Experimental and techno-economic analyses of low-cost battery materials and processes

Batteries have the potential to facilitate broader global access to reliable and affordable clean energy. Therefore, battery research focused on reducing the lifetime costs, manufacturing and waste impacts, and mining pressures associated with existing battery technologies are of utmost importance. To contribute to this effort, this dissertation delves into technological innovations at several key steps in the battery supply chain, from end-of-life assessment and reuse to raw material selection prior to manufacturing.

Chapter 2 focuses on developing a holistic assessment of the state-of-health (SOH) of batteries after use in a hybrid-electric vehicle with the goal of selecting cells for reuse in a second-life application. Four hundred LiFePO₄ (LFP) cells were sampled from a retired hybrid-vehicle battery pack and tested using constantcurrent cycling, the current gold standard, and electrochemical impedance spectroscopy (EIS), a faster and cheaper alternative to cycling. In contrast to existing battery diagnostic studies, which only report cell capacity and resistance, this chapter analyzes correlations between battery diagnostic parameters, the spatial dependence of SOH within the pack, and correlations between cycling and EIS test results. We found that SOH variation within the pack is high in used EV packs, making it unfavorable to reuse them without refurbishment. We also found strong correlations between EIS and cycling tests under specific EIS test conditions, suggesting that low-cost, fast EIS tests are potentially valuable substitutes for cycling tests. Furthermore, using two distinct acceptance criteria, we estimated that 75% or 94% of modules and 90 or 99% of cells would likely be useful in second-life applications, respectively. 

Chapter 3 presents an assessment of the economic and technical feasibility of repurposing spent LFP cells based on the state-of-health insights from Chapter 2. We developed a model for investigating whether battery reuse can provide an economically sustainable solution to the problem of spent battery management while also addressing grid storage needs. Using Monte Carlo simulations, we probabilistically compared the pack lifetime, manufacturer revenue, and levelized cost of storage for end-users across two common battery chemistries—LiFePO₄ (LFP) and Li-Ni-Mn-Co (NMC)—within two applications—residential load-following and renewables firming. We found that LFP outperforms NMC in terms of revenue, enduser costs, and lifespan across various pack processing strategies. We also found that residential loadfollowing demonstrated superior revenue potential over renewables firming, despite the lower capital costs associated with the latter. Our results suggest that reusing entire packs might offer longer lifetimes and potentially lower costs for end-users than other strategies, but future research should study the safety implication of this method. However, pack sizing sensitivity analysis suggested that producing smaller packs drastically increased the optimistic estimates of lifetime. 

Chapter 4 focuses on recycling pathways for commercial batteries. In particular, we assessed the possibility of directly recycling LFP cathodes. We tested three types of cathode relithiation treatments—lithium iodide (LI), lithium hydroxide with citric acid, annealed (LHC), and lithium iodide, annealed (LIA). Cathode materials' crystal structure, morphology, composition, and electrochemical performance were characterized before and after relithiation. Based on insights from the LI and LHC methods, we developed the novel LIA method, which combines a galvanic relithiation process with a solid-state annealing process, yielding the best performance results for all states of health. In contrast, the LI method, which lacked a solid-state annealing process, restored the LiFePO₄ crystal structure of the harvested material but did not fully restore the characteristic electrochemical discharge profile of LiFePO₄. The LHC method worked well on samples from two healthier cells but severely degraded the sample from the least healthy cell. We implemented a techno-economic analysis, which showed that LI and LIA relithiation are considerably more expensive than LHC relithiation due to the high cost of lithium iodide. However, all three relithiation methods were shown to be cost-competitive with LFP manufacturing for specific ranges in lithium replacement fractions. LHC relithiation was the lowest cost manufacturing process, with estimated breakeven recovery costs ranging from 5 to 30 USD per kg (10-61 USD per kWh) at 100% lithium replacement. 

Finally, Chapter 5 showcases an innovative alkaline MnO₂ cathode and electrolyte pairing. The system explored in this concluding project offers advantages in terms of lower costs, raw material abundance, enhanced safety, and comparable full-cell capacity to LFP. We found that electrochemical synthesis and cycling of layered MnO₂ (birnessite) can be improved using lower-cost NaOH electrolyte, which has not been studied in relevant existing literature, instead of the higher-cost KOH electrolyte, which has been repeatedly studied in existing literature. Furthermore, the birnessite cathode’s cycling stability was significantly improved in electrolyte-starved environments, demonstrating a stable capacity of approximately 400 mAh/g for over 400 cycles in NaOH. To our knowledge, NaOH and electrolyte-starved environments have not been studied in existing birnessite cathode literature. Thus, this study demonstrates essential directions for future research in improving the commercial viability of this low-cost, environmentally sustainable alternative to Li-ion batteries.