法拉第效率
卤化
电极
钝化
材料科学
电解质
硅
纳米技术
化学工程
无定形固体
卤化物
表面改性
复合数
图层(电子)
锂(药物)
能量转换效率
相间
氟
极化(电化学)
非晶硅
作者
Haosheng Li,Yaru Li,Guantai Hu,Ying Li,Caijin Xiao,Liang Zhao,Huiqin Huang,Haochang Zhang,Wei Xia,Ning Lin
标识
DOI:10.1038/s41467-025-67985-x
摘要
Silicon-based solid-state batteries are promising next-generation high-energy-density technologies. However, poor (electro)chemical compatibility between silicon negative electrodes and solid electrolytes (e.g., Li6PS5Cl) plus sluggish interfacial kinetics severely limits their reversibility and Coulombic efficiency. Here, we propose a surface halogenation strategy that transforms the native amorphous SiO2 passivation layer on silicon particles into a functional Al(Si)OCl composite surface via controlled reaction with AlCl3. This artificial interphase reconciles interfacial incompatibility and enables fast ionic/electronic transport, suppressing irreversible lithium loss. The optimized negative electrode achieves a high initial Coulombic efficiency of 94.3% in half-cells and 85.6% initial Coulombic efficiency (86.6% with pre-lithiation) in full cells paired with LiNi0.88Co0.09Mn0.03O2. Enhanced reversibility further delivers long-term cyclability. The optimized negative electrode delivers 86% capacity retention and 99.998% average Coulombic efficiency over 200 cycles. Even at high-loading ( > 10 mAh cm-2, and no adhesives/conductive carbon/electrolyte), it retains 72% capacity after 500 cycles. The full cells maintain 80% capacity after 200 cycles at 1 C, with an average Coulombic efficiency exceeding 99.95%. The versatility of this halogenation strategy underscores halide chemistry’s broad potential in advancing high-performance, reversible silicon-based solid-state batteries. Silicon negative electrodes in solid-state batteries exhibit poor reversibility. Here, the authors demonstrate surface halogenation engineering that suppresses irreversible lithium loss, achieving 94.3% initial Coulombic efficiency and 72% capacity retention over 500 cycles at 25°C.
科研通智能强力驱动
Strongly Powered by AbleSci AI