Structural and Dynamic Heterogeneities of Supercooled Liquids and Metallic Glasses


Student thesis: Doctoral Thesis

View graph of relations



Awarding Institution
Award date11 Jun 2018


When a melt is cooled fast enough to avoid crystallization, a metallic glass (MG) forms. In the glass transition process, the dynamics of supercooled liquid slows down dramatically and becomes increasingly heterogeneous. Nevertheless, the remarkable change of dynamics is not accompanied by obvious structural evolution. The structural mechanism of slow dynamics and dynamic heterogeneity is the hottest topic in glass science. Meanwhile, the disordered structure of MGs inherited from supercooled liquids gives rise to many unique properties. Therefore, unravelling the structural and dynamical characteristics of supercooled liquids and MGs is vital not only in fundamental science but also for various applications. In this thesis, we studied both supercooled liquids and MGs. Firstly, we studied the structure and dynamics and their correlations in supercooled metallic liquids comprehensively in second part. We characterized the structure from a new perspective based on configuration overlap rather than using simple geometrical segmentation methods. By carrying out large-scale molecular dynamics simulations, we found that it is not local geometrical orderings like icosahedra extracted from instantaneous configurations but the intrinsic correlation between static configurations that captures the structural origin governing slow dynamics. It is also revealed that dynamical heterogeneity is the consequence of slow dynamics rather than the origin. Part three shows how pressure influences the structure and dynamics of a ternary supercooled metallic liquid by computer simulations. The results demonstrated that isothermal compression could also trigger glass formation akin to conventional isobaric cooling. However, the structural evolution and dynamical properties including fragility are quite different in the two processes. The effects of pressure on dynamical heterogeneity were also thoroughly studied by high-order correlation functions. The relationship between timescales and length scales of dynamic heterogeneity was explored. Furthermore, the average dynamics, dynamic heterogeneities including the high-order dynamic correlation length, and static structure of the system were found to be well described by thermodynamic scaling with the same scaling exponent in fourth part, although the liquid itself is not strongly correlating thermodynamically. This indicates that the metallic liquid could be treated as a single-parameter liquid, which may make it easier to study amorphous materials in the future. Secondly, we aimed to investigate the properties of MGs. Although heterogeneity is commonly believed to be intrinsic to MGs, how to distinguish and characterize the heterogeneity at the atomic level is still debated. In part five, we proposed a new statistical method to characterize the heterogeneities in MGs. The structural features of the heterogeneities were also discussed. Since diverse local structures are crucial to create numerous active sites for high-efficiency catalysis, we then intended to explore new functional applications of MGs in electrocatalysis taking advantage of the inherent heterogeneities in part six. It is demonstrated that MGs would be promising for catalyzing water splitting by exhibiting superior efficiency and anomalous durability. This finding opens a new avenue to explore the applications of MGs.