As an efficient means of energy storage, rechargeable batteries, especially the lithium-ion batteries (LIBs) have been a vital component in solving the upcoming energy crisis and environmental problems. Recently, the development of electric vehicle market puts new requirement on the next generation LIBs, including superior energy density, safety and cycling stability, etc. Compared with experimental investigation, Physics-based models provide a surrogate method to not only tackle the underlying physics of the complex battery system, but also optimize the design of battery systems. In this thesis, I will show how I use the physics-based continuum model and cooperate with some experimental methods to understand the lithium transport phenomena inside the multiscale battery electrode systems, based on which the models are then applied to guide the experimental optimization of battery electrode design and to quantitively understand the degradation of high-performance electrodes.
The thesis is divided into three parts. First part (Chapter 2) presents a systematical model selection study on the multiscale LiNi₀.₃₃Mn₀.₃₃Co₀.₃₃O₂ (NMC₁₁₁) electrode. Discharge and voltage relaxation curves, interrogated with theory, are used to distinguish between lithium transport impedance that arise on the scale of the active crystal and on the scale of agglomerates (secondary particles) comprised of nanoscale crystals. Model-selection algorithms are applied to determine that the agglomerate scale transport is dominant in the NMC₁₁₁ electrode studied here. This study not only discovers the dominant length scale for lithium transport, but also provide a validated model (the agglomerate model) for later study.
The second part (Chapter 3 & 4) talks about understanding & optimization of ion transport in porous electrodes. In Chapter 3, multi-scale physics-based models for different active material systems, which have been parameterized and validated with discharge experiments, are optimized by varying porosity and mass loading to achieve maximum volumetric energy density. The optimization results show that with a re-scaling of the current rate, the optimal results follow a general design rule that is captured in a convenient correlation. Chapter 4 extends the model to simulate the performance of advanced electrode architectures utilizing aligned channels, by quantifying the impact of aligned channel electrode structures on cell rate capability. Then the optimization algorithm in Chapter 3 is applied to these aligned-channel electrodes.
The final part (Chapter 5) shows how I use the physics-based model to quantitatively analyze the battery degradation. The validated model is applied to cycling data to obtain parameter estimates indicative of degradation modes. It’s found that growth rates of interfacial impedance and active material loss are greater at 4.5 V, as might be expected. However, when charged to 4.5V, degradation rates are lower at a cycling C-rate of 1.0 h⁻¹ than at 0.5 h⁻¹. Once performance changes are quantified, we use further simulation to evaluate the contribution of individual degradation modes to fade of cell performance metric such as capacity, power density, and energy density.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/vw31-t946 |
| Date | January 2022 |
| Creators | Hui, Zeyu |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0024 seconds