1 In-life range modularity for electric vehicles: the...

Sustainability 2017, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability

Article 1

In-life range modularity for electric vehicles: the 2

environmental impact of a range-extender trailer 3

system 4

Nils Hooftman 1,*, Maarten Messagie 1, Frédéric Joint 2, Jean-Baptiste Segard 2 and Thierry 5 Coosemans 1 6

1 Electrotechnical Engineering and Energy Technology, MOBI Research Group1, Vrije Universiteit Brussel, 7 Pleinlaan 2, Brussels 1050, Belgium ([email protected]) 8 9

2 EP Tender, 22 rue Gustave Eiffel, 78300 Poissy, France ([email protected]) 10 11

* Correspondence: [email protected]; Tel.: +32 (2) 629 37 67 12

Academic Editor: name 13 Received: date; Accepted: date; Published: date 14

Abstract: 15 Purpose: The range of an electric vehicle (EV) remains an important hurdle for a wide-spread 16 consumer adoption. Opting for an EV with an over-dimensioned battery not only significantly 17 increases the capital expenditure, it also increases the vehicle’s impact on climate change and 18 diminishes the EV’s inherent benefit concerning local air quality levels. The concept of in-life 19 modularity by means of a range-extender trailer is investigated as an alternative for high battery 20 capacities. Both a generator as a battery based range-extender are discussed. 21

Method: A life cycle assessment (LCA) is performed for comparing the combination of 40 kWh EV 22 and the trailer with a range of conventional cars and EVs, differentiated per battery capacity. An 23 application factor of 5% is assumed for the range-extender. In addition, environmental breakeven 24 distances are calculated for either replacing an existing ICE-based vehicle or for a one-on-one 25 comparison between EVs, the EV + generator combination and ICE-based vehicles. 26

Results: While offering the same electric range, the 40 kWh EV + generator trailer contributes 30% 27 to climate change, compared to a 90 kWh EV. In case of a 50 kWh battery trailer, this reduction 28 increased to 40%. This indicates in-life modularity has a positive impact on climate change. 29 Breakeven distances indicate that replacing an ICE-car with another one is no solution, when climate 30 change is considered. The range-extended 40 kWh EV becomes a better solution after approximately 31 45.000 km in case of replacing an existing car, while this distance is reduced to less than 10.000 km 32 when it is compared to a new PHEV. 33

Keywords: electric vehicles, range-extender, CO2, air quality, mobility needs, LCA, Paris 34 Agreement 35

36

1. Introduction 37

Air quality levels across Europe maintain to be problematic, and this especially in urban regions. 38 These are found to be hotspots for nitrogen oxides (NOx) and particulate matter (PM) [1]. Europe’s 39 domestic transportation has a substantial contribution to this ongoing issue and is nearly entirely 40 represented by the road transport sector [2]. In the light of mitigating local air quality levels, this 41 sector has been regulated for its pollutant emissions for more than two decades, i.e. by the so-called 42

1 VUB-MOBI Group is member of Flanders Make.

mailto:[email protected]



Sustainability 2017, 9, x FOR PEER REVIEW 2 of 27

Euro emission standards [3]. These standards are determined for both the heavy duty (i.e. long 43 haulage trucks) sector on the one hand and the light duty (cars + vans) sector, on the other hand. 44 Concerning transport’s impact on climate change, the light-duty sector is imposed to a target for 45 carbon dioxide (CO2) emissions. Therefore, each car manufacturer is required to obtain a corporate 46 average fleet fuel economy of 95 grams of CO2 per kilometre by 2021, as described in Regulation 47 2009/443 [4]. Battery electric vehicles (EV) are key assets in reaching this target, as they are given extra 48 weight in the balance by means of so-called super-credit factors [5]. These allow EVs to count for more 49 than one car in the fleet average calculation. An indirect effect of Regulation 443 is that car 50 manufacturers can, in fact, increase the emissions of their conventional technologies, as electric 51 vehicles bring down the fleet’s average [6]. This can lead to a higher uptake of conventional vehicles, 52 as an indirect allowance of higher CO2 emissions results in cheaper conventional cars, as their engines 53 require less cost-intensive calibrations [7]. The importance of the 2021 and future CO2 targets is 54 strengthened by the 2015 Paris Agreement, in which the majority of the world’s nations agreed to 55 strictly reduce greenhouse gas emissions (GHG, represented by CO2-equivalent gasses), in order to 56 maintain the global temperature increase well below 2°C, relative to pre-industrial levels [8]. Despite 57 the current lack of a well-defined roadmap towards this goal, a net-zero GHG economy is required 58 by 2050 or shortly thereafter [9]. For this reason, the EU strives to a minimum GHG reduction by 60% 59 for its transport sector by 2050, while the entire European economy is bound to reduce its GHG 60 contribution by 80 to 95% [10]. In the light of decarbonizing the light-duty fleet, electrification is 61 believed to play a major role. 62

63 Regardless of an absent post-2021 target for the passenger car sector’s CO2 emissions, car 64

manufacturers seem to have understood the potential of EVs as nearly all of them have either 65 introduced – or plan to launch EV models on the short-term [11]–[16]. Despite the growing number 66 of EV models presented by the manufacturers, the adoption of the technology remains marginal This 67 can be concluded from the 2016 sales share of EVs in Europe, which remained below 1 percent for 68 purely electric vehicles [17]. The absence of a pan-European approach for enabling electromobility 69 through a set of incentives is in contrast with China’s plans. With its major cities suffering severely 70 from air pollution, China will both issue strict NOx limits from 2020 onwards and plans to impose 71 quota for EV registrations to the automotive industry, from 2018 onwards [18],[22],[23]. Thus, China 72 targets a 20% market share for EVs by 2025. Such a top-down approach might prove to be the only 73 solution the induce a substantial shift away from conventional car technologies on the short-term. 74

A widespread adoption of EVs requires a paradigm shift in the mind of the consumer, as the 75 technology is characterized by a limited on-board energy storage. Both fiscal and financial incentives 76 to bring down the capital expenditure (CAPEX) for the technology as well as the widespread 77 availability of publicly available charging infrastructure have proven to be highly desired up to date 78 and will be required to be applied (at least) up until the point where cost parity with a conventional 79 car of the same segment is reached [24],[25]. Whereas the electric range is repeatedly indicated as one 80 of the major hurdles for EV breakthrough, it is a fact that the technology cannot cover the mobility 81 needs of every consumer. Large-scale travel surveys are an important source for estimating real-world 82 needs for the daily range for passenger car users. Examples of such studies can be found in Pearre et 83 al. [23] and Needell et al. for American studies [24]; and Pasaoglu et al. [25] and Corchero et al. [26] 84 for European variants. These different sources all indicate that the clear majority of daily range needs 85 are in the 0 to 80 km range. Moreover, ranges exceeding 150 km/day are found to occur only on a 86 limited number of days per year, i.e. for only 5% of the daily trips, confirming the findings by Khan 87 et al. [27] and Gonder et al. [28]. An exemplary distribution of the daily driven distances based on 88 Redelbach et al. is given in Figure 1 [29]. Characterizing for this distribution is that over 98% of the 89 daily driven distances by car is below 150 km. These examples show the potential for substituting 90 conventional cars with EVs for a substantial part of the year while relying on alternatives for the days 91 when larger distances are travelled. 92


93

94

Figure 1: Example of an average daily driven distance distribution (based on [29]) 95

Alternatives could either be a conventional car or to opt for an EV with a bigger battery pack. A 96 trend noticeable among the available EV models is precisely this increasing battery capacity. One 97 could reflect on this trend until which point it is remains both economically and environmentally 98 feasible, as the more battery cells are produced, the higher the EV’s environmental impact becomes 99 and the smaller the differences become with ICE-vehicles. Following the results from Needell et al., a 100 40 kilowatt-hour (kWh) EV could cover up to 95% of the average daily range needed. In this 101 perspective, a 90 kWh EV would have to carry on 50 kWh of capacity as a ‘burden’ that is only rarely 102 applied. The use of this over-capacity is from here on referred to as the marginal application. Next to its 103 adverse effects caused during its manufacturing, other consequences have a monetary nature. 104 Despite the price per kWh is dropping faster than expected for EV battery packs [30][31], it remains 105 the most cost-intensive part of the vehicle. Opting for an oversized battery hence affects the purchase 106 price of the vehicle. On the other hand, there is the linear relation between vehicle mass and its energy 107 consumption when urban driving is considered. Due to the frequent stopping and going in urban 108 situations, a poor ‘energy economy’ results in an increase of the operational expenditure (OPEX) of 109 the vehicle. 110

Another alternative solution brought up by the automotive sector is a plug-in hybrid electric 111 vehicle (PHEV), combining both a powerful internal combustion engine (ICE) with an externally 112 chargeable, electric powertrain. Although PHEVs seemed rather an inventive way for obtaining 113 super-credit factors for manufacturers (up until 2016) and for bypassing vehicle taxes, than a means 114 to effectively reduce local air quality pollution [12][13][34], the question remains whether this 115 expensive dual technology will be the flagship for reaching the future emission targets or if it will 116 simply become obsolete due to the general adoption of the EV. 117

A more pragmatic way for covering the marginal application of the EV is to occasionally connect 118 it to a range-extender (ReX), while performing every other moment when the nominal electric range 119 suffices with an adequate battery capacity of e.g. 40 kWh. A possible concept could consist of a trailer, 120 fitted with either a mobile petrol generator or an extra battery pack, allowing the EV’s battery to be 121 used in a charge-sustaining mode for an additional 300 km or so. Whereas extra range is typically 122 required to cover long distances over highways, this range-extender concept could complement fast-123 charging stations located near highways. In an idealized situation, an automated connection of the 124 trailer to the EV would take place, hence minimizing the interference of the EV user. 125

126 The objective of this paper is to assess the effect the range-extender trailer can have on an EV’s 127

battery capacity, given its environmental impact and the impact on urban air quality. Moreover, the 128

0%

20%

40%

60%

80%

100%

0%

5%

10%

15%

20%

25%

Distance [km]

Distribution of the daily average driven distances by car

Frequency Cumulative


40 kWh + trailer combination is compared a range of mid-sized family cars, based on different 129 powertrains. With the aim of proving that an over-dimensioned battery pack has a negative 130 environmental impact besides the economic disadvantage, one high-end EV with a battery pack 131 capacity of 90 kWh was added to the vehicles list. The scope of this investigation focusses on the 132 European market and power production mix. The discussed EVs are assumed to be charged with the 133 average EU mix at an average CO2 emission intensity of 276 grams per kWh of electricity produced 134 [35]. The authors of this paper deliberately chose not to address marginal energy production for 135 generating the electricity for EVs, as EVs are thought to be part of the total load system, confirming 136 the viewpoints found in [36][37][38]. A secondary objective is to determine after how many 137 kilometres the EV+ReX combination becomes a more environmentally solution when climate change 138 is concerned. This is done by means of calculating the so-called breakeven points with both ICE-based 139 powertrains and different EVs. 140 141

2. Methodology 142

a. Life cycle assessment 143

An environmental Life Cycle Assessment (LCA) is used to compare the impacts, damages and 144 benefits of the combination of an EV+ReX trailer while considering all the associated emissions, both 145 direct and indirect. This process considers a Life Cycle Inventory (LCI) of every emission and raw 146 materials that are used throughout the different product stages - manufacturing, use and end of life. 147 The advantage of separating the different product life stages enables the identification of the causes 148 of specific impacts and emissions per stage in the product’s value-chain [39]. A full LCA is performed 149 for assessing the impact on climate change. Considering the impact on urban air quality, a deliberate 150 choice was made to focus only on the emissions which are produced locally, based on Hooftman et 151 al. [39]. Therefore, refinery-to-tank (RTT) emissions are applied instead of the conventional well-to-152 tank (WTT) emissions. The reason for this choice is that e.g. the impact from mining raw materials 153 for an EV’s battery pack shouldn’t be considered when local emissions are concerned. A graphical 154 overview of the applied assessment scope with the relevant life cycle stages is given in Figure 2. 155

156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173

174

Figure 2: Flowchart of the LCA approach (based on [40][39]) 175

Product life cycle

OPERATION

Well-to-wheel life cycle

Tank-to-wheels

Energy resource

extraction

Energy carrier

production

Energy carrier

distribution

Energy

conversion Maintenance

Product

manufacturing

Material

production

End-of-life

Well-to-tank

Complete life cycle

Emissions within the national border

Refinery-to-tank

Non-exhaust emissions

Emission of non-regulated pollutants


The selected impact assessment methodology which was applied in the SimaPro 8.3 software is 176 ReCiPe midpoint (H) [41]. Out of a set of 18 midpoint impact categories, 4 are discussed in this paper 177 as they represent both GHG emissions and air quality in urban environments. These are Climate 178 Change (CC), Photochemical Oxidant Formation (POF), Human Toxicity (HT), and Particulate Matter 179 Formation (PMF). These midpoint indicators serve as an intermediate between the emission source 180 and the ‘endpoint’, representing the recipients of the environmental effects caused by anthropogenic 181 activities, as there are Human Health, Ecosystem Quality and Natural Resources [42]. No endpoint 182 indicators are discussed in this paper. 183

Concerning the midpoint indicator for climate change, CO2-equivalents represent the group of 184 greenhouse gasses. The main drivers of POF are elements from the family of benzenes, nitrogen 185 oxide(s) and other non-methane organic compounds, which are precursor gasses for ground-level 186 ozone (O3) formation. POFs are in this paper represented by the group of non-methane volatile 187 organic compounds (NMVOC). As for HT, dioxins, cadmium, silver, and zinc among others 188 contribute to the impacts in this category, grouped as 1,4-dichlorobenzene equivalents (1,4-DB). 189 Particulate Matter Formation highlights the impacts of primarily formed particulates as well as 190 particulates formed by the condensation of nitrogen oxides sulphur oxides, ammonia and non-191 methane volatile organic compounds (secondary PM). PMF is represented by the emission of PM10-192 equivalents, i.e. particles with an aerodynamic diameter smaller or equal to 10 micrometres. Both 193 PMF from combustion as from tyre, brake and road wear is considered. 194

The applied methodology is based upon the work by Messagie et al [36] and allows to distinguish 195 the different impacts per technology per functional unit, in this case, one kilometre driven. The entire 196 life cycle impact is calculated by considering a lifespan of 209.460 km. Concerning the use phase 197 impacts, an update of the official TTW emission factors was performed to reflect real-world driving 198 emissions (RDE), based upon Hooftman et al [39]. The reason hereto is that the European type-199 approval process, based upon the New European Driving Cycle (NEDC), significantly 200 underestimates both the emissions of greenhouse gasses and toxic pollutants [37][38]. Moreover, the 201 emission of non-regulated pollutants and non-exhaust PM are included as well, to allow an extensive 202 comparison of the impact on local air quality levels. 203

b. Environmental breakeven points 204

Environmental breakeven points (BEP) are calculated to indicate the distance after which the 205 impact of vehicle technology X exceeds the impact on climate change of technology Y. BEPs are 206 calculated for two scenarios. In the first scenario, the replacement of an existing ICE car with a new 207 EV is considered. Therefore, the manufacturing of the EV generates an impact which can be read as 208 an initial offset compared to the ICE car’s case, as the latter was manufactured earlier on. The slope 209 of the impact curves for the different technologies in the comparison over time represent the use stage 210 and includes maintenance. Where curves intersect, the BEP is reached for the given combination of 211 vehicle technologies. Equation 1 shows how this distance is calculated [45]. 212

𝑑𝐵𝐸𝑃 =𝐼𝐸𝑜𝐿,𝑣𝑒ℎ𝑖𝑐𝑙𝑒 1+𝐼𝑀𝑎𝑛,𝑣𝑒ℎ𝑖𝑐𝑙𝑒 2

𝑖𝑈𝑠𝑒,𝑣𝑒ℎ𝑖𝑐𝑙𝑒 1−𝑖𝑈𝑠𝑒,𝑣𝑒ℎ𝑖𝑐𝑙𝑒 2𝑘𝑚 Equation 1 213

Where dBEP is the distance at which a BEP is reached, IEoL, vehicle 1 is the impact of recycling the old 214 vehicle, IMan, vehicle 2 is the impact of manufacturing the new vehicle and iUse, vehicle j is the impact of the 215 use phase (well-to-wheel and maintenance) of the respective vehicle. In the second scenario, the 216 different powertrain technologies are compared on the same basis, in fact simulating the situation 217 where a new vehicle is to be bought and both manufacturing and use impacts are considered. Like in 218 scenario 1, the outcome is distance at which the purchase of an EV becomes environmentally more 219 interesting. 220

c. Assumptions 221

For the comparison of the different powertrain technologies, the 40 kWh EV is chosen as the 222 reference to be combined with the range-extender. Also, the marginal application covered by the 223


trailer is assumed to be 5% of the vehicle’s lifetime driven distance. Moreover, one trailer is assumed 224 to be shared by 15 users, resulting in the fact that its manufacturing impact is subsequently shared 225 over these 15 users. For the remaining 95% of the EV’s lifetime, it is assumed to drive purely electric 226 whilst charging with the European electricity mix, characterised by a CO2 emission intensity of 276 g 227 per kWh produced [35]. Equation 2 shows how the well-to-tank emission is based on the EV’s average 228 consumption (kWh/km), while the ReX trailer is assumed to be applied for covering long distances 229 over highways. For this reason, the EV’s highway consumption added with the influence of towing 230 the trailer was used for the reflecting the impact of the 5% of the time the ReX trailer is used. An 231 arbitrary extra consumption due to the trailer of 5% was chosen, corresponding to the marginal need 232 for longer ranges than provided by the current 40 kWh EVs [24], [27], [28]. The specifications for the 233 generator trailer concept are given in Table 1. 234

𝑊𝑇𝑇𝐸𝑉40+𝑅𝑒𝑋 = (0,95×𝑊𝑇𝑇𝐸𝑉40,𝑎𝑣𝑔) + (0,05×(𝑊𝑇𝑇𝐸𝑉40,ℎ𝑖𝑔ℎ𝑤𝑎𝑦 + 𝑊𝑇𝑇𝑅𝑒𝑋)) Equation 2 235

Table 1: Overview of the generator trailer characteristics 236

Generator range-extender

Rated power [kW]

Mass [kg]

25

265

Fuel tank [l] 35

Fuel type Petrol

Range [km] 300

Average consumption [l/kWh] 0,44

Average consumption [l/100 km] 7,5

237

The presented work is based upon the Life Cycle Inventory (LCI) created by Messagie et al, which is an 238 elaborated source of virtually all needed materials, chemicals, energies and emissions related to the fulfilment 239 of the functional unit [36]. This database has been updated as it originally considered the Belgian electricity mix 240 (190 gCO2/kWh) on the one hand and covered only one EV model (the 2011 Nissan Leaf (24 kWh)) on the other 241 hand. The different EVs used in this paper have been derived from the 24 kWh Leaf by scaling the parameters 242 which are influenced by 1) the mass in running order, 2) the battery size, and 3) the energy consumption. The 243 LCI data for the conventional technologies have been kept up to date to reflect the most recent Euro 6 standard 244 and its related characteristics. Real-world emission factors for the tank-to-wheel emissions for conventional cars 245 are based upon Hooftman et al [39]. A deliberate divergence from the official emission factors was chosen to 246 allow a fair comparison between the different technologies. For the EVs, the energy consumption data published 247 by the U.S. Department of Energy (DOE) was used [46]. The specifications of the assessed EVs are given in Table 248 2, while the assumed energy consumption factors for the discussed ICE cars are given in Table 3. 249

Table 2: Specifications of the discussed electric vehicles (based upon [46]) 250

Parameter [unit] 30 kWh EV 40 kWh EV 60 kWh EV 90 kWh EV

Capacity [kWh] 30 40 60 90

Mass in Running Order [kg] 1591 1450 1624 2200

Weight battery [kg] 272 305 435 540

Average consumption

[kWh/km] 0,15 0,15 0,15 0,25

Highway consumption

[kWh/km] 0,20 0,20 0,20 0,26

Highway consumption with

ReX trailer [kWh/km] 0,210 0,210 0,210 0,273

251


Table 3: Overview of the fuel consumption indicators per technology 252

Unit Petrol Petrol

hybrid Diesel

Plug-in

electric

vehicle

[l/100 km] 6,8 5,6 5,3 3,4

3. Life cycle inventory 253

Based upon a life cycle inventory (LCI), the elementary flows which are linked to the various 254 vehicle technologies need to be converted to the different impact categories. These allow a 255 quantification and a comparison between the potential impacts. This step is referred to as the life 256 cycle impact assessment (LCIA). Note that the different presented EVs are based upon existing 257 models for the European market. An exemplary full LCI of a 30 kWh EV is given in Table 8 in the 258 appendix. Concerning the environmental performance of the generator ReX trailer concept itself, the 259 product’s detailed LCI can be found in Table 9 and Figure 10 in the appendix while Table 5 260 summarizes the parameters of the different impact categories. The emission factors for a suitable 261 generator are given in Table 4 . 262

Table 4: Emission factors for the generator ReX trailer [47] 263

Unit Average generator emissions

HC CO NOx CO2 HC+NOx

[g/kWh] 2,525 40,485 1,099 999,406 3,624

264 In Table 5, a deliberate distinction was made between the production of the ‘trailer body’ and 265

the production of the generator set, while the operation of the latter was analysed during its use 266 phase. This choice was made to allow a better insight in the allocation of their respective contribution 267 to the midpoint indicators. The total lifetime of the trailer was chosen to be identical to that of a 268 passenger car itself, namely 209.470 km. The European electricity mix is included, representing the 269 95% of the time during which the ReX trailer is decoupled from the EV. For all four midpoint 270 indicators, it’s this ‘EV part’ which is responsible for the approximately three quarters of the 271 respective impact. The fact that the generator ReX trailer has a significant impact for being active only 272 5% of the time emphasizes the potential environmental improvements if the generator would be 273 substituted by a battery pack, which will be discussed further on. Keep in mind that in the remainder 274 of this paper, the impacts of the trailer’s assembly for both the bodyworks and the generator are 275 divided by 15, as the product is developed to be shared by the same number of users. 276

Table 5: LCI of the 40 kWh EV + generator ReX for the various impact categories, assuming the ReX is applied 277 for 5% of the time 278

Impact category Unit Total

Trailer

Assembly excl.

generator

Generator

manufacturing

Generator

operation

EV Electricity

(EU mix)

Climate change kg CO2-eq. 1,09E-01 2,12E-03 3,67E-03 1,27E-02 9,09E-02

Human toxicity kg 1,4-DB-eq. 6,34E-02 2,33E-03 9,94E-03 4,49E-04 5,07E-02

Photochemical oxidant

formation kg NMVOC 2,61E-04 1,03E-05 1,46E-05 4,37E-05 1,92E-04

Particulate matter

formation kg PM10-eq. 1,59E-04 7,38E-06 1,25E-05 8,88E-06 1,30E-04

279

280


4. Results 281

a. Benchmark exercise 282

The results of the benchmarking exercise which considers the different powertrain technologies 283 are presented graphically in Figure 3, Figure 4, Figure 5 and Figure 6. In the following paragraphs, 284 the impacts per midpoint indicators are discussed separately. 285

286

287

Figure 3: Impact per kilometre on climate change considering the entire LCA 288

Concerning the impact on climate change, a full life-cycle assessment was performed to consider 289 all contributing phases of the product. Notice how the absence of the TTW share for EV technologies 290 inherently causes an advantageous situation. Moreover, as the WTT emissions for EVs are based on 291 the European electricity mix (276gCO2/kWh), an indication is given for the GHG reduction potential 292 if smart charging mechanisms would allow EVs to recharge only when there is an abundance of solar 293 and/or wind energy, or in case of an energy production focussed on renewable sources. Once more, 294 EVs are considered to be part of the total load system and, therefore, aren’t deemed responsible for 295 marginal energy production of cheaper feedstocks such as coal. If the impact of the powertrain cycle 296 is considered, the larger the battery pack is designed, the higher its impact on climate change 297 becomes. This is also the case to a lesser extent for the hybrid technologies based on an ICE. The 40 298 kWh EV + generator ReX trailer combination is found to have a total impact which is less than 10% 299 bigger than for the 60 kWh EV, although it offers a bigger range. Compared to the 90 kWh EV, the 300 ReX combination allows a reduction of more than 40%, although the same range is available. This 301 indicates the environmental potential of the occasional use of a range-extender, instead of constantly 302 driving around with a substantial part of the battery capacity that is only rarely addressed. 303

PetrolEuro 6

DieselEuro 6

HybridEuro 6

Plug-inHybridEuro 6

EV 30kWh

EV 40kWh

EV 60kWh

EV 90kWh

EV 40kWh +

gen. ReX

Powertrain cycle 2.65E-03 2.65E-03 6.72E-03 1.24E-02 1.59E-02 2.00E-02 2.83E-02 4.08E-02 2.03E-02

Vehicle cycle 1.23E-02 1.28E-02 1.55E-02 1.76E-02 1.46E-02 1.35E-02 1.49E-02 1.96E-02 1.36E-02

TTW 1.63E-01 1.42E-01 1.35E-01 8.08E-02 0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.27E-02

RTT 3.80E-02 2.33E-02 2.95E-02 7.42E-02 4.17E-02 4.17E-02 4.17E-02 6.93E-02 4.44E-02

0.00E+00

5.00E-02

1.00E-01

1.50E-01

2.00E-01

2.50E-01

[kgC

O2

-eq

./km

]

Climate change


304

305

Figure 4: Impact per kilometre on photochemical oxidant formation during use of the vehicle (including the 306 refinery-to-tank (RTT) emissions upstream) 307

When photochemical oxidant formation is considered, the impacts from the manufacturing 308 stage are excluded. In addition, the system boundaries have been narrowed down to reflect only the 309 emissions from the fuel refinery onwards, assuming refinery takes place within the national borders. 310 This allows a comparison based on the emissions created locally. As can be seen in Figure 4, a 311 significant impact for diesel technology originates from the TTW phase. Whereas real driving 312 emissions (RDE) are considered, the struggle for car manufacturers to get a hold on NOx emissions is 313 emphasized. The influence of the electricity production process indicates the potential for further 314 reductions when renewable energy sources would be applied. In case of renewables, the impact of 315 EVs on POF can be virtually eliminated. The absence of local emissions makes EV technology very 316 suitable to counter poor air quality levels in e.g. urban regions. Due to the emissions generated during 317 the use phase of the generator ReX trailer, its POF contribution lies in between the hybrid and PHEV. 318 Keep in mind that the ReX trailer generator is petrol-driven, resulting in relatively low POF compared 319 to diesel technology. 320

PetrolEuro 6

DieselEuro 6

HybridEuro 6

Plug-inHybridEuro 6

EV 30 kWh EV 40 kWh EV 60 kWh EV 90 kWhEV 40 kWh+ gen. ReX

TTW 8.45E-05 5.95E-04 7.54E-05 4.52E-05 0.00E+00 0.00E+00 0.00E+00 0.00E+00 4.37E-05

RTT 1.38E-05 1.86E-05 1.23E-05 7.38E-06 2.75E-05 2.75E-05 2.75E-05 4.54E-05 2.87E-05

0.00E+00

1.00E-04

2.00E-04

3.00E-04

4.00E-04

5.00E-04

6.00E-04

7.00E-04

[kgN

MV

OC

/km

]Photochemical oxidant formation


321 Figure 5: Impact per kilometre on particulate matter formation during use of the vehicle (including the 322

refinery-to-tank (RTT) emissions upstream). Tank-to-wheel (TTW) emissions are further disaggregated to 323 exhaust TTW and non-exhaust PMF (NEx) 324

Concerning the influence of particulate matter formation on local air quality, the manufacturing 325 of the vehicles is again excluded. Non-exhaust PMF is added to the comparison, based on emission 326 factors used by Hooftman et al. in [39]. These are scaled for the different EV powertrains relative to 327 the vehicle weight. The contribution of brake wear for EVs is assumed to be one-third of a 328 conventional car’s, while road wear is multiplied by a factor 1,1 due to the extra weight of the vehicle 329 [39]. For the ReX trailer, brake wear was excluded as the trailer is assumed not to be provided with a 330 braking system. Road and tire wear for the trailer are assumed to be one fifth of the petrol car’s wear 331 factors. Again, diesel technology is found to have the largest TTW contribution, as it inherently 332 produces high levels of PM during its combustion process. Despite the successful application of 333 filtration systems, this impact remains significant compared to the other ICE technologies. 334 Considering the EV technologies, the impact of the energy production process (i.e. the RTT phase) 335 indicates the room for improvement in case of the application of renewables. Results for the EV+ReX 336 combination are in line with the 90 kWh EV’s, although the ReX concept is designed in the light of a 337 sharing economy, making it more likely that the trailer will be dropped off and picked up near 338 highways. Therefore, the impact of the EV+ReX combination is less relevant concerning local air 339 quality levels. 340

PetrolEuro 6

DieselEuro 6

HybridEuro 6

Plug-inHybridEuro 6

EV 30kWh

EV 40kWh

EV 60kWh

EV 90kWh

EV 40kWh +

gen. ReX

RTT 3.39E-06 7.82E-06 3.02E-06 1.81E-06 6.28E-06 6.28E-06 6.28E-06 1.02E-05 6.55E-06

Exhaust 1.50E-05 1.29E-04 1.34E-05 8.02E-06 0.00E+00 0.00E+00 0.00E+00 0.00E+00 8.88E-06

Non-exhaust 2.37E-05 2.37E-05 2.01E-05 2.01E-05 1.95E-05 1.78E-05 1.99E-05 2.70E-05 2.11E-05

0.00E+00

2.00E-05

4.00E-05

6.00E-05

8.00E-05

1.00E-04

1.20E-04

1.40E-04

1.60E-04

1.80E-04

[kgP

M1

0-e

q./

km]

Particulate matter formation


341

342 Figure 6: Impact per kilometre on human toxicity impacts during the vehicle’s manufacturing stage, for which 343

the powertrain cycle has been disaggregated 344

Finally, for the comparison of the impact on human toxicity, the focus is on the production stage 345 of the different technologies. Notice that the vehicle cycle mainly covers the impact of producing the 346 body shell and is linked to the vehicle mass. No disaggregation was done for the powertrain cycle of 347 the conventional petrol and diesel cars, whereas this is the case for the other technologies. What 348 catches the eye is the impact of producing the lithium battery, as a function of the battery’s capacity. 349 Notice the similar impact on human toxicity for manufacturing both the PHEV and the 30 kWh EV 350 and to a lesser extent the 40 kWh EV. Concerning HT, the ReX trailer combination is found to have a 351 similar impact as the 40 kWh EV itself, indicating the trailer itself barely has an influence. 352

b. Breakeven points 353

In the following figures, the calculated BEPs are presented for either replacing existing ICE-354 based technology with an EV, or for comparing a new ICE car with the given EVs. The EV+ReX 355 combination is added to this comparison to show the environmental breakeven distances with the 356 ICE-cars. The difference with the ‘replacement’ scenario is that no manufacturing impacts are 357 considered for the respective, existing ICE car. By means of the intersections with the impact curves 358 of the EV cars, the driven distance is known by which the latter becomes a more environmentally 359 solution. End-of-life impacts, which are in fact negatively adding to the total impact, are included in 360 the totals. For the ease of comparing, both the breakeven distances are discussed as well as their 361 translation to the number of years of driving the respective car. Therefore, an average annual 362 kilometrage of 15.000 km (15k) is assumed. Given the widening gap of CO2 emissions for 363 conventional cars between laboratory results and real-world driving data over the last two decades 364 [48], [49], fuel consumption for both old and new ICE-cars is kept the same. 365

PetrolEuro 6

DieselEuro 6

HybridEuro 6

Plug-inHybridEuro 6

EV 30kWh

EV 40kWh

EV 60kWh

EV 90kWh

EV 40kWh +

gen. ReX

Generator 6.62E-04

Onboard charger 0.00E+00 1.97E-03 1.97E-03 1.97E-03 1.97E-03 1.97E-03 1.97E-03

DC/DC converter 2.31E-03 2.31E-03 2.31E-03 2.31E-03 2.31E-03 2.31E-03 2.31E-03

AC/DC converter 6.01E-03 6.12E-03 6.12E-03 6.12E-03 6.12E-03 6.12E-03 6.12E-03

Electric Motor 5.72E-03 5.72E-03 5.72E-03 5.72E-03 5.72E-03 5.72E-03 5.72E-03

Li battery 1.10E-03 4.38E-03 8.22E-03 1.10E-02 1.64E-02 2.47E-02 1.10E-02

Powertrain cycle 7.06E-03 7.06E-03

Vehicle cycle 1.28E-02 1.33E-02 1.51E-02 1.72E-02 1.43E-02 1.31E-02 1.46E-02 1.94E-02 1.45E-02

0.00E+00

1.00E-02

2.00E-02

3.00E-02

4.00E-02

5.00E-02

6.00E-02

7.00E-02

[kg

1,4

-DB

-eq

./km

]Human Toxicity


366

Figure 7: The impact on climate change of either replacing an existing petrol car by EV technology (solid grey 367 line) and buying a new car (dashed grey line). For the latter situation, the manufacturing impact of the petrol 368

car is included 369

The comparison of a petrol car versus the different EV options is shown in Figure 7. Notice the 370 offset of roughly 2,5 tonnes of CO2 for manufacturing the new car at kilometre zero. When an existing 371 petrol car is to be replaced, it becomes clear that a new petrol variant as a substitute isn’t the best 372 choice, concerning climate change. Given that the CO2 impact increases with the battery capacity, 373 both replacing an existing car as buying a new one reaches a BEP first with the 30 kWh EV, i.e. after 374 38k and 21k km, respectively. In terms of year, this means the 30 kWh EV becomes a more 375 environmentally solution after 2,5 years and 1,5 years, respectively. 376

Considering the ReX trailer combination with the 40 kWh EV, a replacement of an existing petrol 377 car becomes interesting after roughly 46k km, while a new purchase would pay off environmentally 378 after roughly 26k km, or approximately 2 years of driving. Whereas the 40 kWh EV + ReX 379 combination was found to emit roughly 10% more CO2 emissions than the 60 kWh EV, the latter 380 becomes the better solution after nearly 130k km, or nearly 9 years of driving. Such a BEP simply isn’t 381 found for the trailer combination and the 90 kWh EV. Concerning the comparison between the 90 382 kWh EV and a new petrol car, the former outruns the latter after approximately 72,5k km or 5 years 383 of driving. For a detailed overview, head to Table 12 in the appendix. 384

385

0

5 000

10 000

15 000

20 000

25 000

30 000

35 000

40 000

45 000

50 000

0 10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

110000

120000

130000

140000

150000

160000

170000

180000

190000

200000

210000

Clim

ate

chan

ge [

kg C

O2-

eq.]

Driven distance [km]

Breakeven points for climate change - petrol vs. EVsReplacement vs. new purchase

Petrol Euro 6 (existing) Petrol Euro 6 (new) EV 30 kWh (new)

EV 40 kWh (new) EV 60 kWh (new) EV 90 kWh (new)

EV 40 kWh (new) + gen. ReX


386

Figure 8: The impact on climate change of either replacing an existing diesel car by EV technology (solid grey 387 line) and buying a new car (dashed grey line). 388

The same comparison is made for replacing a diesel car and choosing between a new diesel car 389 and new EVs according to their battery capacity. Compared to the petrol car’s situation, the presented 390 diesel technology is characterized by a better fuel economy, which is translated in more horizontal 391 slope in Figure 8. Also here, a diesel for a diesel makes no sense when the impact on climate change 392 is considered, indicated by the fact that no BEP is reached within a reasonable time/kilometrage. 393 Notice that for the newly produced diesel car, environmental breakeven with the 30 kWh occurs 394 around 26k km, compared to 21k km for the petrol car shown in Figure 7. Concerning the 90 kWh 395 EV, an intersection is reached only after 98k km and 128k km for the new and the replacement 396 scenarios, which can be translated to 6,5 years and over 8,5 years of driving a diesel car, respectively. 397 The 40 kWh EV + ReX trailer combination becomes interesting after 33k km and 67k km when a new 398 purchase and a replacement are considered, respectively. In terms of the average number of years 399 driving a passenger car, this translates to roughly 2,5 years and 4,5 years, respectively. 400

401

0

5 000

10 000

15 000

20 000

25 000

30 000

35 000

40 000

0 10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

110000

120000

130000

140000

150000

160000

170000

180000

190000

200000

210000

Clim

ate

chan

ge [

kg C

O2-

eq.]


Breakeven points for climate change - diesel vs. EVsReplacement vs. new purchase

Diesel Euro 6 (existing) EV 30 kWh (new) EV 40 kWh (new)

EV 60 kWh (new) EV 90 kWh (new) EV 40 kWh (new) + gen. ReX

Diesel Euro 6 (new)


402

Figure 9: The impact on climate change of either replacing an existing petrol hybrid car by EV technology 403 (solid grey line) and buying a new car (dashed grey line). 404

When petrol hybrid technology is compared to the presented EV technologies, its low fuel 405 consumption translates in longer distances/years of driving the petrol hybrid before its impact on 406 climate change exceeds that of the various EV technologies. Therefore, the replacement of an existing 407 hybrid car is even less prioritized as this was the case for the petrol or diesel car. Due to the higher 408 impact of the hybrid’s manufacturing stage, the comparison for the new purchase looks entirely 409 different. In this case, it takes only 15k km before the 30 kWh EV becomes more interesting 410 considering the impact on climate change. This means that in case of buying a new car, EV technology 411 could pay off environmentally in approx. one year. The ReX combination performs better than a new 412 hybrid after 23k km, or roughly 1,5 years. As for the 90 kWh EV, over 85 km are to be performed 413 before a BEP with a new hybrid is reached. This indicates once more the extent of the battery pack’s 414 influence. 415

416

0

5 000

10 000

15 000

20 000

25 000

30 000

35 000

40 000

45 000

0 10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

110000

120000

130000

140000

150000

160000

170000

180000

190000

200000

210000

Clim

ate

chan

ge [

kg C

O2-

eq.]


Breakeven points for climate change - hybrids vs. EVsReplacement vs. new purchase

Hybrid Euro 6 (existing) EV 30 kWh (new) EV 40 kWh (new)


Hybrid Euro 6 (new)


417

Figure 10: The impact on climate change of either replacing an existing petrol plug-in hybrid car by an EV 418 (solid grey line) and buying a new car (dashed grey line) 419

Finally, petrol-fuelled PHEV technology is compared with the presented EV technologies. What 420 catches the eye is that due to the PHEV’s manufacturing impact on climate change, the 30 kWh EV 421 becomes the better option as early as after 1,6k km. This is in contrast with the replacement scenario, 422 where the same EV becomes interesting climate change-wise after roughly 54k km or after nearly 4 423 years of driving. For the ReX trailer combination, a replacement and a new purchase become 424 interesting after respectively 74k km and 6k km. Replacing an existing PHEV with a 90 kWh EV is no 425 viable solution if a positive impact on climate change is envisioned, as a BEP occurs only after 143k 426 km. 427

428 What can be concluded from the BEPs for climate change is that replacing an existing ICE-based 429

car with a new one makes no sense as no intersections of the impact graphs are reached within a 430 reasonable time or kilometrage. Replacing a petrol or diesel car with an EV becomes interesting 431 starting from roughly two and a half to three years of driving, respectively. When the manufacturing 432 impacts are included, hybrids are surpassed by EVs after one year, if 15.000 km per year on average 433 are assumed. For PHEVs, this BEP is reached nearly from the start. As conventional (non-hybridized) 434 ICE vehicles are assumed to phase-out on the short-term, given the strict pollutant and CO2 emission 435 targets imposed by the EU, EV technology clearly becomes a competitive opponent for hybrid 436 technologies. The inherent advantage for EVs when it comes to local emissions only adds to this 437 statement. 438

439 440 441 442 443 444

0

5 000

10 000

15 000

20 000

25 000

30 000

35 000

40 000

45 000

0 10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

110000

120000

130000

140000

150000

160000

170000

180000

190000

200000

210000

Clim

ate

chan

ge [

kg C

O2-

eq.]


Breakeven points for climate change - PHEV vs. EVsReplacement vs. new purchase

PHEV Euro 6 (existing) EV 30 kWh (new) EV 40 kWh (new)


PHEV Euro 6 (new)


5. Discussion 445

a. Sensitivity analysis 446

Whereas the results presented in this paper are based on certain assumptions, a sensitivity 447 analysis is recommendable to highlight possible weaknesses and/or potentials for the ReX trailer, 448 combined to an EV. An overview is given of examples which are thought to have an influence on the 449 results of this exercise. 450

i. Energy consumption 451

The impact of the energy consumption (kWh/km) of an EV is noticeable in the WTT contribution 452 to any of the given midpoint indicator. As R&D is focussed on light-weight battery chemistries which 453 offer higher energy densities, it is likely that the overall energy consumption will slightly reduce in 454 the near-future. The total influence isn’t expected to be substantial unless the paradigm of heavy and 455 powerful cars is surpassed. Given the current occupation rate for passenger cars (approx. 1,1 persons 456 per car), a shift towards e.g. two-seater vehicles in combination with light-weight battery packs could 457 result in decreasing energy consumption levels. 458

ii. The electricity production mix 459

For the comparisons made in this paper, the EU electricity mix was applied, representing a 276g 460 of CO2 produced per kWh. If by example the Belgian mix would be used, this would be reduced to 461 approx. 210g/kWh [35]. As EVs are only exploited to their full environmental potential when its 462 electricity originates from renewable sources, this is where the focus of European policy concerning 463 energy production should be on. This is emphasized by the agreed commitments towards carbon-464 neutral economy by 2050. As indicated by Messagie et al, the climate change impact of EVs powered 465 by energy from renewable sources could be reduced to approx. 40g of CO2-equivalents per kilometre 466 driven. The other way around; if energy production would be based on fossil fuels like oil or coal, 467 the impact per kilometre would significantly exceed the impact of conventional cars. This is shown 468 in Figure 11, for which the 40 kWh EV’s impact is distinguished per energy production source (based 469 on [36]). A drastic shift to renewable sources would thus mean the WTT emissions for EVs could be 470 marginalized. This will also have a substantial impact on the other three discussed midpoint 471 indicators. Nuclear power isn’t considered a renewable source, given that the current generation of 472 nuclear power plants requires uranium, which is a finite element and hence not renewable. 473

474

475

Figure 11: Impact of the energy production mix in grams per kilometre (based on [36]) 476

PetrolEuro 6

DieselEuro 6

HybridEuro 6

Plug-inHybridEuro 6

EV 40kWh

Nuclear

EV 40kWhWind

EV 40kWh

Hydro

EV 40kWh EU

Mix

EV 40kWhGas

EV 40kWh Oil

EV 40kWhCoal

Powertrain cycle 2.65 2.65 6.72 12.35 20.04 20.04 20.04 20.04 20.04 20.04 20.04

Vehicle cycle 12.33 12.81 15.54 17.61 13.46 13.46 13.46 13.46 13.46 13.46 13.46

TTW 163.19 142.29 134.89 80.82 0.00 0.00 0.00 0.00 0.00 0.00 0.00

WTT 38.04 23.26 29.49 74.20 2.37 3.48 9.98 41.69 159.00 276.67 325.00

0.00

50.00

100.00

150.00

200.00

250.00

300.00

350.00

400.00

[gC

O2-

eq./

km]

Climate change


In the current absence of large-scale power storage facilities for renewable sources such as wind 477 and solar power, EVs might prove to be a solution as well. Slow-charging facilities could, therefore, 478 be maximally provided with green power, although this is far less likely for fast-charging facilities. 479 In case of the latter, energy is needed immediately at high power rates, which is why the share of 480 renewable energy is likely to be on the low side. 481

482

iii. The use pattern for the ReX trailer 483

An important aspect of this sensitivity analysis is the marginal application factor of the ReX 484 trailer. If the ReX trailer isn’t applied for 5% of the annually driven kilometres, but e.g. for 10%, this 485 can have a significant influence on the comparison of the different midpoint indicators. For climate 486 change, applying the ReX trailer for 10% of the time would result in a relative increase by 17%. 487

Another aspect of the use pattern is the intrinsic issue of people tending to cover longer trips 488 during weekends, e.g. to visit family. This can result in an unevenly balanced demand for the ReX 489 concept and thus, at peak demand in a lower number of users for one trailer. This sensitivity has been 490 calculated for 5 users and 10 users concerning the impact on climate change. Results showed no 491 significant impact on either the vehicle cycle nor the powertrain cycle, as the increase was found to 492 remain below 2,5% in case of 5 users. 493

494

b. Generator vs. battery 495

As a side-track, the case is investigated for which the ReX trailer’s petrol generator would be 496 substituted by a battery pack. The question here is to determine which capacity would then be needed 497 to offer at least the same additional range and which impact this would have on climate change. 498 Therefore, a system level energy density of 140 Wh/kg [50] and trailer body weight of 130 kg 499 (excluding the battery) is assumed, to assess the influence of the total battery trailer on the energy 500 consumption of the 40 kWh EV. This is plotted as a function of the trailer’s capacity in Figure 12. The 501 additional power consumption during highway driving, relative to the installed trailer capacity, is 502 assumed to increase linearly from +5% for 20 kWh of trailer capacity to +15% for 100 kWh. The 503 argumentation behind this increase is that during motorway driving, vehicles mainly require extra 504 power to overcome the forces on the frontal area. As the trailer is likely to improve the EV’s drag 505 coefficient (CX) as it improves the aerodynamics behind the EV, the overall impact of the trailer is 506 reduced to the rolling resistance of its wheels. Moreover, this extra power consumption is 507 marginalised by means of the assumption that the trailer is only connected to the EV for 5% of the 508 time. A 50 kWh battery capacity was arbitrarily chosen as the reference for the battery trailer concept, 509 characterized by a total weight of 480 kg. For the 40 kWh EV towing this trailer, a highway energy 510 consumption of 0,22 kWh/km was determined. Table 6 highlights the impact on climate change per 511 kilometre driven for both the generator-based ReX trailer and the battery-based ReX trailer. 512

513


514

Figure 12: Impact of the battery capacity on the total trailer weight (blue) and the average consumption of the 515 40 kWh EV towing the trailer (orange) 516

Table 6: LCI of the 40 kWh EV with either the generator or the battery trailer 517

kg CO2-eq./km WTT TTW Vehicle cycle Powertrain cycle Total

40 kWh EV + Generator trailer

4,44E-02 1,27E-02 1,36E-02 2,03E-02 9,10E-02

40 kWh EV + 50 kWh trailer

4,26E-02 0,00E+00 1,36E-02 2,12E-02 7,75E-02

518 Considering the BEPs for climate change, the battery trailer proves to allow shorter breakeven 519

distances compared to the generator trailer. This can be seen in Table 7. Both trailer concepts are 520 compared to the default 60 kWh and 90 kWh EVs. Whereas for the generator trailer, the 60 kWh EV 521 becomes environmentally beneficial after roughly six years of driving, no viable BEP was found with 522 the battery trailer. The latter indicates that the default 60 kWh EV will never be a more 523 environmentally solution. Concerning the 90 kWh EV, both trailer concepts prove to be the better 524 solution from the start onwards as well. 525

Table 7: Breakeven points in kilometres driven for the different trailer topologies according to the impact on 526 climate change 527

[km] 40 kWh EV

+ Generator trailer 40 kWh EV

+ 50 kWh trailer

Replacement New Replacement New

Petrol 46.639 27.718 43.633 26.133

Diesel 61.985 35.920 56.297 32.894

Hybrid 62.663 22.996 56.839 21.503

PHEV 68.654 8.661 61.572 8.922

60 kWh EV - 87.220 - 1.486.517

90 kWh EV - Negative result: No

BEP -

Negative result: No BEP

528 Finally, Figure 13 shows the emission factors for climate change in grams of CO2 per kilometre. 529

The 40 kWh EV is the default vehicle for the combination with either the ReX trailer generator trailer 530 or the battery trailer. Both the original 60 kWh EV and 90 kWh EV are added to the comparison. No 531 combined capacities higher than 100 kWh are considered, as these are unlikely to become mainstream 532

2.00E-01

2.05E-01

2.10E-01

2.15E-01

2.20E-01

2.25E-01

2.30E-01

2.35E-01

0.00E+00

1.50E+02

3.00E+02

4.50E+02

6.00E+02

7.50E+02

9.00E+02

20

24

28

32

36

40

44

48

52

56

60

64

68

72

76

80

84

88

92

96

10

0

[kWh/km][kg]

Trailer battery capacity [kWh]

Trailer weight and energy consumption as a function of the installed trailer capacity [kWh]

Trailer weight [kg] EV consumption [kWh/km]


in the EV car market. Two situations are chosen to highlight the excessive CO2 emissions per 533 kilometre as a function of the over-capacity on board in the default EVs. 534

In situation A, the 40 kWh EV is combined with a 20 kWh trailer range-extender for 5 percent of 535 the time. Thus, the same highway range is obtained as the 60 kWh EV’s. The difference with the 536 default 60 kWh EV is pronounced in a 111% higher emission factor. In case of situation B, the 537 combined capacity of both the 40 kWh EV + generator ReX as the combination with the 50 kWh 538 battery trailer equals the default 90 kWh EV’s capacity, as well as its highway range. Here, the 539 difference with the original 90 kWh EV is substantial, as a 143% and 167% higher emission factor was 540 found for the generator and battery trailer, respectively. Compared to the 60 kWh EV, the battery 541 trailer combination emits nearly 10% less CO2. 542

543

Figure 13: An overview of the CO2 emissions per kilometre for the different EV technologies and the trailer 544 combinations 545

6. Conclusions 546

Electric vehicles are characterized by a limited on-board energy content. Assuming EV users 547 have the possibilities to recharge their vehicles daily, the noticeable trends towards battery capacities 548 exceeding 40 kWh – or an average range of 300 km – creates a substantial share of this capacity that 549 remains unused except for rare occasions. As this marginally applied capacity share comes at a 550 significant price, both for the consumer and the environment, in-life modularity by means of an 551 external range-extender concept is discussed. This modularity could be realised by means of a trailer, 552 equipped with either a petrol generator or an additional battery pack. 553

A life-cycle assessment is performed to determine the potential of the trailer concept, when 554 connected to a 40 kWh EV. This combination is benchmarked to a range of both conventional and 555 electric cars, for their respective impact per kilometre driven. Concerning the impacts on local air 556 quality and human health, the LCA’s system boundaries are narrowed down to exclude the emissions 557 which occurred outside of the national borders, e.g. during material extraction, as these have no 558 influence locally. Therefore, well-to-tank (WTT) emissions are cropped to refinery-to-tank (RTT) 559 emissions, assuming refining within the national borders. In addition to the LCA approach, 560 environmental breakeven points (BEP) were determined for both replacing an existing internal 561 combustion engine vehicle (ICEV) with an EV, and for the most environmentally solution in case of 562 a one-on-one comparison between both ICEVs and EVs. In the latter case, both the manufacturing 563 and use phase of the vehicle are included, whereas no manufacturing impacts were considered in 564 case of the replacement scenario. 565

76.684.9

77.5

91.0

129.7

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

EV40 + 20 kWhTrailer

EV60 EV40 + 50 kWhTrailer

EV40 + ICE Trailer EV90

[gCO2/km]

CO2 emissions factors

A: Highway range 300 km

B: Highway range ~500 km


Results confirm the findings from literature [36], [39], [51], stating that diesel powertrains are 566 the worst solution for local air quality, due to high contributions to photochemical oxidant formation 567 (POF) and particulate matter formation (PMF). Due to the absence of tank-to-wheel (TTW) emissions, 568 EVs are found most suitable for densely trafficked and populated areas such as cities. Addressing the 569 limited range of EVs by means of a generator range-extender positively impacts the contribution to 570 climate change, as for the same combined range as the 90 kWh, the 40 kWh EV + generator ReX emits 571 nearly 30% less CO2, while the battery ReX emits 40% less CO2. Although the ReX trailer is rather 572 designed for use over long distances and thus highway travelling, its impact on local air quality is of 573 importance as well. Compared to the 40 kWh EV alone, the ReX combination adds 3 times more to 574 POF, while PMF contributions increase by 150%. Compared to the default 90 kWh EV, POF emissions 575 reduce by nearly 25% while PMF contributions are reduced by nearly 40%. The contributions by the 576 various EVs to climate change, PMF and POF are mostly determined by the energy production mix, 577 during the Well-To-Tank stage (WTT). Power consumption by EVs is assumed to be part of the total 578 load system, which is why the average carbon intensity of the European energy mix is used. In this 579 point of view, WTT emissions by EVs can be substantially reduced as the European energy 580 production shifts towards renewable sources. 581

582 Concerning breakeven points for climate change, the 40 kWh EV becomes a more 583

environmentally replacement for an existing ICEV car after roughly 3 years of driving, if an annual 584 kilometrage of 15.000 km is assumed. The same is true in case of the 40 kWh EV with the generator 585 ReX. When the manufacturing phase is included for both vehicles in the comparison, the pace at 586 which a BEP is reached is related to the hybridization rate of the conventional car. This makes the 587 breakeven distance between a PHEV and e.g. the 40 kWh EV exceptionally short, i.e. about 7.000 km 588 or less than half a year of driving. Due to the high manufacturing impact for the 90 kWh EV, a BEP 589 in comparison with other new ICEVs is reached typically after approx. 5 years of driving. For the 40 590 kWh EV + ReX trailer combination this ranges from half a year to roughly two years. Replacing an 591 ICEV with an ICEV, no matter which technology, proves not to be advantageous when climate 592 change is concerned. 593

594 Concluding, the objective of this paper was to investigate the environmental feasibility of in-life 595

range modularity, as an alternative for high-capacity battery electric vehicles. Results show the 596 benefit of relying on a range-extension only when needed, both for contributions to climate change 597 as to local air quality. 598

Acknowledgements 599

This research was made possible by means of funding by EP Tender. We acknowledge Flanders 600 Make for the support to our research group. 601

602 603


Appendix 604

Table 8: LCI of a 30 kWh EV for the European electricity mix (276 gCO2/kWh) (based on [36][52]) 605

Midpoint

indicator

WTT TTW Vehicle cycle Powertrain cycle

WTT

Public

charging

station

Tire

abrasion

Road

abrasion

Brake

abrasion TTW

Body

shell

Lead

battery

Mainten

ance

Li

battery

Electric

Motor

AC/DC

converter

DC/DC

converter

On-

board

charger

Catalytic

converter

Starter

and

generator

Engine

Control

Unit

CC

[kgCO2/km] 4,14E-02 2,95E-04 0,00E+00 0,00E+00 0,00E+00 0,00E+00 1,30E-02 6,29E-05 1,52E-03 1,24E-02 1,19E-03 1,35E-03 5,08E-04 4,14E-04 0,00E+00 0,00E+00 0,00E+00

POF

[kgNMVOC/km] 2,68E-05 7,42E-07 0,00E+00 0,00E+00 0,00E+00 0,00E+00 4,08E-05 2,56E-07 4,94E-06 4,54E-05 4,68E-06 5,66E-06 1,99E-06 1,98E-06 0,00E+00 0,00E+00 0,00E+00

PMF

[kgPM10/km] 5,83E-06 4,51E-07 7,05E-06 1,00E-05 2,46E-06 0,00E+00 3,12E-05 1,91E-07 2,46E-06 3,16E-05 4,77E-06 3,21E-06 1,19E-06 1,07E-06 0,00E+00 0,00E+00 0,00E+00

HT

[kg1,4-DB/km] 2,03E-04 4,43E-04 6,37E-04 4,01E-06 7,35E-04 0,00E+00 1,34E-02 3,03E-04 5,86E-04 8,22E-03 5,72E-03 6,12E-03 2,31E-03 1,97E-03 0,00E+00 0,00E+00 0,00E+00

606

607


Sustainability 2017, 9, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability

Table 9: Overview of the LCI of the generator ReX trailer’s manufacturing phase 608

ReX Trailer CC HT POF PMF

kg CO2 eq kg 1,4-DB eq kg NMVOC kg PM10 eq

Chassis 6,83E+01 7,22E+01 2,47E-01 2,45E-01

Jockey Wheel 4,84E+00 2,78E+00 1,99E-02 1,27E-02

Sec. Wheel Ass. 6,81E+01 5,06E+01 2,46E-01 2,13E-01

Main Wheel Ass. 2,32E+02 1,02E+02 8,76E-01 6,01E-01

Shock Absorber 3,43E+01 2,80E+01 1,29E-01 1,44E-01

Sundries 1,16E+01 6,61E+00 4,33E-02 6,55E-02

12V Motor 1,23E+01 5,74E+01 6,60E-02 6,65E-02

Junction Box 3,63E+01 2,02E+02 1,98E-01 2,00E-01

Bodywork 1,73E+02 8,93E+01 8,94E-01 5,62E-01

Cat. Converter 1,00E+01 2,12E+01 6,35E-02 8,48E-02

Generator 7,69E+02 2,08E+03 3,06E+00 2,61E+00

Total 6,50E+02 6,33E+02 2,78E+00 2,20E+00

609

610

Table 10: Impact of the manufacturing stage of the generator 611

GENERATOR CC HT POF PMF

kg CO2 eq kg 1,4-DB eq kg NMVOC kg PM10 eq

Metal working, aluminium 1,57E+02 5,69E+01 4,50E-01 4,49E-01

Copper 1,99E+01 4,63E+02 2,17E-01 2,75E-01

Metal working, steel 7,84E+01 4,69E+01 2,33E-01 2,32E-01

Electronic control unit 2,67E+01 1,69E+02 1,22E-01 9,74E-02

Aluminium 1,65E+02 8,83E+01 5,73E-01 4,86E-01

Polyethylene 1,93E+01 2,12E-01 8,64E-02 1,98E-02

Steel 7,74E+01 1,12E+02 3,38E-01 3,23E-01

Batteries 1,42E+01 1,13E+02 1,08E-01 8,72E-02

Polybutadiene 9,97E+00 1,55E-01 4,76E-02 1,38E-02

Aluminium alloy, AlMg3 3,91E+01 2,08E+01 1,79E-01 1,29E-01

Recycling 1,63E+02 1,01E+03 7,10E-01 4,98E-01

Total 7,69E+02 2,08E+03 3,06E+00 2,61E+00

612

613


Table 11: Overview of the manufacturing impact and the total impact per midpoint indicator and technology. 614 The difference between the manufacturing stage and the total is the impact of the use stage. 615

Technology Impact Climate Change

Photochemical Oxidant

Formation

Particulate Matter

Formation

Human Toxicity

kg CO2-eq. kg NMVOC kg PM10-eq. kg 1,4-DB-eq.

Petrol Manufacturing 2.756 11 10 4.008

Total 45.290 33 20 4.747

Diesel Manufacturing 2.856 11 11 4.120

Total 37.916 141 45 4.855

Hybrid Manufacturing 4.281 16 14 7.298

Total 39.095 36 23 8.036

Plug-in Hybrid Manufacturing 5.895 22 18 8.688

Total 38.749 35 25 9.602

30 kWh Manufacturing 6.073 21 15 7.972

Total 15.125 28 21 8.518


Total 15.751 30 23 8.818


Total 17.786 38 28 10.304


Total 27.161 54 39 13.177

40 kWh + ReX trailer Manufacturing 6.704 24 17 8.688

Total 19.052 41 25 9.326

616

617

Table 12: Overview of the BEPs considering the impact on climate change 618

Breakeven distance Climate Change [km]

30 kWh 40 kWh 60 kWh 90 kWh 40 kWh +

ReX trailer

Total impact

[kg]

Petrol Replacement 38.092 42.019 54.783 93.457 46.639 42.534

New 20.804 24.731 37.494 72.560 27.718 45.290

Diesel Replacement 49.038 54.093 70.524 127.986 61.985 35.060

New 25.980 31.035 47.466 98.337 35.920 37.916

Hybrid Replacement 49.505 54.609 71.197 129.560 62.663 34.815

New 14.614 19.717 36.305 84.573 22.996 39.095

PHEV Replacement 53.583 59.106 77.060 143.653 68.654 32.854

New 1.575 7.098 25.052 74.961 8.661 38.749

619 620


References 621

[1] EEA, “Air quality in Europe — 2016 report,” Copenhagen, 2016. 622

[2] Dieselnet, “EU: Cars and light truck,” 2016. [Online]. Available: 623

https://www.dieselnet.com/standards/eu/ld.php. 624

[3] EEA, “EEA greenhouse gas - Data viewer,” 2016. [Online]. Available: http://www.eea.europa.eu/data-625

and-maps/data/data-viewers/greenhouse-gases-viewer. 626

[4] DG Clima, “Revision of Regulation (EU) No 443/2009 and Regulation (EU) No 510/2011 regulating CO2 627

emissions from light duty vehicles,” Brussels, 2016. 628

[5] S. Díaz, U. Tietge, and P. Mock, “CO2 emissions from new passenger cars in the EU: Car manufacturers’ 629

performance in 2015,” Berlin, 2015. 630

[6] A. Rusich and R. Danielis, “Total cost of ownership, social lifecycle cost and energy consumption of 631

various automotive technologies in Italy,” Res. Transp. Econ., vol. 50, pp. 3–16, Aug. 2015. 632

[7] J. Massiani, “Cost-Benefit Analysis of policies for the development of electric vehicles in Germany: 633

Methods and results,” Transp. Policy, vol. 38, pp. 19–26, Feb. 2015. 634

[8] “Report of the Conference of the Parties on its twenty-first session, held in Paris from 30 November to 635

13 December 2015 Addendum Part two: Action taken by the Conference of the Parties at its twenty-first 636

session,” in 2015 United Nations Climate Change Conference, 2015. 637

[9] J. Rockström, H. J. Schellnhuber, B. Hoskins, V. Ramanathan, P. Schlosser, G. P. Brasseur, O. Gaffney, C. 638

Nobre, M. Meinshausen, J. Rogelj, and W. Lucht, “The world’s biggest gamble,” Earth’s Futur., vol. 4, no. 639

10, pp. 465–470, Oct. 2016. 640

[10] EC, “Roadmap to a Single European Transport Area - Towards a competitive and resource efficient 641

transport system,” Brussels, 2011. 642

[11] P. McGee, “Volkswagen plans to ‘leapfrog’ Tesla in electric car race,” 2017. [Online]. Available: 643

https://www.ft.com/content/a43ac2ce-3198-11e7-9555-23ef563ecf9a. [Accessed: 09-Jul-2017]. 644

[12] N. Pollard, “Geely’s Volvo to go all electric with new models from 2019 | Reuters,” 2017. [Online]. 645

Available: http://www.reuters.com/article/us-volvocars-geely-electric-idUSKBN19Q0BJ?il=0. 646

[Accessed: 09-Jul-2017]. 647

[13] Reuters, “Mercedes-Benz Speeds Up Electric Car Plans | Fortune.com,” 2017. [Online]. Available: 648

http://fortune.com/2017/03/29/mercedes-benz-electric-emissions/. [Accessed: 09-Jul-2017]. 649

[14] D. Fickling, “Renault-Nissan’s Electric Lead Vanishes - Bloomberg Gadfly,” 2017. [Online]. Available: 650

https://www.bloomberg.com/gadfly/articles/2017-04-24/renault-nissan-needs-to-turn-over-a-new-leaf. 651

[Accessed: 09-Jul-2017]. 652

[15] F. Lambert, “BMW announces production expansion for its upcoming electric vehicles | Electrek,” 2017. 653

[Online]. Available: https://electrek.co/2017/05/02/bmw-electric-vehicle-production-expansion/. 654

[Accessed: 09-Jul-2017]. 655

[16] F. Lambert, “GM says it will work with PSA on electric vehicles, but no word on Bolt EV/Ampera E | 656

Electrek,” 2017. [Online]. Available: https://electrek.co/2017/03/06/gm-psa-electric-vehicles-bolt-ev-657

ampera-e/. [Accessed: 09-Jul-2017]. 658

[17] EAFO, “EAFO | European Alternative Fuels Observatory,” 2017. [Online]. Available: 659

http://www.eafo.eu/. [Accessed: 24-Apr-2017]. 660

[18] H. He and L. Yang, “China’s Stage 6 emisison standard for new light-duty vehicles (final rule),” Berlin, 661

2017. 662

[19] N. Shirouzu, “China likely to unveil NEV car quotas soon, but delay implementation: sources - NY Daily 663


News,” 2017. [Online]. Available: http://www.nydailynews.com/newswires/news/business/china-664

unveil-nev-car-quotas-delay-implementation-sources-article-1.3454664. [Accessed: 31-Aug-2017]. 665

[20] M. Martina and N. Shirouzu, “China Upholds Strict Electric Car Sales Quotas Despite Industry Protests 666

| Technology News | US News,” 2017. [Online]. Available: 667

https://www.usnews.com/news/technology/articles/2017-06-13/china-upholds-strict-electric-car-sales-668

quotas-despite-industry-protests. [Accessed: 31-Aug-2017]. 669

[21] P. Slowik and N. Lutsey, “EXPANDING THE ELECTRIC VEHICLE MARKET IN U.S. CITIES,” 2017. 670

[22] P. Hummel, D. Lesne, J. Radlinger, C. Golbaz, C. Langan, K. Takahashi, D. Mulholland, A. Stott, G. 671

Haire, M. Mittermaier, N. Gaudois, and L. Shaw, “UBS Evidence Lab Electric Car Teardown – Disruption 672

Ahead? 1,” 2017. 673

[23] N. S. Pearre, W. Kempton, R. L. Guensler, and V. V. Elango, “Electric vehicles: How much range is 674

required for a day’s driving?,” Transp. Res. Part C Emerg. Technol., vol. 19, no. 6, pp. 1171–1184, Dec. 2011. 675

[24] Z. A. Needell, J. McNerney, M. T. Chang, and J. E. Trancik, “Potential for widespread electrification of 676

personal vehicle travel in the United States,” Nat. Energy, vol. 1, no. 9, p. 16112, 2016. 677

[25] G. Pasaoglu, D. Fiorello, A. Martino, G. Scarcella, A. Alemanno, C. Zubaryeva, C. Thiel, and European 678

Commission. Joint Research Centre. Institute for Energy and Transport., “Driving and parking patterns 679

of European car drivers : a mobility survey.,” European Commission, Sevilla, 2012. 680

[26] C. Corchero, S. Gonzalez-Villafranca, and M. Sanmarti, “European electric vehicle fleet: driving and 681

charging data analysis,” in 2014 IEEE International Electric Vehicle Conference (IEVC), 2014, pp. 1–6. 682

[27] M. Khan and K. M. Kockelman, “Predicting the market potential of plug-in electric vehicles using 683

multiday GPS data,” Energy Policy, vol. 46, pp. 225–233, Jul. 2012. 684

[28] J. Gonder, T. Markel, M. Thornton, and A. Simpson, “Using Global Positioning System Travel Data to 685

Assess Real-World Energy Use of Plug-In Hybrid Electric Vehicles,” Transp. Res. Rec. J. Transp. Res. Board, 686

vol. 2017, pp. 26–32, Dec. 2007. 687

[29] M. Redelbach, E. D. Özdemir, and H. E. Friedrich, “Optimizing battery sizes of plug-in hybrid and 688

extended range electric vehicles for different user types,” Energy Policy, vol. 73, pp. 158–168, Oct. 2014. 689

[30] A. Hirtenstein, “Move Over Tesla, Europe’s Building Its Own Battery Gigafactories - Bloomberg,” 2017. 690

[Online]. Available: https://www.bloomberg.com/news/articles/2017-05-22/move-over-tesla-europe-s-691

building-its-own-battery-gigafactories. [Accessed: 27-Jun-2017]. 692

[31] A. Mahmoudzadeh Andwari, A. Pesiridis, S. Rajoo, R. Martinez-Botas, and V. Esfahanian, “A review of 693

Battery Electric Vehicle technology and readiness levels,” Renew. Sustain. Energy Rev., vol. 78, pp. 414–694

430, Oct. 2017. 695

[32] ICCT, “European Vehicle Market Statistics - Pocketbook 2015/16,” 2015. 696

[33] N. E. Ligterink and R. T. M. Smokers, “Monitoring van plug-in hybride voertuigen (PHEVs) april 2012 697

t/m maart 2015,” TNO, Delft, 2015. 698

[34] European Commission, “Reducing CO2 emissions from passenger cars | Climate Action,” 2017. 699

[Online]. Available: https://ec.europa.eu/clima/policies/transport/vehicles/cars_en. [Accessed: 09-Jul-700

2017]. 701

[35] “Overview of electricity production and use in Europe — European Environment Agency.” [Online]. 702

Available: https://www.eea.europa.eu/data-and-maps/indicators/overview-of-the-electricity-703

production-2/assessment. [Accessed: 07-Jul-2017]. 704

[36] M. Messagie, F.-S. Boureima, T. Coosemans, C. Macharis, and J. Mierlo, “A Range-Based Vehicle Life 705

Cycle Assessment Incorporating Variability in the Environmental Assessment of Different Vehicle 706


Technologies and Fuels,” Energies, vol. 7, no. 3, pp. 1467–1482, Mar. 2014. 707

[37] A. Nordelöf, M. Messagie, A.-M. Tillman, M. Ljunggren Söderman, and J. Van Mierlo, “Environmental 708

impacts of hybrid, plug-in hybrid, and battery electric vehicles?what can we learn from life cycle 709

assessment?,” Int. J. Life Cycle Assess., vol. 19, no. 11, pp. 1866–1890, Nov. 2014. 710

[38] S. Rangaraju, L. De Vroey, M. Messagie, J. Mertens, and J. Van Mierlo, “Impacts of electricity mix, 711

charging profile, and driving behavior on the emissions performance of battery electric vehicles: A 712

Belgian case study,” Appl. Energy, Apr. 2015. 713

[39] N. Hooftman, L. Oliveira, M. Messagie, T. Coosemans, and J. Van Mierlo, “Environmental Analysis of 714

Petrol, Diesel and Electric Passenger Cars in a Belgian Urban Setting,” Energies, vol. 9, no. 2, p. 84, Jan. 715

2016. 716

[40] A. Nordelöf, M. Messagie, A.-M. Tillman, M. Ljunggren Söderman, and J. Van Mierlo, “Environmental 717

impacts of hybrid, plug-in hybrid, and battery electric vehicles—what can we learn from life cycle 718

assessment?,” Int. J. Life Cycle Assess., Aug. 2014. 719

[41] M. Goedkoop, R. Heijungs, M. Huijbregts, A. De Schryver, J. Struijs, and R. van Zelm, “A life cycle 720

impact assessment method which comprises harmonised category indicators at the midpoint and the 721

endpoint level,” 2013. 722

[42] L. Oliveira, M. Messagie, S. Rangaraju, J. Sanfelix, M. Hernandez Rivas, and J. Van Mierlo, “Key issues 723

of lithium-ion batteries – from resource depletion to environmental performance indicators,” J. Clean. 724

Prod., vol. 108, pp. 354–362, 2015. 725

[43] P. Mock, U. Tietge, V. Franco, J. German, A. Bandivadekar, N. Ligterink, J. Kühlwein, and I. Riemersma, 726

“From laboratory to road: A 2014 updata of official and ‘real world’ fuel consumption and CO2 values 727

for passenger cars,” no. September, 2014. 728

[44] U. Tietge, N. Zacharof, P. Mock, V. Franco, J. German, A. Bandivadekar, N. Ligterink, and U. Lambrecht, 729

“From laboratory to road: A 2015 update of official and ‘real-world’ fuel consumption From laboratory 730

to road: A 2015 update of official and "real-world" fuel consumption and CO2 values for 731

passenger cars in Europe,” Berlin, 2015. 732

[45] M. Messagie, F. Boureima, N. Sergeant, J.-M. Timmermans, C. Macharis, and J. Van Mierlo, 733

“Environmental breakeven point: an introduction into environmental optimization for passenger car 734

replacement schemes,” in WIT Transactions on The Built Environment, 2012, vol. 128, pp. 39–49. 735

[46] US Department of Energy, “Fuel Economy of New All-Electric Vehicles.” [Online]. Available: 736

http://www.fueleconomy.gov/feg/PowerSearch.do?action=alts&path=3&year1=2016&year2=2017&vtyp737

e=Electric&srchtyp=newAfv. [Accessed: 14-Jun-2017]. 738

[47] CERAM, “Test Report N°16/10681,” Montlhéry Cedex, 2017. 739

[48] G. Fontaras, N.-G. Zacharof, and B. Ciuffo, “Fuel consumption and CO2 emissions from passenger cars 740

in Europe – Laboratory versus real-world emissions,” Prog. Energy Combust. Sci., vol. 60, pp. 97–131, 741

2017. 742

[49] N. G. Zacharof and G. Fontaras, “Review of in use factors affecting the fuel consumption and CO2 743

emissions of passenger cars,” Ispra, 2016. 744

[50] A. Dautfest, “Potenzial zur Steigerung der Energiedichte von HV-Speichern.” Frauenhofer-Institut für 745

Betriebsfestigkeit und Systemzuverlässigkeit LBF, 2017. 746

[51] L. Oliveira, S. Rangaraju, M. Messagie, and J. Van Mierlo, “Increasing The Environmental Potential Of 747

Electric Vehicles And Renewable Energies With Grid Attached Energy Storage,” in EVS28, 2015, pp. 1–748

9. 749


[52] EEA, “Overview of electricity production and use in Europe — European Environment Agency,” 2016. 750

[Online]. Available: https://www.eea.europa.eu/data-and-maps/indicators/overview-of-the-electricity-751

production-2/assessment. [Accessed: 10-Jul-2017]. 752

753

1 In-life range modularity for electric vehicles: the...

Documents

Transcript of 1 In-life range modularity for electric vehicles: the...