Battery Modeling to Enable Electric Aviation

Few technologies have changed the world more than lithium-ion batteries or heavier than air aircraft. Lithium-ion batteries allow for small, safe, and portable electrical energy storage. They are safer, last longer, and have higher specific energies and powers than most previous electrochemical energy sources. In the last few decades, cars and trucks powered by lithium-ion batteries have gained adoption due to sustainability concerns about the global warming impact of fossil-fuel burning vehicles. While aviation has had an outsized positive impact on the global economy, it too has come under increased scrutiny for its environmental impact. Aircraft require large amounts of power and energy, and have secondary negative societal effects such as the formation of contrails, the release of harmful pollutants such as nitrous oxides, and high noise levels. One proposed solution to these problems is to electrify aircraft using electrochemical batteries. Electric aircraft emit no pollutants, and electric motors are much quieter than their combustion-based counterparts. This promise has led to the development of numerous production, prototype, and conceptual electric aircraft. Of particular interest, electrification has been cited as an enabling technology for urban air mobility (UAM). UAM proposes to reduce traffic and speed commutes at a comparable energy efficiency to electric vehicles, mainly through the use of small electric vertical takeoff and landing (EVTOL) aircraft. However, despite the enormous interest, the requirements, safety, performance, cost, and design of batteries for electric aircraft remain unclear. 

In this thesis, I explore the requirements, performance, projected lifetime, safety considerations, and design of batteries for electric aircraft. I begin by using the fundamental equations of aircraft performance to elucidate the performance metrics required of next generation batteries to electrify commercial aircraft. I then compare the aircraft that could be electrified using next-generation chemistries such as lithium-air. I find that even in the most optimistic scenario, only regional aviation and some narrow body flights can be electrified using the most advanced battery chemistries known, and that only small aircraft such as urban air mobility aircraft or general aviation aircraft can be electrified using current batteries. Next, I identify two main safety concerns for batteries in electric aircraft: functional failures, which correspond to failures of the battery to meet the power demands of the aircraft, and thermal failures, which correspond to catastrophic failure due to excess heat, fire, and explosions. In order to mitigate the functional failures, I develop a large dataset of battery cycling data. To gain deeper insight into battery degradation, I conduct a model-based diagnosis of battery degradation for electric aircraft. Using an electrochemical model capable of simulating the baseline EVTOL aircraft power profile in 385 µs or a 1C discharge in 98 µs and a parameterization of that model that enables us to physically constrain the modes. I independently sample the degradation modes for each of the 21,392 EVTOL cycles across 22 cells, finding significant trends in positive and negative electrode active material loss and resistance gain. The model shows a median error of 32.5 mV across all cycles. I show that resistance and loss of active material in each electrode are identifiable. I find that the rate of positive electrode active material loss is most strongly correlated with minimum current (maximum charging current) (ρ = 0.5), the rate of negative electrode active material loss is most strongly correlated with depth of discharge (ρ = −0.422), and the rate of resistance increase is most strongly correlated with depth of discharge (ρ = 0.413). I introduce a protocol for early identification of active degradation mechanisms by computing the cosine distance between the identified degradation modes and the modes expected for each mechanism, identifying electrolyte oxidation and active material dissolution as the most likely causes of positive electrode degradation and SEI growth as the most likely cause of negative electrode degradation within the first 50 cycles. Next, I optimize parameters for multiple degradation mechanisms, and show that a diffusion-limited SEI model combined with a plating model and a model for stress-driven loss of active material in the positive electrode best explain the evolution of the posterior distributions of the degradation modes. I supplement the physics-based model with a data-driven model based on neural ordinary differential equations, and show that the hybrid physics and machine learning model outperforms the physical model alone. I then apply the developed diagnosis method to batteries in satellites in low earth orbit. I find that charging to the maximum upper cutoff voltage leads to a “kneepoint” caused by large losses of positive electrode active material. I simulate LEO satellite battery degradation, including mechanical stress, using an electrochemical-thermal model. By incorporating a concentration-dependent partial molar volume, I show that charging to the maximum upper cutoff voltage results in a 35% increase in the positive electrode surface tangential stress. I introduce a mechanistic degradation model that accurately predicts large losses of positive electrode active material at high upper cutoff voltages and very little at lower upper cutoff voltages. I use the model to develop a protocol for accelerating aging of satellite batteries. Finally, I explore the tradeoffs of various cooling methods for electric aircraft battery packs. I develop 1-dimensional thermal network models for various battery pack cooling systems, including indirect liquid cooling, forced air cooling, natural air cooling, and immersion cooling. I conduct trade studies for the immersion cooling systems for a variety of coolants on the market. I compare the degradation and cell to cell variation of each cooling method. Finally, I close with a discussion of the outlook for electric aircraft.