Accelerating Multimetallic Catalyst Design for Electrochemical CO2 Reduction using Quantum Chemical Modeling and Machine Learning

本文提出的工作是一项计算研究，旨在确定新的多金属纳米材料，用于将二氧化碳（CO2）有效地电化学转化为增值化学品和燃料。这有双重好处：减少温室气体二氧化碳的排放，并在太阳能和电化学转换过程相结合的闭环碳循环的基础上，更接近可持续能源的未来。先前的研究表明，铜（Cu）纳米立方在碳碳键形成方面表现出显著的选择性，但其电效率太低，无法在商业上可行。这项研究是基于这样一个假设，即由精确混合的低成本金属组成的多金属纳米立方体可以比单独的铜纳米立方体以更高的效率和选择性将二氧化碳转化为有用的化学物质和燃料。研究人员将密度泛函理论计算和从头算分子动力学的专业知识结合起来，在先进的机器学习算法的帮助下，预测材料组合，降低电化学将二氧化碳还原为乙烯和乙醇的过电位。该研究基于三步法，首先揭示CO2在Cu纳米立方体上还原的活性位点和反应机制，然后通过机器学习模型建立将纳米颗粒组成和结构与表面反应性联系起来的预测模型，最后开发一个加速催化剂发现的集成框架。这项工作的广泛影响将通过教育外展活动和对项目过程中开发的工具的开源访问而得到加强。

1604984Xin, HongliangThe proposed work is a computational study aimed at identifying novel multimetallic nanomaterials for the efficient electrochemical conversion of carbon dioxide (CO2) to value-added chemicals and fuels. This has the dual benefit of reducing the emissions of the greenhouse gas CO2 and moving closer to a sustainable energy future based on a closed loop carbon cycle fueled by a combination of solar energy and electrochemical conversion processes. Prior research has demonstrated that copper (Cu) nanocubes exhibit remarkable selectivity towards carbon-carbon bond formation, but with electrical efficiency too low to be commercially viable. The study is based on the hypothesis that multimetallic nanocubes consisting of precisely mixed, low-cost metals can convert CO2 to useful chemicals and fuels at higher efficiency and selectivity than the Cu nanocubes alone. The researchers bring together expertise in density functional theory calculations and ab initio molecular dynamics - aided by advanced machine-learning algorithms - to predict materials combinations that lower the over-potential for electrochemical reduction of CO2 to ethylene and ethanol. The research is based on a three-step approach that first unravels the active site and reaction mechanism of CO2 reduction on Cu nanocubes, then creates predictive models linking nanoparticle composition and structure to the surface reactivity by machine-learning models, and lastly, develops an integrated framework for accelerating catalyst discovery. The broader impact of the work will be enhanced through educational outreach activities and open-source access to the tools developed during the course of the project.