Understanding the Structural Dynamics of Polarons in Transition Metal Oxide Semiconductors by Vibrational Spectroscopy and Charge-Carrier Mobilities

通过光催化或光电化学产生太阳能燃料肯定是当今最重要和最有前途的技术过程之一。尤其是光催化分解水，将水分离为气态氧和氢，近年来引起了人们的广泛兴趣，因为它被认为是一种关键的储能技术。值得注意的是，尽管过去几十年来在研究方面做出了巨大努力，但还没有可用的光催化剂来经济地分解水。一类可以用来驱动光催化分解水的材料是过渡金属氧化物。这些氧化物是半导体，当光子被吸收时，电子从价带激发到导带。从而发生电荷分离，所产生的自由电子和电子空穴可以分别用作还原剂或氧化剂。这些载流子通常被困在过渡金属氧化物中，因为它们形成极化子，电荷周围的原子在那里移位，从而产生潜在的最小值。在原子水平上描述极化子状态时，理论方法是不可避免的。然而，这种通常基于混合密度泛函的电子结构计算可能是模棱两可的，以至于极化子的确切几何形状在很大程度上(甚至在定性上)取决于所选择的泛函。动力学相关的性质，如迁移率或电子转移速率常数，显然取决于极化子的几何形状。因此，精确的极化子结构的详细知识对于后续的研究项目至关重要。在这个项目中，我提出了对BiVO4中极化子态的高级理论模拟研究，BiVO4是高效光催化分解水的主要候选者。在杂化密度泛函理论水平上对BiVO4中的空穴极化子进行了从头算分子动力学模拟。从这些模拟，振动光谱(红外和拉曼)和电荷载流子迁移率将被计算。所有的模拟，包括振动光谱的计算，都将通过机器学习来加速，这对于获得统计收敛的数据是绝对必要的。安排了与著名实验者的合作，他们将测量相应的振动光谱和迁移率。因此，本项目的目的是为电子结构理论提供一个基准；在这种情况下，确定BiVO4中正确的电子空穴极化子几何形状，但所应用的方法也适用于任何其他金属氧化物材料。此外，我还将从理论和实验的联合数据中对BiVO4中电子空穴极化子的结构动力学有一个详细的了解。这对于寻求将太阳能转换为燃料的技术作为低碳能源系统的一部分进行优化的社区来说，可能具有重大价值。

The generation of solar fuels by photocatalysis or photoelectrochemistry is certainly one of the most important and promising technological processes nowadays. Especially photocatalytic water splitting, where water is separated into gaseous oxygen and hydrogen, attracted a wide interest recently because it is envisaged as a key technology for energy storage. Remarkably, there is yet no photocatalyst available which can split water economically, despite the vast efforts in research over the last decades. One class of materials which can be used to drive photocatalytic water splitting are transition metal oxides. These oxides are semiconductors, where an electron is excited from the valence to the conduction band when a photon is absorbed. Thereby charge separation occurs and the resulting free electron and the electron hole can be used individually as reducing or oxidizing agent, respectively. These charge carriers are often trapped in transition metal oxides since they form polarons, where the atoms in the surrounding of the charge displace and thus create a potential minimum. When it comes to characterizing polaronic states on an atomistic level, theoretical methods are inevitable. However, such electronic structure calculations, which are usually based on hybrid DFT, can be ambiguous such that the exact geometry of a polaron significantly (and even qualitatively) depends on the chosen functional. Dynamically relevant properties, such as mobilities or electron transfer rate constants, obviously depend on the geometry of the polaron. A detailed knowledge of the exact polaronic structure is therefore crucial for following research projects. In this project, I propose an advanced theoretical simulation study of polaronic states in BiVO4 which is a leading candidate for efficient photocatalytic water splitting. Ab initio molecular dynamics simulations on the hybrid DFT level of theory are to be conducted of the hole polaron in BiVO4. From these simulations, vibrational spectra (IR and Raman) and charge-carrier mobilities are to be calculated. All simulations, including the calculation of the vibrational spectra, will be accelerated by machine learning which is absolutely necessary to obtain statistically converged data. Collaborations with renowned experimentalists are arranged, who will measure the corresponding vibrational spectra and mobilities. The aim of this project therefore is, to provide a benchmark for the electronic structure theory; in this case to definitively identify the correct electron hole polaron geometry in BiVO4, but the applied methods are transferable to any other metal oxide material as well. From the joint theoretical/experimental data, I will moreover gain a detailed understanding of the structural dynamics of the electron hole polaron in BiVO4. This could be of significant value to the community seeking to optimise solar-to-fuel conversion technologies as part of a low-carbon energy system.