Combining High-Throughput Analysis and Data-Driven Electrochemistry for Advanced Characterization of Si-Ge-Sn Thin Films as Li-Ion Anodes

In this study, a thin film of Si-Ge-Sn alloy with a gradient composition distribution was synthesized using combinatorial magnetron sputtering. This thin-film ternary material library was electrochemically characterized as a lithium-ion anode on a millimeter-scale using a three-electrode half-cell scanning droplet cell. The experimental protocol was driven by automated on-the-fly electrochemical analysis and Bayesian optimization with a Gaussian process to find the chemical composition with the highest specific capacity. Comparative analysis revealed a significant improvement in Li-ion battery performance with the best-performing Si-Ge-Sn anodes outperforming those with conventional graphite anodes, with increases in gravimetric energy densities of over 30%. Advanced high-throughput µ-XRF and Raman spectroscopy provided valuable insights into the composition-structure-property relationships of the Si-Ge-Sn system. Additionally, the results of XPS and EIS studies were used to elucidate the discrepancies between theoretical and experimental reversible capacities of Si- and Ge-rich compounds.

本研究采用组合式磁控溅射技术，合成了具备梯度成分分布的硅-锗-锡（Si-Ge-Sn）合金薄膜。针对该薄膜三元材料库，研究团队采用三电极半电池扫描液滴池装置，在毫米尺度下开展锂离子负极的电化学表征工作。本实验流程依托实时自动电化学分析与基于高斯过程的贝叶斯优化驱动，以筛选出比容量最高的化学组分。对比分析结果显示，锂离子电池性能得到显著提升：性能最优的Si-Ge-Sn负极性能优于传统石墨负极，重量能量密度提升幅度超过30%。先进的高通量微区X射线荧光光谱（µ-XRF）与拉曼光谱技术，为揭示Si-Ge-Sn体系的成分-结构-性能关联提供了重要见解。此外，本研究借助X射线光电子能谱（XPS）与电化学阻抗谱（EIS）的测试结果，阐明了富硅与富锗化合物的理论可逆容量与实验可逆容量之间的偏差。