□ 연구개요 - Liquid metals (LMs) have been studied for the use as heat transfer fluids in nuclear energy systems such as liquid Na for sodium cooled fast reactors (SFRs) and energy storage systems. In addition to these traditional applications, LMs have been attracting large attention as advanced functional materials in various fields, such as flexible electronics, biomedical application, soft robot, catalyst, etc. - For these applications, the chemistry of LMs, such as reactivity and corrosion characteristics, is important. However, the comprehensive database and understanding of the LM chemistry are still not available. - In this project, we aimed to first develop computational methods to efficiently and accurately calculate the solution enthalpy and diffusion coefficients as important physio-chemical properties of LMs, and then to construct all-element database taking liquid Na as a test case. □ 연구 목표대비 연구결과 - < Task-1: Acquisition of first-principles calculation data > We obtained first-principles molecular dynamic (FPMD) calculation data for 96 elements (from H to Cm) in liquid Na for the use in Tasks-2, 3 and 4. - < Task-2: Construction of all-element database for chemical states of impurity > Excluding 2 elements which showed bad convergence in the first-principles calculations and are not important for liquid Na application in nuclear engineering (Sm-62 and Yb-70), we determined the charge state, atomic radius, coordination number, etc, of 94 elements as impurity atoms in liquid Na at 600 K and 1000 K. - < Task-3: Development of machine-learning potential models based on FPMD simulation results > We established an efficient and automated method to construct accurate machine-learning (ML) moment tensor potentials (MTPs) based on FPMD calculation results. Using the established method and the FPMD calculation data prepared in Task-1, we developed MTPs for 94 elements in liquid Na. - < Task-4: Calculation of solution enthalpy and diffusion coefficients by classical molecular dynamics (CMD) using MTPs > We established all-element database (= 94 elements) for solution enthalpies and diffusion coefficients of impurities in liquid Na at 600 K and 1000 K by CMD using MTPs constructed in Task-3. - < Task-5: Validation of calculation results > We compared the calculated solution enthalpies and diffusion coefficients with available experimental data, and confirmed good agreement with experiments for solution enthalpies, except for several elements whose experimental data are likely to be affected by oxide formation. For diffusivity, due to very low availability of experimental data, we only compared the calculation and experiment for self-diffusion coefficient of liquid Na, showing good agreement. - < Task-6: Derivation of correlation between fundamental impurity properties with solution enthalpy and diffusion coefficient > We analyzed the correlation between fundamental properties of impurities, such as electronegativity and atomic radius, with solution enthalpy and diffusion coefficient, and established model equations to describe all-element data of solution enthalpy and diffusion coefficient as functions of fundamental impurity properties. - < Task-7: Conclusion and identification of future research directions > As a future research direction, we proposed new liquid alloy discovery for nuclear energy systems applications by means of materials informatic approach using the computational methods established in this project. □ 연구개발성과의 활용 계획 및 기대효과(연구개발결과의 중요성) - The constructed all-element database can be used as fundamental data to improve the safety on the use of liquid Na. For example, the database can be used to effectively design new materials/devices to mitigate the consequence of Na fire and to establish reliable and efficient safety regulation. - Through this project, we established an efficient computational framework to construct all-element database. This framework can be easily applied to other LMs, such as liquid Pb, and liquid Li, and can help design new LM alloys of low chemical reactivity as well as corrosion-resistant structural materials for the applications in nuclear energy systems, flexible electronics, soft robotics, etc. The methods can be also applied to molten salts for the R&D of molten salt reactors (MSRs). - The all-element database established in this study is a first-of-kind database in the world, and significantly advanced the chemistry of LM. (source : 연구결과 요약문 2p)
- 연구책임자 : Takuji Oda
- 주관연구기관 : 서울대학교
- 발행년도 : 20230300
- Keyword : 1. 액체 금속;용해 엔탈피;확산 계수;불순물;원자론적 시뮬레이션; 2. liquid metal;solution enthalpy;diffusion coefficient;impurity;atomistic simulation;
