13th International Colloquium Fuels - September 2021 167 Beyond GHG: Cradle-to-grave impact of renewable energyfuelled mobility on water consumption and eutrophication Victor Gordillo Zavaleta* Aramco Fuel Research Center, Aramco Overseas BV, 232 Avenue Napoléon Bonaparte, 92500 Rueil-Malmaison, France Alexander Stoffregen and Oliver Schuller Sphera, Hauptstraße 111-113, 70771 Leinfelden-Echterdingen, Germany * Corresponding author: victor.gordillo@aramcooverseas.com +31 88 262 24 99 Summary The study focuses on the environmental impacts of a selected number of vehicles and energy propulsions at horizon 2030, in particular water consumption and freshwater eutrophication. Greenhouse gas emissions are also analysed. Powertrains include conventional internal combustion engine vehicles, battery electric vehicles and fuel cell electric vehicles. For each powertrain type, renewable energy sources like biofuels and power-to-liquid fuels are compared to conventional sources like fossil fuels and the grid electricity mix. The differences found in the study show that both the manufacturing and use phase play an important role in the life cycle impacts. Electrified solutions show higher impacts during their manufacturing stage, which are generally compensated by better engine efficiencies during their use phases, but under certain conditions conventional thermal powertrains can achieve similar results, especially for greenhouse gas emissions and water consumption. The source of energy chosen for vehicle propulsion and PtL production is critical. The study also shows the substantial impact of materials on water consumption and eutrophication, in particular for battery electric vehicles. The study concludes that for light duty vehicles, multiple parameters and criteria should be used to provide an environmental impact assessment. 1. Introduction In recent years, renewable energy-fuelled transport solutions have increasingly gained attention because of their generally accepted potential to mitigate global warming. Liquid renewable fuels in particular are considered as a viable solution, as they can be used in the legacy fleet of road, air and marine transportation. In the particular case of drop-in liquid advanced fuels, they would also claim the advantage of minimizing the need of significant modifications in the fuel supply chain or engine retro-fitting, making them an solution well-suited during the period of energy transition. While the discussions about the main challenges for the adoption of these mobility solutions generally include the intermittency of renewable power, direct and indirect land use, the recognition of carbon sources as sustainable, and the guarantee of origins of the renewable energy used in their production, the question of water use by these alternative pathways is vital. Impacts on water resources are present in critical steps such as electrolysis, battery and equipment manufacturing, agriculture and waste treatment processes. The present study addresses two main impacts affecting water bodies: The impact of water consumption and the eutrophication of freshwater. Both impact factors are assessed for a selected number of mobility solutions, including battery electric vehicles and internal combustion engine vehicles running on advanced renewable drop-in fuels, such as second generation biofuels and power-to-liquid (PtL) fuels. Europe in 2030 has been chosen as time and location reference for the manufacturing processes. The analysis applies the principles of Life Cycle Assessment (LCA) on a modelling platform built on software application GaBi (licenced by Sphera). This same platform was used to carry out greenhouse gas (GHG) emission studies on vehicles in the past years [1] [2]. The present study is the first published work focusing on water and eutrophication impact categories based on this same model. 2. Scope 2.1 Methodology For the environmental assessment, a life-cycle approach following ISO 14040: 2006 [3] and ISO14044: 2006 [4] Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 168 13th International Colloquium Fuels - September 2021 was chosen. Data for the foreground system (carbon capture, hydrogen and syngas production, fuel synthesis, passenger vehicles and lithium-ion battery manufacturing) were taken from various literature sources. Background data, such as life-cycle inventory (LCI) data for energy (electricity, conventional fuels) and material supply were taken from Sphera’s GaBi LCI databases 2021 [5]. The LCA models were created using the GaBi software system for life-cycle engineering [5]. 2.2 System boundaries and cut-off All impact categories considered are analysed on a cradle-to-grave (CTG) basis, meaning that they include direct emissions from power, fuel and vehicle manufacturing, but also plant infrastructure building and end-of-life. The methodological approach is Attributional LCA (ALCA), so all emissions are calculated as averages based on one specific life cycle inventory. Most energy and material flows as well as emissions for the foreground system are included in the model. No specific cut-off criteria was defined for the assessment. 2.3 Allocation In cases where several products were produced from the PtL synthesis (e.g. Methanol-to-Gasoline, Fischer- Tropsch pathways) and for conventional fuels, allocation to split-up impacts between the different products was done based on the energy content (lower heating value, LHV) of the products. 2.4 Impact categories considered The main focus of the study are two impact categories from the EF 3.0 [6] methodology chosen for the life-cycle impact assessment (LCIA): • EF 3.0 Water use (ILCD Level of recommendation III): A midpoint impact category measuring the water consumption impact as the quantity of water deprived (water scarcity), according to the model Available WAter REmaining (AWARE) [7]. This impact is recommended by the Product Environmental Footprint (PEF) initiative by the European Commission, and measured in units of cubic meters of world waterequivalent (m 3 world eq). • EF 3.0 Freshwater eutrophication (ILCD Level of recommendation II): A midpoint impact category measuring the excess nutrient availability in freshwater bodies, leading to phenomena such as harmful algal blooms (HABs) [8]. This impact factor is calculated according to the EUTREND model, used in LCA method ReCiPe 2008. Contrary to terrestrial or marine eutrophication, both expressed in units of nitrogen equivalent, in freshwater environments phosphorus is considered the limiting factor in pure form or in the form of phosphates and phosphoric acid. For this reason, freshwater eutrophication is measured in kg of phosphorus-equivalent (kgPO 4 -eq) • Alternatively and to provide the context of previous studies, this paper also provides figures for the impact category Climate Change, represented by its corresponding characterisation factor in EF 3.0. This impact is calculated using a baseline model of 100 years of the IPCC 2013 model [9]. 3. Life Cycle Inventories 3.1 Fossil fuels Fossil gasoline and diesel fuels are obtained via crude oil distillation and upgrading in oil refineries. Typically, the calculation of environmental impacts of a conventional fuel takes into account specific refinery configurations related to its production. For the present study and considering the variety of such configurations in Europe, impact category values come from an average European refinery production scheme and crude slate, as available in the GaBi LCI database [5]. For the calculation, direct refinery emissions were allocated based on the mass of the co-products when required. Emissions related to crude oil consumption where allocated by energy content (lower heating value) [10]. 3.2 Hydrogen Hydrogen (H 2 ) can be directly used as fuel-in-fuel cell electric vehicles or as feedstock, together with carbon dioxide, for PtL fuels. Although hydrogen is today predominantly produced via production routes using fossil energy, such as steam methane reforming (SMR) [11], PtL fuels as such are defined by the usage of electricity as energy input and therefore use water electrolysis to produce the hydrogen. Within the electrolysis, water is converted through an electrochemical process into hydrogen and oxygen: = +286 kJ/ mol Several electrolysis technologies exist. Alkaline electrolysis (AE) and proton exchange membrane (PEM) electrolysis working on low temperature (70-90°C) and solid oxide electrolysis cells (SOEC) working on high temperature (650-850°C) requiring an additional heat source beside the electricity supply. Alkaline electrolysis, as the oldest and proved water electrolysis technology [12], were currently installed in industry scale with capacities up to 10 MW and proposed projects in the 100 MW [13] scale are underway. Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 13th International Colloquium Fuels - September 2021 169 Table 1: Technical parameters of alk aline electrolysis Reference year 2030 Efficiency (Low heating value) [11] 68.0% Power consumption [kWh/ kg H 2 ] 49.0 Water consumption [kg/ kg] 15 Pressure (outlet) [bar] 30 This study assumes two types of hydrogen: Conventional hydrogen from SMR (also called grey hydrogen), and hydrogen produced from alkaline electrolysis of water and wind electricity (green hydrogen). Table 1 summarises the technical specifications of alkaline electrolysers used to produce green hydrogen. 3.3 Carbon capture PtL fuels require a carbon source in the form of carbon dioxide (CO 2 ) that can be either extracted from ambient air or from flue gases or gas mixtures from which CO 2 needs to be separated. Especially for the separation of CO 2 from an industrial gas (e.g., synthesis gas from a steam methane reformer), a multitude of proven technologies for industrial scale exist [14]. These technologies include chemical absorption with amine, physical absorption with water or an organic solvent, pressure swing adsorption (PSA), membrane or cryogenic methods. Carbon sources for specific PtL plants are ideally located next to the plant to avoid CO 2 transportation. Beyond 2030, point sources of CO 2 are expected to become scarce, and PtL plants will be probably installed in areas close to renewable electricity generation farms, usually in remote areas. In this context, direct air capture (DAC) could be a promising solution. An important difference between DAC and carbon capture from point sources is the concentration of CO 2 in the used gas (0.04% for ambient air, 3-15% for power plants and up to 100% for, e.g., ammonia plants) [15] and the related energy consumption for CO 2 separation and purification. Table 2 shows the electricity and heat demands for a low temperature DAC per kg of captured CO 2 used in this study. For the sake of comparison, in a coal power plant with 13% CO 2 concentration in its flue gas, the energy consumption by chemical absorption using amines would drop to 0.024 kWh electricity and 3.0 MJ of thermal energy per kg of captured CO2. [14] Table 2: Technical parameters of low-temperature Direct Air Capture (DAC) unit Power consumption [kWh/ kg CO 2 ] of which power for CO 2 desorption [16] power for CO 2 purification [17] 0.504 0.40 0.104 Heat consumption [kWh/ kg CO 2 ] [16] 1.60 3.4 Fischer-Tropsch diesel Fischer-Tropsch (FT) fuels, such as synthetic diesel, gasoline or kerosene from fossil-derived syngas are proven technologies at commercial scale. Some examples are gasification of coal, like Sasol’s 160,000 barrels per day Coal-to-Liquid facility in South Africa [18], and steam reforming of natural gas, like Shell’s 140,000 barrels per day Pearl Gas-to-Liquid plant in Qatar [19]. So far, PtL projects using FT synthesis to produce hydrocarbons, like the Sunfire pilot plant in Dresden (Germany), are still in a demonstration/ research scale with low capacities. Nordic Blue Crude together with Sunfire and Climeworks have announced their intention to build a 20 MW plant (related to electricity input) by 2022 [20]. The simplified FT reaction can be described as the following: = -152 kJ/ mol [21] Carbon monoxide (CO) is a constituent of syngas, but in the case of PtL projects, it has to be obtained from reduction of the captured CO 2 . The reverse water gas-shift reaction (RWGS) allows this conversion at high pressures (30 bar) and temperatures (800-1000 °C). The selectivity can be improved by recycling the unreacted constituents, although reducing the energy efficiency of the process [22]. = +42 kJ/ mol [22] Beside alkanes, some alkenes, alcohols and carboxylic acids are formed. For all products, the molar ratio of the syngas is approximately 2 mol H 2 to 1 mol CO. The product slate is predominantly influenced by the temperature of the FT reaction, on average resulting in lighter hydrocarbons for high temperatures FT (320-350°C) and heavier hydrocarbons for low temperatures FT (190-250°C). To maximize the yield of transport fuels (gasoline, diesel, kerosene) a low temperature FT was chosen for this study, combined with a hydrocracking step to convert the produced wax into LPG, gasoline, diesel and kerosene. This results in a product slate indicated in of 37% gasoline, 28% diesel, 32% kerosene and 3% LPG [23]. Mass balance, electricity and thermal energy inputs for the FT synthesis as well as refining/ hydrocracking information are shown in Table 3. Table 3: Technical parameters of Reverse Water Gas Shift + Fischer-Tropsch + Hydrocracking system H 2 consumption [kg/ kg fuel ] 0.50 CO 2 consumption [kg/ kg fuel ] 3.80 Light ends production [kg/ kg fuel ] 0.27 Fuel production [kg fuel ] 1 Water production [kg/ kg fuel ] 3.03 Power consumption [kWh/ kg fuel ] 0.53 Net heat production [kWh/ kg fuel ] 2.56 Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 170 13th International Colloquium Fuels - September 2021 3.5 Methanol-to-Gasoline (MTG) Synthetic methanol from electricity for the methanolto-gasoline (MTG) synthesis can be either supplied via two-step synthesis using a synthesis gas (2-step route) or a one step process that uses CO 2 directly as feedstock (direct route). A first example of a PtL plant using the direct methanol synthesis is the George Olah plant in Iceland, run by CRI with a capacity of 4,000 t/ a [24], whose material balance was used as a reference. = -49.2 kJ/ mol [24] The methanol-to-gasoline (MTG) process was developed by Mobil in the 1970s. Commercial applications of the MTG process with capacities up to 1 Mt/ a (Shanxi, China) have been realised [25]. Within the process, methanol is partially dehydrated to dimethyl ether (DME), followed by olefin and aromatic/ paraffin formation using a zeolite catalyst [26]: The overall mass and energy balances of both steps of the reaction are shown in Table 4. Table 4: Technical parameters of Methanol synthesis + Methanol-to-Gasoline reaction H 2 consumption [kg/ kg fuel ] 0.47 CO 2 consumption [kg/ kg fuel ] 3.42 Light ends production [kg/ kg fuel ] 0.08 Fuel production [kg fuel ] 1 Water production [kg/ kg fuel ] 2.81 Power consumption [kWh/ kg fuel ] 1.26 Net heat production [kWh/ kg fuel ] 1.44 Like for the FT pathway, the additional energy supply required for carbon capture would drop if a higher concentration point source were used instead of atmospheric CO 2 . 3.6 Hydrotreated Vegetable Oil (HVO) Hydrotreated Vegetable Oil (HVO) is a biofuel, constituted by a mix of paraffinic hydrocarbon chains obtained from hydrotreatment and hydrocracking of triglycerides of vegetable origin. Triglycerides, the main components of vegetable fats, are esters derived from glycerol and linear carboxylic acids known as fatty acids. By the action of hydrogen, triglycerides undergo several reaction steps: • Hydrogenation to saturate the double bonds present in the linear structure of the vegetable oils. • Hydrocracking of the saturated triglycerides to break their esteric structure into three fatty acids and propane. The latter, along with small quantities produced of LPG and naphtha (around 5% the mass of HVO) can be used to produce electricity and steam. • Either deoxygenation or decarboxylation of the fatty acids, which produces water or carbon dioxide respectively along with linear hydrocarbons of around 17 to 18 carbons. These molecules have similar carbon length as those of fossil diesel with a more paraffinic nature, leading to higher cetane number but lower density and poorer cold properties. Further processing (isomerisation) can be applied to improve some of these properties. The chemical reactions are represented below: Triglycerides are also the raw material to produce another type of biofuels known as Fatty Acid Methyl Esters (FAME). FAME is formed by a mix of esters from esterification of vegetable fats, therefore its chemical nature is different from HVO’s. We propose two types of HVO feedstock for this study: A mix of rapeseed and sunflower oil (50% of each on an energy basis) to represent first-generation (1G) biofuels, and used cooking oil (UCO) for second-generation (2G) biofuels. In our assumptions, all raw vegetable oils as well as the used cooking oil are produced in Europe. UCO is assumed to be a burden-free waste and as such, it does not support any allocation of any impact. Hydrogen consumed in the process is assumed to come from steam methane reforming (grey hydrogen). The material and energy balances used for HVO production are taken from public data from Neste Oil [27] [28] and shown in Table 5. Table 5: Technical parameters of HVO (2G) production UCO consumption [kg/ kg fuel ] 1.21 H 2 consumption [kg/ kg fuel ] 0.04 Light ends production [kg/ kg fuel ] 0.05 Fuel production [kg fuel ] 1 Water production [kg/ kg fuel ] 0.10 Waste production [kg/ kgfuel] 0.10 Power consumption [kWh/ kg fuel ] 0.04 Heat consumption [kWh/ kg fuel ] 0.19 The UCO supply (collection and transport) is considered to be carried out as follows: • Feedstock collection: 50 km collection route by truck within urban area, 50% payload, 14-20 t gross vehicle weight (GVW) Feedstock transport: 100 km distance by truck between collection location and HVO plant, 34-40 t GVW. 3.7 Passenger vehicles The reference vehicles considered for the present study are passenger vehicles of the C-Segment (medium cars), similar to VW Golf, Ford Focus or Opel Astra. Powertrains include 4 variants of 3 types of powertrains: Internal combustion engine vehicles (ICEV) running on either gasoline or diesel, battery electric vehicles (BEV), and fuel cell hydrogen electric vehicles (FCEV). The weights of the vehicles in 2030 are based on the JEC Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 13th International Colloquium Fuels - September 2021 171 TTW report v5 [29], with some adjustments for electrified vehicles in link with the weight of the batteries. 3.8 Batteries Lithium-ion batteries used by BEVs, PHEVs, HEVs and FCEVs were assumed to be built with chemistry technology Nickel-Manganese-Cobalt in proportions 80: 10: 10 respectively (NMC 811), with an energy density of 250 Wh/ kg. Europe was chosen as the manufacturing location at horizon 2030. The emission factors of aluminium, electricity and thermal sources were calibrated consistently. The weights of the batteries were calculated from the life cycle inventories taken from the BatPac [30] model. 4. Model parameters 4.1 Assumptions for vehicles and batteries The main assumptions and input parameters to calculate TTW GHG emissions are summarized in Table 2. TTW emissions are calculated based on energy consumption of vehicles in MJ per km and fuel emission factors per MJ. Vehicle energy consumptions (MJ/ km in WLTP) are derived from 2025+ figures in the JEC TTW study v5 [29]. The lifetime average vehicle mileage is assumed to be 200,000 km for all vehicle types. During the end-of-life stage of vehicles and batteries, two types of emission credits are assumed: • Credits related to the recovery of materials such as steel, aluminium, polypropylene, copper, glass or magnesium, present mainly in the chassis and internal structure of the vehicles. Battery acid as well as fluids are also partly recovered. For batteries, this is extended to the recovery of rare materials from the cathode (lithium, cobalt) and for fuel cells, to platinum. • Credits related to the production of energy from material incineration, mostly tyres, polymers and carbon fiber. Table 6: Assumptions for passenger vehicles Powertrain WLTP Energy consumption [MJ/ km] Battery size [kWh] ICE Gasoline 1.41 (4.4 L/ 100 km) -- ICE Diesel 1.30 (3.6 L/ 100 km) -- BEV (200 km range) 0.43 (11.9 kWh/ 100 km) 24.7 FCEV* 0.70 (0.58 kg/ 100 km) 1.2 *Assuming a fuel cell power of 100 kW 4.2 Assumptions for fuels and electricity Electricity and thermal energy sources are used throughout all the processes of fuel production and vehicle manufacturing. The sources are consistent with the location chosen, Europe. For the electricity consumed in PtL and hydrogenation processes, we assume the use of 100% renewable electricity as a default value for PtL production in this study. By default, heat integration was active for the PtL processes, meaning that the heat produced by the exothermic reactions (typically fuel synthesis) is recovered and made available for upstream processes requiring heat (carbon capture, reverse water-gas shift) assuming an 80% recovery. Additional heat required in the process is obtained from electricity via a heat pump with an efficiency of 90% heat recovered from the electricity consumed. Table 7 summarises the impact categories for all the sources of energy used in this study per unit of energy (MJ) of fuel. Table 7: Assumptions for vehicle energy sources Fuel Fuel type Lower heating value (LHV) [MJ/ kg] EF 3.0 Climate change factor [g CO 2 -eq/ MJ] EF 3.0 Water use [Lworld-eq/ MJ] EF 3.0 Eutrophication, freshwater [mg P-eq/ MJ] Diesel Fossil 43.1 87.7 0.2 0.02 HVO (1G) 44.0 38.4 11.7 9.38 HVO (2G) 12.0 ~0 0.01 e-FT diesel 9.9 1.9 0.10 Gasoline Fossil 43.2 94.4 0.2 0.05 e-MTG 9.6 1.9 0.10 Electricity EU mix* - 86.2 15.2 0.42 Wind 2.9 0.15 0.01 Hydrogen Grey 120.0 85.7 0.1 0.16 Green 5.4 3.5 0.04 HVO (2G) = Second generation hydrotreated vegetable oils / FT = Fischer-Trosch / MTG = Methanol-to-Gasoline *Assuming mix from IEA Current Policies scenario (2030) 5. Results and Discussion 5.1 Greenhouse gas emissions The results of cumulated greenhouse gas emissions generated during the life cycle of a passenger vehicle for Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 172 13th International Colloquium Fuels - September 2021 all the vehicle/ fuel combinations are shown in Figure 1. Both fossil fuelled vehicles emit the largest amount of GHGs (26.4 and 30.1 t CO 2 -eq for diesel and gasoline ICEVs respectively), essentially produced during the use phase. These emissions include not only tailpipe gases (representing 73-83% of the total use phase) but also emissions from the production of the fuel and crude oil extraction. A small part of the impact of this use phase (2.0-2.5%) comes from diverse operating materials consumed like fluids or tyres. Results for ICEV compared to BEVs (12.4 t CO 2 -eq when using grid electricity) and FCEVs (17.4 t CO 2 -eq when using grey hydrogen), show that the difference in use phase emissions largely outweighs the initial disadvantage of electrified alternatives related to their manufacturing phase emissions. This disadvantage comes from their more energy-intensive manufacturing process associated to elements such as the lithium-ion battery, the fuel cell and the compressed hydrogen tank. The carbon burden associated to these elements is quickly compensated by the combination of the higher engine efficiency and, when using cleaner sources like wind electricity or green hydrogen, a lower carbonintensive energy source. This apparently inherent disadvantage for ICEVs can be drastically reduced when replacing the fossil fuel with a renewable alternative, either biofuel (HVO) or PtL fuel (MTG gasoline, e-FT diesel). The GHG emissions from a conventional diesel ICEV can be almost cut by half when substituting the fossil fuel by a first generation biofuel. The reduction can be further halved if the biofuel is produced from waste (2G). These calculations are based on a specific allocation methodology applied to biofuels. Variations of this methodology could result in meaningful differences not covered by the present study but explored in a separate piece of work [2]. The benefits of switching to renewable fuels also applies to electrified transport solutions, as shown in Figure 1. The impact of the use of wind electricity in BEVs or green hydrogen produced with wind electricity for FCEV (green H 2 ) also lowers substantially the climate change impact of the conventional variants. When focused on the use phase of these vehicles, the reduction reaches levels close to 90% in both cases. This relative reduction is higher than in the case of ICEVs, but it does not affect the initial burden of the vehicle and battery manufacturing stage. It is important to notice that the burden of electrified solutions running on full renewable energy are of the same order of magnitude than conventional vehicles using renewable fuels. Figure 1: Cumulated GHG emissions of passenger vehicles in 2030 5.2 Water use Figure 2 depicts the results of water use (measured according to the Water Scarcity impact category from EF 3.0) for all the transportation options analysed. Conventional fossil-fuelled ICEVs and FCEVs show similar levels of water consumption, of around 0.9 million litres of world-equivalent water per vehicle. BEVs powered with the EU Grid mix from 2030 (based on the IEA’s New Policies Scenario) represent a consumption of more than 2.5 times the conventional options, with burdens coming from both the manufacturing and use phases. Figure 2: Cumulated water scarcity (AWARE) of passenger vehicles in 2030 Similar to the case of the Climate Change impact, the Water Scarcity level of BEVs and FCEVs of the manufacturing stage is significantly higher than their ICEV equivalents, representing an even higher difference of about 75%, although the credits from the end-of-life partially compensate this difference. Figure 3 shows a Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 13th International Colloquium Fuels - September 2021 173 breakdown of the different contributions to water scarcity during the manufacturing phase. More than half of the water consumption comes from the manufacturing of the battery, and particularly of the cathode. Three main factors contribute to the impact the cathode: The energy consumption, the production of nickel and the production of lithium, which consumes substantial quantities of water. The electric engine is the third contributor (almost 11% of manufacturing water consumption), explained by the use of copper and aluminium. Direct water utilization for vehicle manufacturing accounts for only 2.5% of the total water scarcity index. Indirect water utilization via the manufacturing process of the different materials scattered in the categories of Figure 3 account for more than 70% of the contribution, with about one third of that contribution coming from rare metals (cobalt, lithium, nickel, manganese) and plastics (water consumption during the olefins production). The rest of the contribution comes from indirect water consumptions for energy generation (heat and electricity). Figure 3: Breakdown of water scarcity contributions of a battery electric vehicle (manufacturing phase) The direct use of water during the vehicle manufacturing (excluding the battery) is shown in Table 8 according to public data provided by two vehicle manufacturers. Most of the water consumed during the assembly processes comes from treated water. Almost two thirds of the water used in the processes is recovered and treated before releasing it to the environment. Table 8: Water consumption during the vehicle manufacturing stage [m 3 per vehicle] BMW [31] VW [32] Water Input 2.32 3.56 of which Ground water 0.29 0.53 Surface water 0.00 0.10 Tap water 2.03 2.93 Water output 1.53 2.67 of which Process waste water 0.44 Sanitary waste water 1.09 Balance (water vapour, embodies, other losses) 0.79 0.89 The use phase of the vehicles in Figure 2 accounts for an important part to the water scarcity index. In the case of ICEVs, the impact is limited for fossil fuels due to a low net use of water during the crude oil and refining process. In the case of first generation biofuels, the impact is the most important of all the options studied, mainly because of the water required for irrigation of the rapeseed and sunflower crops required to their production. These impacts are reduced to levels comparable to fossil-fuelled ICEVs when second generation fuels are used assuming the used cooking oil feedstock does not carry any burden from its original use. ICEVs running on PtL fuels show higher water consumption levels than fossil and 2G options because of the indirect use of water for electricity generation. The water used for hydrogen production (electrolysis) is mostly recovered during the synthesis process, therefore carrying a relatively low water intensity per MJ (Table 7). The effect of water use in electricity generation compared to a fossil source can be also observed in the case of FCEVs. The use of green hydrogen increases the impact of water use because of its water intensity. Combined with a fixed contribution from the wheels replacement, this results in the use phase water consumption for a renewable-fuelled fuel cell vehicle to be multiplied by almost 2.5 compared to another running with conventional hydrogen. In the case of BEVs, the switch from an electricity mix to a wind electricity cuts down effectively the overall water scarcity factor of the vehicle by a factor of 2.3. The origin of the electricity plays therefore a major role in the water consumption during the use phase. The EU energy mix of 2030 is beyond 100 times more water-intensive than wind electricity. 5.3 Freshwater eutrophication In Figure 4 we observe the impacts of the different vehicle/ energy alternatives on freshwater eutrophication. The results for this environmental impact, measuring the amount of excess nutrients added to freshwater bodies, Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 174 13th International Colloquium Fuels - September 2021 mostly rivers, lakes and ground reservoirs, are extremely high for the case of first-generation HVO. This is explained by the use of phosphate-based fertilizers in agriculture, directly affecting the sources of blue water. The rest of the results are less differentiated for ICEVs and FCEVs, regardless of the type of fuel consumed. BEVs stand out as a separate category of considering the impact of the vehicle manufacturing only: Freshwater eutrophication of BEV manufacturing accounts for more than twice the impact of conventional and fuel cell vehicles. The eutrophication credits from the end-of-life cycle, however, neutralise these oversized impacts to levels comparable to fuel cell electric vehicles. Figure 4: Cumulated freshwater eutrophication of passenger vehicles in 2030 The main impacts related to the manufacturing stage are represented in Figure 5. The biggest contributors to these impacts are the production of cobalt and lithium. For cobalt, the blasting stage has been investigated as the main responsible for causing eutrophication, because of the nitrogen oxides emitted during the blasting stage [33]. Nitrogen oxide emissions are at the origin of smog and acid rain, which in turn cause eutrophication. In the case of lithium, eutrophication comes from phosphates released in the process water used to treat spodumene, a mineral from which lithium is extracted in Australia or China (this is not the case in South America where lithium comes mainly brine). Sodium phosphates are used to precipitate lithium from alkaline solutions [34]. Another important contributor to eutrophication is the wheel manufacturing, related to the nitrogen oxide emissions during rubber manufacturing [35]. Note that the value for wheels in this graph only accounts for the initial set, so the total contribution that includes the impact from the tyres in the use phase is estimated to 5 times the value represented in this graph. The differences in the use phase of the vehicles are not clearly distinguished. General trends can be observed, which show higher levels of eutrophication related to the use of PtL fuels. Like in the case of water consumption, the main explanation for this difference comes from the indirect pollution during the electricity generation phase. For BEVs, the change from the EU grid mix to renewable electricity represents a drop of 40% of use phase eutrophication, which is mainly explained by a reduction of 75 times the phosphorus intensity. The change of tyres included in this phase makes this difference less noticeable in the graph. Figure 5: Breakdown of freshwater eutrophication contributions of a battery electric vehicle (manufacturing phase) 5.4 Limitations of the study The present study is introduced as a first assessment for a limited number of powertrain combinations and only one sensitivity on fuel type per vehicle. Several parameters play a role in determining the level of pollutants: Most of the PtL processes are still in demonstration status or do not exist at all in the assessed configuration, operational data about auxiliary consumption and especially process emissions into air and water (besides CO 2 ) are rather restricted, which is a limitation of the study, especially for impact categories other than climate change. Concerning water, the results are limited to a specific regional impact. An extension of the study could include the variation of these impacts depending on the geographical zone they cover, as the water treatment standards change from country to country. Also, a more detailed breakdown of different sources of water, such as ground or desalinated seawater which could provide better insights about particular actions that can be taken to tackle the environmental effects. Similar regionalised studies can be applied to some of the vehicle materials, in particular aluminium, copper and lithium. Impact of second-generation biofuels is highly dependent on methodological parameters. Results can be significantly impacted if some sort of economic allocation can be applied to used cooking oil, considering its market value similar to raw oil. Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 13th International Colloquium Fuels - September 2021 175 Further options potential improvements investigated: - The impact of different sources of electricity for the production of PtL fuels, especially hydrogen. This could be extended to the hydrogen used in the biofuels production process. The utilization of blue hydrogen (produced from an SMR with an integrated carbon capture system) can also be assessed. - The impact of the use of renewable fuels for transportation of biomass. The present study assumes fossil fuels consumed by trucks. - The utilization of hybrid mobility solutions (hybride, plug-in electric vehicles). 6. Conclusions The present study aimed at analysing the impact factors for passenger vehicles beyond the usual assessment of greenhouse gas emissions, including water scarcity and freshwater eutrophication. The following conclusions can be drawn from this piece of work: • The generalisation of the use of renewable energy for transport drastically diminishes the initially clear advantage of electrified technologies with respect to internal combustion engine vehicles related to greenhouse gas emissions. • Water use during the vehicle manufacturing phase is 70% constituted by indirect consumption to produce vehicle materials, especially aluminium, copper and lithium, but also plastics. In battery electric vehicles, this indirect use represents more than half of the consumption. • Fossil and PtL fuels do not show significant differences in water consumption, despite the use of water for hydrogen electrolysis. Most of the water produced in the process is recovered. • The source of electricity is critical not only for greenhouse gas emissions, but also for water use and freshwater eutrophication. With respect to wind electricity, the average electricity mix in Europe at horizon 2030 is expected to have water use and eutrophication impacts per unit of energy roughly 100 and 75 times higher respectively. This has a direct impact for BEVs but also indirect for ICEVs runnion on PtL fuels. • First generation biofuels show relative advantages in Climate change impact with respect to fossil fuels, but they compare very poorly in water use and especially freshwater eutrophication, due to the large amounts of irrigation water and phosphate-based fertilizers used in agriculture. • The impact of freshwater eutrophication during the manufacturing phase of a battery electric vehicle comes mainly from the Li-ion battery component, and is related to the indirect emissions from cathode components, mainly cobalt and lithium. Their impact can be reduced depending on the level of recycling targeted for these rare materials. • Passenger vehicle technologies should not be dissociated from the source of energy when performing environmental assessments. The single impact of the energy efficiency of a given technology does not determine totally its impacts on the environment. For this purpose, Life Cycle Assessment (LCA) has proven to be a valuable tool to perform holistic comparisons between different transportation options. References [1] V. Gordillo et al, “Greenhouse Gas Emissions of Passenger Vehicles from a Cradle-to-Grave perspective,” in SIA Powertrains & Energy, 2020. [2] M. Yugo, V. Gordillo, E. Shafiei and A. Megaritis, “A look into the life cycle assessment of passenger cars running on advanced fuels,” in SIA Powertrain & Power Electronics, 2021. [3] International Organization for Standardization (ISO), “ISO 14040: 2006 Environmental Management - Life Cycle Assessment: Principles and Framework,” 2006. [4] International Organization for Standardization (ISO), “ISO 14044: 2006 Environmental Management - Life Cycle Assessment: Principles and Framework,” 2006. [5] Sphera, “GaBi 10.0 LCA Software & LCI databases v2021.1,” February 2020. [Online]. Available: http: / / www.gabi-software.com/ international/ index/ . [6] Joint Research Centre (JRC), “Supporting information to the characterisation factors of recommended EF Life Cycle Impact Assessment method,” Ispra: European Commission, 2018. [7] A.-M. Boulay et al., “The WULCA consensus characterization model for water scarcity footprints: assessing impacts of water consumption based on available water remaining (AWARE),” The International Journal of Life Cycle Assessment, vol. 23, p. 368-378, 2018. [8] B. Morelli et al., “Critical review of eutrophication models for life cycle assessment,” Environmental Science & Technology, vol. 17, no. 52, p. 9562- 9578, 2018. [9] Intergovernmental Panel on Climate Change (IPCC), “Climate Change 2013: The Physical Science Basis - Contribution to the Fifth Assessment Report of the IPCC,” 2013. [10] O. Schuller, “The GaBi LCI Refinery Model 2021,” 2021. [Online]. Available: https: / / sphera.com/ wpcontent/ uploads/ 2020/ 04/ The-GaBi-LCA-Refinery-Model.pdf. [11] Fuel Cell and Hydrogen 2 Joint Undertaking (FCH JU), “Hydrogen Roadmap Europe - A sustainabi- Beyond GHG: Cradle-to-grave impact of renewable energy-fuelled mobility on water consumption and eutrophication 176 13th International Colloquium Fuels - September 2021 lity pathway for the European energy transition,” 2019. [Online]. Available: https: / / www.fch.europa. eu/ publications/ hydrogen-roadmap-europe-sustainable-pathway-european-energy-transition. [12] P. Millet et al., “Chapter 2 - Water Electrolysis Technologies,” in Renewable Hydrogen Technologies, 2013, pp. 19-41. [13] International Energy Agency (IEA), “IEA Hydrogen Project Database,” 2019. [14] Intergovernmental Panel on Climate Change (IPCC), “Special Report on Carbon Dioxide Capture and Storage,” 2005. [15] N. von der Assen, “Selecting CO2 Sources for CO2 Utilization by Environmental-Merit-Order Curves,” Environmental Science & Technolgy, 2016. [16] C. Beuttler et al., “The Role of Direct Air Capture in Mitigation of Anthropogenic Greenhouse Gas Emissions (Frontiers in Climate),” Frontiers in Climate, 2019. [17] Element Energy, “Shipping CO2 - UK Cost Estimation Study: Final report for Business, Energy & Industrial Strategy Deparment,” 2018. [18] Sasol, “Coal Gasification and Liquefaction - SA Experiences and Opportunities,” 2012. [Online]. [19] Shell, “Knowledge Guide: GTL Fuel - Synthetic Technology for Cleaner Air v2.0,” 2016. [20] A. Winther Mortensen, “Nordic GTL - a pre-feasibility study on sustainable aviation fuel from biogas, hydrogen and CO2,” 2019. [21] D. H. König, „Techno-ökonomische Prozessbewertung der Herstellung synthetischen Flugturbinentreibstoffes aus CO2 und H2,“ Universität Stuttgart, 2016. [22] R. Detz, “CO production via reverse water gas shift,” 2019. [23] A. De Klerk, “Fischer-Tropsch Process,” in Kirk- Othmer Encyclopedia of Chemical Technology, 2013. [24] Carbon Recycling International (CRI), “Power and CO2 emissions to methanol,” European Methanol Policy Forum Brussels, 2015. [25] ExxonMobil, “ExxonMobil methanol to gasoline (MTG),” in Syngas Technology Conference, Colorado Springs, 2017. [26] S. J. a. Y. Zhu, “Techno-economic analysis for the conversion of lignocellulosic biomass to gasoline via the methanol-to-gasoline (MTG) process,” U.S. Department of Energy, 2009. [27] S. Nikander, “Greenhouse gas and energy intensity of product chain: Case transport biofuel,” Helsinki University of Technology, 2008. [28] Neste Oil, “NExBTL Renewable Diesel Singapore Plant - North American Technical Corn Oil Pathways Description,” 2014. [29] JEC, “JEC Tank-to-Wheel report v5: Passenger cars,” Publications Office of the EU, Luxembourg, 2020. [30] Argonne National Laboratory (ANL), “BatPaC: Battery Manufacturing Cost Estimation,” 2019. [Online]. Available: https: / / www.anl.gov/ tcp/ batpac-battery-manufacturing-cost-estimation. [31] BMW Group, “Sustainable Value Report 2020,” 2020. [Online]. Available: https: / / www.bmwgroup. com/ content/ dam/ grpw/ websites/ bmwgroup_com/ responsibility/ downloads/ en/ 2020/ 2020-BMW- Group-SVR-2019-Englisch.pdf. [32] Volkswagen Corporation, “Sustainability Report,” 2020. [Online]. Available: https: / / www.volkswagenag.com/ presence/ nachhaltigkeit/ documents/ sustainability-report/ 2020/ Nonfinancial_Report_2020_e.pdf. [33] S.-H. Farjana et al., “Life cycle assessment of cobalt extraction process,” Journal of Sustainable Mining, 2019. [34] Y. Song et al., “A promising approach for directly extracting lithium from α-spodumene by alkaline digestion and precipitation as phosphate,” Hydrometallurgy, 2019. [35] K. Piotrowska et al., “Assessment of the Environmental Impact of a Car Tire throughout Its Lifecycle Using the LCA Method,” Materials, 2019. [36] JEC, “JEC Well-to-Tank report v5,” Publications Office of the European Union, Luxembourg, 2020. [37] International Energy Agency, “The Future of Hydrogen,” 2019. [Online]. Available: https: / / www.iea. org/ reports/ the-future-of-hydrogen.