液态金属最大过冷能力的定量预测

Undercooling is the prerequisite of solidification, and hence the undercoolability of liquid metals are of great significance on solidification. Although the classical nucleation theory is the basis of solidification theory, it has big limitation on predicting the maximum undercooling of liquid metals. In this project, we use intrinsic undercooling to denote the maximum undercoolability of liquid metals. The correlations of intrinsic undercooling with thermophysical properties and the liquid structure are investigated for the quantitative prediction of intrinsic undercooling. In previous works on undercooling prediction, the liquid/solid interfacial energy are often derived from experimental undercooling according to the classical nucleation theory. However, this 'fitting' process results in a correlation between liquid/solid interfacial energy and undercooling, and it is therefore not applicable to the quantitative prediction of intrinsic undercooling. In this project, we calculate undercooling and liquid/solid interfacial energy independently from the inter-atomic potential directly with molecular dynamics (MD) simulation. To elucidate the cooling effect on simulated undercooling, MD and phase field method are combined to investigate the homogeneous nucleation of liquid metals. Using electrostatic levitation, the maximum undercoolings of liquid metals are investigated, and both the density and volume expansion coefficient are measured. The melting enthalpy and liquidus temperature are measured by DSC. On one hand, the experimental results are used for comparison with simulation ones. On the other hand, the experimental results are applied directly to investigate the quantitative correlation of intrinsic undercooling with thermophysical properties. Local atomic structure is studied by MD. The atomic cluster and the five-fold symmetry are used to elucidate the correlation of intrinsic undercooling with liquid structure. The maximum undercooling of new alloys are predicted on the basis of the derived prediction formula of intrinsic undercooling, and electrostatic levitation experiment is used for validating the reasonability of the prediction. This project is important for the development of solidification theory and the novel materials.

过冷是凝固的前提，液态金属过冷能力对于凝固具有重要影响。经典形核理论是凝固理论的基础，然而它在预测液态金属最大过冷能力上存在很大局限性。项目利用本征过冷度表征最大过冷能力，在研究本征过冷度与液态金属热物理性质和结构定量耦合关系的基础上实现其定量预测。针对“过冷度反推法”间接获得的液/固界面能不适用于本征过冷度定量预测这一问题，项目利用分子动力学直接从势函数独立计算液/固界面能和过冷度。耦合分子动力学和相场模拟研究均质形核过程，澄清“急冷”对于模拟过冷度的影响。利用静电悬浮研究液态金属的最大过冷能力，实现密度和体积膨胀系数的测定，结合同步辐射研究液态金属的微观结构。利用分子动力学研究液态金属原子团簇类型及化学短程有序与本征过冷度的定量耦合关系。对新合金体系的最大过冷能力进行预测，并用静电悬浮实验验证本征过冷度定量预测公式的合理性。本项目的开展对于凝固理论的发展以及新材料的开发具有重要意义。

过冷是凝固的前提，液态金属的最大过冷能力对于凝固具有重要影响。项目围绕液态金属最大过冷能力的定量预测这一关键科学问题，利用实验和分子动力学模拟相结合的方法，对金属材料的过冷能力进行了研究。利用临界晶胚法对十二种纯金属的液/固界面能进行了模拟，并探讨了其与基本热物性参数之间的耦合关系。在液/固界面能研究的基础上，建立了液态金属本征过冷度与热物理性质的定量耦合关系，研究表明熔化熵对最大过冷能力具有重要影响。以纯金属Al为模型金属，探讨了由冷速引起的过冷能力、熔体及凝固结构的变化。对过冷液体中的类固态团簇进行了分析，揭示了冷速与过冷度相关性的结构根源。本项目的开展对于凝固理论的发展以及新材料的开发具有重要意义。