This report aims to provide knowledge on how to decrease the environmental impact of electric vehicles by optimizing the size of the battery carried by the vehicle. Two different sizes of batteries, permanently installed in a fictional car, has been compared using life cycle assessment. A third alternative, swapping between two sizes of batteries in a swapping station has also been evaluated. All alternatives have been investigated with two charging regimes: charging availability at home or at work, or no such possibilities (thus having to rely on fast charging), has been investigated. The system boundary includes the full life cycle of batteries and charging infrastructure, but excludes the rest of the vehicle. Some results are however extended to a complete vehicle scenario to enable comparison with other studies. The use phase is modelled by the electricity required to drive the vehicle including charging losses.

The results indicate the following conclusions: • The large (70 kWh) permanently installed battery gives the largest climate impact in (almost) all investigated scenarios. • The climate impact for the swapping alternative ends up in between the small (35 kWh) and the large (70 kWh) battery cases, under most conditions. Only when the number of cars per swapping station are reduced by a factor of 10 will the swapping alternative give the largest climate impact. However, the data currently obtained and used for swapping stations is incomplete and therefore any related conclusion is insecure. • Home charging with low voltage alternating current (AC) electricity and on-board charger, gives a bit more climate impact than direct current (DC) fast charging due to higher losses, both during distribution and charging. Possibilities for home charging to charge during non-peak hours with assumingly lower carbon footprint as well as potential to provide grid services, could change this around, i.e. making AC-charging the preferred option from a climate impact point of view. • The charging infrastructure contributes with 1.2-3.5% to the total climate impact per vehicle kilometre. The charging losses could amount to 3.2-5.7% of the total climate impact per vehicle kilometre. These percentages would be smaller in a complete vehicle scenario. • Sodium ion chemistry (compared to NMC chemistry) could provide 15-25% less total climate impacts and 32-42% less resource depletion. Also LPF chemistry perform better than NMC chemistry, both in terms of climate impact and resource depletion, but less so than sodium ion chemistry. These percentages would be smaller in a complete vehicle scenario. • When average Chinese electricity mix is used for charging, the total carbon impact is 3-4 times higher than with the assumed base case global 2030 mix. The use phase dominates the total carbon impact. With Chinese electricity, but otherwise base case conditions, the 70 kWh vehicle scores 255 grams CO2-eq/km, which is the highest total climate impact calculated. This figure would be 333 grams CO2-eq/km in a complete vehicle scenario. • When average Swedish electricity is used for charging, the use phase carbon impact is 4-5 times lower than with the assumed base case 2030 mix. The battery production phase dominates the total carbon impact. With Swedish electricity, but otherwise base case conditions, the 35 kWh vehicle scores 27 grams CO2-eq/km, which is the lowest total climate impact calculated. This figure would be 65 grams CO2-eq/km in a complete vehicle scenario.