Comparative performance study of electric vehicle batteries repurposed for electricity grid energy arbitrage

Abstract

Electric vehicle (EV) batteries can provide extended value beyond EV service if they are repurposed for a “second life” in electricity grid applications. However, because batteries from different EV makes and models vary significantly by size, shape, chemistry, and thermal management, there is uncertainty regarding their relative performance in second-life applications. This experimental study evaluates seven different EV batteries in their original modules and/or packs, featuring four unique positive active materials, two negative active materials, three cell formats, and four thermal management designs. Each battery is subjected to deep-discharge cycling at 4 h, 2 h, and 1 h constant-power rates to emulate performance in electricity grid energy arbitrage. Test results are evaluated based on six battery performance metrics in three key performance categories, including two energy metrics (usable energy capacity and charge–discharge energy efficiency), one volume metric (energy density), and three thermal metrics (average temperature rise, peak temperature rise, and cycle time). Significant differences in performance arise from the variety of chemistries and thermal management systems tested, dominating any influence from battery state of health. Chevrolet Volt and EnerDel batteries (both from hybrid EVs using NMC chemistry) give the best usable energy capacity (≥94%) and energy efficiency (≥97%), while Tesla Model S batteries (from long-range EVs using NCA chemistry) give the lowest usable energy capacity (≥84%) and energy efficiency (≥89%). However, the ModelS batteries give roughly double the energy density (half the physical footprint) of the Volt and EnerDel batteries. The Volt battery experiences no more than 2 °C of warming even during a 1 h discharge, thanks to its active (forced) liquid thermal management and high energy efficiency. This contrasts with the Leaf and Lishen batteries, which use passive (natural convection) thermal management and consequently experience over 17 °C of warming during a 1 h discharge, and then require over 4 h of standby time to cool down by less than 10 °C. Novel analytical techniques are applied to the experimental results to rank the tested EV batteries in the three aforementioned performance categories to illustrate their relative commercial performance expectation in second-life energy arbitrage. This new performance ranking system can be employed by industry in conjunction with economic models to select the most appropriate used EV batteries for specific energy storage applications.