Abstract The successive oxide failure statistics theory that arises from the clustering model is used for investigating the time-to-breakdown distributions of Al2O3/HfO2-based nanolaminates. These gate dielectrics in metal-insulator-semiconductor structures are intended to combine a high injection barrier material (Al2O3) with a high-K material (HfO2). When a constant voltage is applied to the structures, the current-time characteristics exhibit stepwise changes that correspond to the sequential opening of multiple conducting channels across the dielectric film. This study demonstrates that a clustering-based approach successfully describes the ordered breakdown statistics associated with the appearance of these conducting channels. Deviations of the experimental data from the expected Weibull statistics at the high percentiles are accounted for by introducing the so-called clustering factor. The origin of the deviations is attributed to the high dispersion of the initial leakage current value and consequently to the initial conducting properties of the oxide film.

中文翻译：

---

使用时间相关聚类模型分析多层 Al2O3/HfO2 栅极堆叠的连续击穿统计

摘要 由聚类模型产生的连续氧化物失效统计理论用于研究基于 Al2O3/HfO2 的纳米层压板的击穿时间分布。金属-绝缘体-半导体结构中的这些栅极电介质旨在将高注入势垒材料 (Al2O3) 与高 K 材料 (HfO2) 结合起来。当对结构施加恒定电压时，电流-时间特性表现出阶梯式变化，对应于跨介电膜的多个导电通道的顺序打开。这项研究表明，基于聚类的方法成功地描述了与这些传导通道的出现相关的有序分解统计数据。通过引入所谓的聚类因子来解释实验数据与高百分位数处的预期威布尔统计数据的偏差。偏差的起源归因于初始漏电流值的高度分散，因此归因于氧化膜的初始导电特性。