragiacomo et al., 2018). 175 15 269 Figure 3. Relative contribution (%) of the different subsystems to the total environmental 270 impact 271 In order to highlight the processes with the highest environmental impact on the life cycle 272 performance of the system, the individual contributions to the impact are broken down in 273 Figure 4. These results show that SOFC manufacturing is the major contribution in almost all 274 impact categories, except for ME. Therefore, the manufacture of SOFC is the main hotspot of 275 the system and must have the highest priority in the improvement actions from the 276 environmental point of view. 277 The second largest contributor to the total environmental impact is electricity consumption, 278 with contributions percentage ranging from 15% in HT to 32.9% in POF. If electricity 279 production is taken into account, its contribution to environmental impact decreases 280 significantly to 4.2% and 9.1% in HT and POF, respectively. The formulation of catalysts 281 presented a uniform distribution of environmental impacts in all categories, with contributions 282 always below 10.6%. The consumption of natural gas to heat the process is only responsible 283 for a maximum of 9.4% in CC and 8.5% in FD, but in the rest of the impact categories, their 284 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% CC OD TA FE ME HT POF PMF TET FET MET FD Relative contribution SS3 SS2 SS1 16 contributions never exceed 6%. The rest of the substances (transport, water consumption and 285 waste treatment) contributed practically insignificantly to the environmental impact. 286 Wastewater treatment is the main contributor to the ME impact, due to the high amount of 287 nitrogen-based compounds such as nitrites and nitrates that are discharged in the treated 288 effluent. This may explain, as seen in Figure 3, why the main contributor to the ME impact 289 category is SS1, due to the wastewater generated during the formulation of the catalyst. 290 291 Figure 4. Relative contribution (%) of the components of alcoholic waste steam reforming to 292 the overall impact 293 The environmental profile of alcoholic waste steam reforming is mainly due to two factors. On 294 the one hand, the use of a raw material that is considered waste from another process and 295 therefore has no associated environmental impact. On the other hand, the production of 296 electricity makes it possible to obtain environmental credits that promote a better 297 environmental profile. 298 299 -20% 0% 20% 40% 60% 80% 100% CC OD TA FE ME HT POF PMF TET FET MET FD Relative contribution Catalyst formulation Transport Water consumption Electricity consumption SOFC manufacture Wastewater Waste treatment Natural gas Electricity production 17 3.2. Sensitivity analysis 300 In order to compare the environmental characterization results of some alternative waste 301 treatments to steam reforming, a sensitivity analysis was performed. The methods selected for 302 this analysis were landfill and incineration. Note that the inventory data for incineration and 303 landfill were taken from the Ecoinvent® database. Figure 5 depicts the environmental 304 performance of the alternative treatments for the alcoholic waste considered. As noted, the 305 steam reforming scenario potentially implied a more acceptable environmental profile than 306 the other scenarios, except for Ozone Layer Depletion, Terrestrial Acidification, Freshwater 307 Eutrophication and Terrestrial Ecotoxicity. In particular, steam reforming makes it possible to 308 reduce GHG emissions by 33% compared to incineration and by 30% compared to landfill. 309 Steam reforming is the largest contributor to OD impact for the emission of harmful gases to 310 the stratospheric ozone layer during some operations such as electricity generation or 311 chemical production. Regarding Terrestrial Acidification and Freshwater Eutrophication, steam 312 reforming presents the worst environmental performance. The consumption of Ni-based 313 compounds in the manufacture of SOFC and some processes derived from the extraction of 314 lanthanum for SOFC are the responsible processes of the poor performance in TA and FE 315 respectively. With respect to TET, steam reforming has worse results, but if the three 316 ecotoxicity categories (TTE, MET and FET) are considered, the environmental impact of steam 317 reforming is lower, improving 95.8% with respect to incineration and 97.8% with respect to the 318 sanitary landfill. 319 18 320 Figure 5. Comparative environmental profile of the alternative treatments for alcoholic waste 321 considering 1 tonne as functional unit 322 3.3. Comparative analysis 323 In addition to the basic scheme, a comparison was made with some processes published in the 324 scientific literature. The FU was changed to 1 kg of hydrogen produced in the plant with 99.9 325 vol% purity by steam reforming (Figure 6), in agreement with other reforming studies using 326 other raw materials for hydrogen production (Hajjaji et al., 2016, 2013; Khila et al., 2016; 327 Susmozas et al., 2016, 2015, 2013), thus allowing the comparison of the environmental profile 328 of different processes. Therefore, the new facility configuration does not consider the 329 operation of the SOFC, consequently the output stream of the system is led to a purification 330 system: First, the WGS process removes carbon monoxide and produces a small amount of 331 additional hydrogen. Additionally, in a COPROX reactor the remaining CO can be further 332 reduced to CO2 in the presence of oxygen. Finally, the PSA process separates H2 from the rest 333 of the gases in the stream, obtaining H2 with 99% purity. Therefore, two additional subsystems 334 were introduced to purify the output hydrogen stream (SS3) and provide cooling water (SS4) 335 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% CC OD TA FE ME HT TET FET MET Steam reforming Sanitary landfill Incineration 19 336 Figure 6. New flowchart considered to compare the steam reforming of alcoholic waste with 337 other published studies. 338 The following processes have been considered: SMR-H2: Steam reforming of methane obtained 339 from natural gas (Susmozas et al., 2013). PG-H2: Poplar biomass gasification. The system 340 includes the cultivation of poplar and its transport to the plant. Once in the plant are included 341 all the operations necessary to obtain hydrogen and the production of electricity from the 342 steam produced in the system (Susmozas et al., 2013). PG&C-H2: Gasification of poplar 343 biomass, as mentioned above, but includes carbon fixation during the cultivation stage 344 (Susmozas et al., 2016). GSR-H2: Glycerol reforming, obtained as a co-product of biodiesel 345 production by transesterification of rapeseed oil. Carbon sequestration during oil production is 346 considered (Susmozas et al., 2015). BSR-H2: Biofuel reforming, including also carbon fixation 347 produced during the growth of biomass used for biofuel (Susmozas et al., 2015). SBR-H2: Steam 348 reforming of bioethanol, including bioethanol production (Hajjaji et al., 2016). BAR-H2: 349 Autothermal reforming of bioethanol (Khila et al., 2016). 350 In order to carry out the comparative analysis, life cycle inventories were modified. In relation 351 to the type and quantity of reaction catalyst, the WGS data were obtained from (Compagnoni 352 et al., 2017). All the necessary data to measure the inputs and outputs in SS3. WGS and PSA 353 20 and SS4. Cooling water supply were obtained from (Susmozas et al., 2015, 2013). Inventory 354 data of WGS, PSA and Cooling water supply can be found in Table 5. 355 Table 5. New life cycle inventory to compare the steam reforming of alcoholic wastes with 356 other published processes 357 Inputs from Technosphere Outputs to Technosphere Materials kg Products kg Alcoholic waste 5.42 H 2 1.00 Water 24.21 Emissions kg SR catalyst 4.41·10-4 CO 2 8.16 WGS catalyst 1.11·10-3 CH 4 0.45 Energy kWh C 2 H 4 9.41·10-3 Electricity 1.36 CO 1.97 Transport t·km Wastes kg Road 0.48 SR catalyst to landfill 1.11·10-3 WGS catalyst to landfill 4.41·10-4 Wastewater from WGS 4.61 358 The results of the comparison between steam reforming of alcoholic residues and other 359 related processes are presented in terms of the impact categories of the CML methodology 360 Global warming potential (GWP – kg CO2 eq), Depletion of abiotic resources (ADP – kg Sb eq), 361 Ozone layer depletion (ODP – kg CFC-11 eq), Photochemical oxidation (POFP – kg C2H4 eq), 362 Acidification potential (AP – kg SO2 eq) and Eutrophication potential (EP – kg PO43-). The 363 magnitudes of the environmental impacts of hydrogen production systems are displayed in 364 Table 6, in order to simplify the comparative study, the results are scaled to 100 and 365 represented in Figure 7. For example, alcoholic waste steam reforming shows the best results 366 in terms of ADP, ODP, AP and EP, but performs worse in GWP and has the worst result in POFC. 367 368 369 370 21 Table 5. Summarized results of comparative life cycle assessment 371 Processes GWP ADP ODP POFP AP EP Reference Present study 9.55 4.25·10-3 5.29·10-8 3.17·10-3 4.20·10-3 8.11·10-4 Present study SMR-H2 10.60 8.90·10-2 1.20·10-6 5.18·10-4 8.40·10-3 1.64·10-3 (Susmozas et al., 2013) PG-H2 0.41 8.57·10-2 1.62·10-7 4.40·10-4 1.19·10-2 2.85·10-3 (Susmozas et al., 2013) PG&C-H2 -14.60 --- 2.85·10-7 8.31·10-4 2.07·10-2 4.60·10-3 (Susmozas et al., 2016) GSR-H2 12.70 5.69·10-2 8.90·10-7 5.16·10-4 6.51·10-2 5.26·10-2 (Susmozas et al., 2015) BSR-H2 3.79 4.13·10-2 5.54·10-7 6.00·10-4 1.56·10-2 3.20·10-3 (Susmozas et al., 2015) SBR-H2 6.81 2.13·10-2 3.96·10-7 1.55·10-3 3.53·10-2 2.54·10-2 (Hajjaji et al., 2016) BAR-H2 7.27 4.87·10-2 3.13·10-6 1.65·10-3 2.76·10-2 2.81·10-2 (Khila et al., 2016) 372 373 Figure 7. Comparison (in %) of different reforming processes to obtain hydrogen. Caption: Dark 374 Blue: Present studio; Light Blue: SMR-H2; Pink: PG-H2; Brown: PG&C-H2; Green: GSR-H2; Purple: 375 BSR-H2; Red: SBR-H2; Yellow: BAR-H2. 376 The comparison between the present study and other published processes is possible because 377 the environmental performance of the different studies is published in some LCA studies with 378 a methodological framework consistent with this study. Steam reforming of alcoholic residues 379 has the highest value in POF, due to direct emissions of CH4, which occur in relatively high 380 quantities during steam reforming. However, this process performs well in terms of ODP with a 381 value around 2% of BAR-H2, which is the process with the worst environmental performance in 382 this category. 383 -20% 0% 20% 40% 60% 80% 100% GWP ADP ODP POFP AP EP Relative contribution -100% 22 Four impact categories are detailed in this section: Depletion of abiotic resources (ADP), 384 Acidification potential (AP), Eutrophication potential (EP) and Global warming potential (GWP). 385 These are the most common and well-established categories for assessing bioenergy systems 386 in LCA studies (Cherubini and Strømman, 2011; Muench and Guenther, 2013; Peters et al., 387 2015). The total GHG emissions of the system are estimated at approximately 9.55 kg CO2 eq 388 per kg of H2 produced. As can be seen in Figure 8.a, this value is relatively higher than that of 389 other technologies but is considerably lower than that of a conventional H2 production system 390 (SMR-H2). Approximately 90% of these emissions are attributed to direct methane emissions 391 from the reforming reactor, as CH4 is 21 times more likely to affect GWP over a 100-year 392 period, according to IPCC. The lowest value in this category corresponds to PG&G-H2, since this 393 process considers CO2 capture during biomass cultivation. This explains the importance of 394 system boundaries in an LCA study, since PG&G-H2 covers from biomass cultivation to 395 hydrogen production with CO2 capture. However, in the present study the limit of the system 396 ranges from alcoholic residues entering the plant to the production of electricity, so carbon 397 sequestration during biomass cultivation is not considered. 398 Some metals, minerals and fossil fuels are used in all H2 production systems. Figure 8b shows 399 that fossil methane to hydrogen system (SMR-H2) has the greatest impact on ADP, as expected, 400 due to the large consumption of fossil fuels in the reforming process. H2 produced from 401 bioethanol consumes considerable non-renewable resources throughout the life cycle when 402 ethanol production phases are considered (Hajjaji et al., 2013). However, in this study, steam 403 reforming of alcoholic waste (mainly ethanol) is the best process in terms of ADP because this 404 ethanol is a waste derived from another process that has no associated impact. With respect 405 to Acidification Potential and Eutrophication Potential (Figure 8c and 8d), steam reforming of 406 alcoholic waste presents the best results. The processes with the highest impact in these 407 impact categories are those that take into account the cultivation phase (SBR-H2, BAR-H2 and 408 23 GSR-H2), mainly due to the use of fertilizers containing nitrate, ammonia and phosphate in the 409 production of bioethanol from wheat or biodiesel from rapeseed oil. 410 411 412 Figure 8. Comparison of the environmental impacts in GWP, ADP, AP and EP categories 413 4. Conclusions 414 From a life cycle perspective, the results suggest that this type of energy systems that produce 415 hydrogen from alcoholic waste through steam reforming has good environmental 416 performance. Overall, steam reforming of this type of alcoholic waste for energy production 417 could play a significant role in future energy systems. 418 The SOFC is the main contributor to environmental impact in most impact categories. 419 Analysing the different processes, the manufacture of SOFC is the process with the greatest 420 environmental impact in all impact categories except in ME, where wastewater treatment is 421 the main contributor. The sensitivity analysis shows the promising performance of this waste 422 treatment, since the treatment of 1 tonne of alcoholic waste produces 351 kg of CO2 eq, this 423 result is 33% and 30% better than incineration and sanitary landfill respectively. 424 It is clear that Life Cycle Assessment is a useful tool to determine the environmental 425 performance of steam reforming of alcoholic waste to produce electricity. However, these 426 -2 0 2 4 6 8 10 12 14 kg CO2eq/FU Global Warming Potential 0,00 0,02 0,04 0,06 0,08 0,10 kg Sb eq/FU Abiotic Depletion Potential 0,0E+00 1,0E-02 2,0E-02 3,0E-02 4,0E-02 5,0E-02 6,0E-02 7,0E-02 kg SO2eq/FU Acidification Potential 0,0E+00 1,0E-02 2,0E-02 3,0E-02 4,0E-02 5,0E-02 kg PO43- eq/FU Eutrophication Potential 24 results have been obtained by evaluating the data taken on a laboratory scale, and more 427 studies on a larger scale will be needed in the future to determine a more accurate estimate of 428 the actual environmental profile of the process. 429 Comparative analysis has allowed us to compare this process with others related to the 430 production of hydrogen from different raw materials. Although steam reforming has some 431 poor results in GWP and POF due to methane emissions, its environmental performance is 432 generally better than other processes published in the scientific literature. 433 434 Acknowledgements 435 This research was supported by the European Projects STAR-ProBio (Grant Agreement Number 436 727740) and life-ECOELECTRICITY. The authors (Mr. Antonio Cortés, Prof. Gumersindo Feijoo 437 and Prof. Maria Teresa Moreira) belong to the Galician Competitive Research Group GRC 438 ED413C 2017/2019 and to the CRETUS Strategic Partnership (ED431E 2018/01), co-funded by 439 FEDER (EU). Prof. María Teresa Moreira and Prof. Antonio Chica acknowledge to Red de 440 Excelencia en biorrefinerías sostenibles (CTQ2016-81848-REDT) 441 References 442 Aguilar-Sánchez, P., Navarro-Pineda, F.S., Sacramento-Rivero, J.C., Barahona-Pérez, L.F., 2018. 443 Life-cycle assessment of bioethanol production from sweet sorghum stalks cultivated in 444 the state of Yucatan, Mexico. Clean Technol. Environ. Policy 20, 1685–1696. 445 https://doi.org/10.1007/s10098-017-1480-4 446 Alamolhoda, S., Vitale, G., Hassan, A., Nassar, N.N., Almao, P.P., 2019. Synergetic effects of 447 cerium and nickel in Ce-Ni-MFI catalysts on low-temperature water-gas shift reaction. 448 Fuel 237, 361–372. https://doi.org/10.1016/j.fuel.2018.09.096 449 Alipour-Moghadam, R., Yusup, S., Azlina, W., Nehzati, S., Tavasoli, A., 2014. Investigation on 450 31 Hydrogen Energy 43, 9514–9528. https://doi.org/10.1016/j.ijhydene.2018.04.024 597 Spallina, V., Matturro, G., Ruocco, C., Meloni, E., Palma, V., Fernandez, E., Melendez, J., 598 Pacheco-Tanaka, A.D., Viviente-Sole, J.L., van Sint-Annaland, M., Gallucci, F., 2018. Direct 599 route from ethanol to pure hydrogen through autothermal reforming in a membrane 600 reactor: Experimental demonstration, reactor modelling and design. Energy 143, 666–601 681. https://doi.org/10.1016/j.energy.2017.11.031 602 Spath, P.L., Mann, M.K., 2001. Life cycle assessment of hydrogen production via natural gas 603 steam reforming, National Renewable Energy Laboratory. 604 Spiridon, I., Darie-Nita, R.N., Hitruc, G.E., Ludwiczak, J., Cianga, I.A., Niculaua, M., 2016. New 605 opportunities to valorize biomass wastes into green materials. J. Clean. Prod. 133, 235–606 242. https://doi.org/10.1016/j.jclepro.2016.05.143 607 Strazza, C., Del Borghi, A., Costamagna, P., Gallo, M., Brignole, E., Girdinio, P., 2015. Life Cycle 608 Assessment and Life Cycle Costing of a SOFC system for distributed power generation. 609 Energy Convers. Manag. 100, 64–77. https://doi.org/10.1016/j.enconman.2015.04.068 610 Strazza, C., Del Borghi, A., Costamagna, P., Traverso, A., Santin, M., 2010. Comparative LCA of 611 methanol-fuelled SOFCs as auxiliary power systems on-board ships. Appl. Energy 87, 612 1670–1678. https://doi.org/10.1016/j.apenergy.2009.10.012 613 Susmozas, A., Iribarren, D., Dufour, J., 2015. Assessing the Life-Cycle Performance of Hydrogen 614 Production via Biofuel Reforming in Europe. Resources 4, 398–411. 615 https://doi.org/10.3390/resources4020398 616 Susmozas, A., Iribarren, D., Dufour, J., 2013. Life-cycle performance of indirect biomass 617 gasification as a green alternative to steam methane reforming for hydrogen production. 618 Int. J. Hydrogen Energy 38, 9961–9972. https://doi.org/10.1016/j.ijhydene.2013.06.012 619 Susmozas, A., Iribarren, D., Zapp, P., Linβen, J., Dufour, J., 2016. Life-cycle performance of 620 32 hydrogen production via indirect biomass gasification with CO2 capture. Int. J. Hydrogen 621 Energy 41, 19484–19491. https://doi.org/10.1016/j.ijhydene.2016.02.053 622 Tian, J., Ke, Y., Kong, G., Tan, M., Wang, Y., Lin, J., Zhou, W., Wan, S., 2017. A novel structured 623 PdZnAl/Cu fiber catalyst for methanol steam reforming in microreactor. Renew. Energy 624 113, 30–42. https://doi.org/10.1016/j.renene.2017.04.070 625 Tian, Y., Zhou, X., Lin, S., Ji, X., Bai, J., Xu, M., 2018. Syngas production from air-steam 626 gasification of biomass with natural catalysts. Sci. Total Environ. 645, 518–523. 627 https://doi.org/10.1016/j.scitotenv.2018.07.071 628 Unrean, P., Lai Fui, B.C., Rianawati, E., Acda, M., 2018. Comparative techno-economic 629 assessment and environmental impacts of rice husk-to-fuel conversion technologies. 630 Energy 151, 581–593. https://doi.org/10.1016/j.energy.2018.03.112 631 Uusitalo, V., Väisänen, S., Inkeri, E., Soukka, R., 2017. Potential for greenhouse gas emission 632 reductions using surplus electricity in hydrogen, methane and methanol production via 633 electrolysis. Energy Convers. Manag. 134, 125–134. 634 https://doi.org/10.1016/j.enconman.2016.12.031 635 Valente, A., Iribarren, D., Gálvez-Martos, J.L., Dufour, J., 2019. Robust eco-efficiency 636 assessment of hydrogen from biomass gasification as an alternative to conventional 637 hydrogen: A life-cycle study with and without external costs. Sci. Total Environ. 650, 638 1465–1475. https://doi.org/10.1016/j.scitotenv.2018.09.089 639 Wang, M., Au, C.T., Lai, S.Y., 2015. H2 production from catalytic steam reforming of n-propanol 640 over ruthenium and ruthenium-nickel bimetallic catalysts supported on ceria-alumina 641 oxides with different ceria loadings. Int. J. Hydrogen Energy 40, 13926–13935. 642 https://doi.org/10.1016/j.ijhydene.2015.07.162 643 Xue, Z., Shen, Y., Zhu, S., Li, P., Zeng, Y., Xi, Z., Cai, Y., 2017. Autothermal reforming of ethyl 644 acetate for hydrogen production over Ni3La7Oy/Al2O3catalyst. Energy Convers. Manag. 645 33 146, 34–42. https://doi.org/10.1016/j.enconman.2017.05.018 646 Zheng, T., Zhou, W., Yu, W., Ke, Y., Liu, Y., Liu, R., San Hui, K., 2019. Methanol steam reforming 647 performance optimisation of cylindrical microreactor for hydrogen production utilising 648 error backpropagation and genetic algorithm. Chem. Eng. J. 357, 641–654. 649 https://doi.org/10.1016/j.cej.2018.09.129 650 651