Department of Chemical and Biomolecular Engineering, University of California Berkeley, Berkeley, California 94720, United States.
Liquid Sunlight Alliance, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States.
ACS Appl Mater Interfaces. 2023 May 17;15(19):23024-23039. doi: 10.1021/acsami.2c21114. Epub 2023 May 8.
A metal-insulator-semiconductor (MIS) structure is an attractive photoelectrode-catalyst architecture for promoting photoelectrochemical reactions, such as the formation of H by proton reduction. The metal catalyzes the generation of H using electrons generated by photon absorption and charge separation in the semiconductor. The insulator layer between the metal and the semiconductor protects the latter element from photo-corrosion and, also, significantly impacts the photovoltage at the metal surface. Understanding how the insulator layer determines the photovoltage and what properties lead to high photovoltages is critical to the development of MIS structures for solar-to-chemical energy conversion. Herein, we present a continuum model for charge-carrier transport from the semiconductor to the metal with an emphasis on mechanisms of charge transport across the insulator. The polarization curves and photovoltages predicted by this model for a Pt/HfO/p-Si MIS structure at different HfO thicknesses agree well with experimentally measured data. The simulations reveal how insulator properties (i.e., thickness and band structure) affect band bending near the semiconductor/insulator interface and how tuning them can lead to operation closer to the maximally attainable photovoltage, the flat-band potential. This phenomenon is understood by considering the change in tunneling resistance with insulator properties. The model shows that the best MIS performance is attained with highly symmetric semiconductor/insulator band offsets (e.g., BeO, MgO, SiO, HfO, or ZrO deposited on Si) and a low to moderate insulator thickness (e.g., between 0.8 and 1.5 nm). Beyond 1.5 nm, the density of filled interfacial trap sites is high and significantly limits the photovoltage and the solar-to-chemical conversion rate. These conclusions are true for photocathodes and photoanodes. This understanding provides critical insight into the phenomena enhancing and limiting photoelectrode performance and how this phenomenon is influenced by insulator properties. The study gives guidance toward the development of next-generation insulators for MIS structures that achieve high performance.
金属-绝缘体-半导体(MIS)结构是一种很有吸引力的光电电极-催化剂结构,可促进光电化学反应,例如质子还原形成 H。金属通过半导体中光子吸收和电荷分离产生的电子来催化 H 的生成。金属和半导体之间的绝缘层可保护后者免受光腐蚀,并且还会显著影响金属表面的光电压。了解绝缘层如何决定光电压以及哪些特性会导致高光电压,对于开发用于太阳能到化学能转换的 MIS 结构至关重要。在此,我们提出了一种从半导体到金属的载流子输运的连续体模型,重点介绍了穿过绝缘体的电荷输运机制。该模型预测的 Pt/HfO/p-Si MIS 结构在不同 HfO 厚度下的极化曲线和光电压与实验测量数据吻合较好。模拟结果揭示了绝缘层特性(即厚度和能带结构)如何影响半导体/绝缘层界面附近的能带弯曲,以及如何调整这些特性可以导致更接近最大可达到光电压(平带电势)的操作。通过考虑绝缘层特性变化引起的隧道电阻变化,可以理解这种现象。该模型表明,最佳 MIS 性能是在具有高度对称的半导体/绝缘层能带偏移(例如,沉积在 Si 上的 BeO、MgO、SiO、HfO 或 ZrO)和低至中等厚度的绝缘层(例如,0.8 至 1.5nm 之间)的情况下获得的。超过 1.5nm 时,填充的界面陷阱位密度很高,会显著限制光电压和太阳能到化学能的转换速率。这些结论对于光电阴极和光阴极都是正确的。这些结论提供了对增强和限制光电电极性能的现象的重要见解,以及这种现象如何受绝缘层特性影响的见解。该研究为开发实现高性能的 MIS 结构的下一代绝缘体提供了指导。