基于跨尺度高通量计算的铁电材料模拟和计算设计

Although high-throughput calculations have got great success in the design of new materials, multiscale high-throughput calculations and design of new ferroelectrics have not received enough attention. The current proposal will try to establish multiscale high-throughput calculations that combine first principles calculations with phase field simulations, and apply them to the analysis of electrocaloric effect in BaTiO3-based ceramics. First, a strategy will be developed to transfer physical properties calculated from first-principles calculations to a phase-field model as the coefficients of free energy function, which links the atomic level and mesoscale simulations for ferroelectric materials. A task management system for high-throughput phase field simulations then will be designed based on the established multiscale model. The functionalities of the system include automatic acquisition of coefficients of phase field model from first principles calculations, automatic generation of computational tasks, querying and monitoring tasks, data storage and so on. By using the multiscale high-throughput calculations, the demonstration application of controlling electrocaloric effect in BaTiO3-based solid solution will be carried out. Outputs of the present project are believed to be not only beneficial to the development of new functionalities in ferroelectrics, but also to the study of computational materials science.

高通量计算近年来在新材料设计方面取得了重要进展，但是跨尺度高通量计算在铁电材料研究领域尚未受到足够的重视。本项目拟构建第一原理和相场模拟相结合的铁电材料跨尺度高通量计算方法，并在此基础上对电卡性质的调控优化进行示范应用。首先，利用第一原理计算材料物理性质并拟合生成相场模型中自由能展开式的参数，从而实现原子尺度与介观尺度计算的有效结合。基于该多尺度计算模型，设计高通量相场模拟的自动化流程计算平台，包括不同材料自由能参数的自动获取、计算任务的自动化生成、计算任务的查询与监控、材料计算性质的分类存储与分析等功能，从而实现铁电材料特定功能的高通量模拟和设计。最后，基于得到的跨尺度高通量计算方法，开展钛酸钡基固溶体电卡性质多自由度调控（成分和应变）的示范应用。本项目成果既能加快新型多功能铁电材料的设计开发，又能促进计算材料学科研究的发展。

铁电、铁弹等铁性功能材料具有力、电、光等各种优异的性能，广泛应用于生产和生活各个领域。本项目采用跨尺度计算方法和高通量相场模拟并结合机器学习，开展了铁性材料电学、力学等性能的调控优化以及新功能的开发和设计。基于第一原理计算、有效哈密顿模型和蒙特卡洛模拟相结合的计算方法，在铁电材料中实现了新型的多铁金属性、二维铁电性等；设计了相场模拟的高通量计算框架，自动化实现材料种类选择、输入文件生成、相场计算程序FEAP的自动调用、计算数据分类存储、计算结果自动分析等流程的相场计算，从而实现材料特定性能的高效计算；利用第一原理计算和高通量相场模拟并结合机器学习，实现了特定力学性能的铁性材料设计，同时也开展了成分和应变多维度调控铁电材料电卡效应的研究。