THERMODYNAMIC MODELING AND MECHANICAL PROPERTIES MODELING OF LONG PERIODIC STACKING ORDERED (LPSO) PHASES
Open Access
- Author:
- Kim, Hongyeun
- Graduate Program:
- Materials Science and Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- July 08, 2019
- Committee Members:
- Zi-Kui Liu, Dissertation Advisor/Co-Advisor
Zi-Kui Liu, Committee Chair/Co-Chair
Ismaila Dabo, Committee Member
Hojong Kim, Committee Member
Adrianus C Van Duin, Outside Member - Keywords:
- LPSO
Hardness model
Hall-Petch - Abstract:
- Recently, there has been an increasing interest in long periodic stacking ordered (LPSO) phases in Mg alloys due to their lightweight, high elastic and mechanical properties. The Vickers indentation hardness and Young’s modulus of LPSO phases have reached 470% and 140%, respectively, of that of pure Mg. Although theoretical and experimental studies have revealed the phase constitutions and crystal structures of LPSO phases including the formation of the noble solute atom clusters, which is also known as L12-type clusters, their phase stabilities and the origin of their enhanced mechanical properties are not yet solved. To further improve the properties and design the alloys, a thorough understanding of the phase equilibria and the origin of the mechanical properties of LPSO phases are therefore needed. In this dissertation, the elastic properties of LPSO phases in the Mg-Al-Gd system were studied using first-principles calculations. Since LPSO phases have been reported to enhance the strength and ductility of Mg alloys due to their high elastic properties, the effects of atomic arrangements in terms of Gd-Al L12-type clusters on LPSOs’ elastic properties in the Mg-Al-Gd system were studied using first-principles calculations. Four types of LPSO phases (10H, 18R, 14H, and 24R) were investigated with and without an interstitial atom in the center of the L12-type clusters. Furthermore, the calculated Poisson’s ratios of each LPSO phases from this study is also used as an important parameter for obtaining thermodynamic properties. Thermodynamic modelling of the four LPSO phases, i.e., 10H, 18R, 14H, and 24R, in the Mg-Al-Gd system was performed using the CALPHAD (calculation of phase diagram) approach with input from the present first-principles calculations and experimental data in the literature. Sublattice models were developed to describe these LPSO phases. Especially, an L12-type clusters in the FCC stacking layers of LPSO phases and the atomic occupancy in the center of L12 cluster were considered based on experimental observations and energetics from first-principles calculations. The calculated phase equilibrium results are in good agreement with experiments about the phase stability of 14H and 18R and the mole fraction of Gd and Al in these LPSO phases. The present modeling provides a new approach to describe the thermodynamic properties of LPSO phases that can be applied to other alloy systems. Material hardness is a good indicator of mechanical properties. However, since there is no hardness model that can be used for LPSO phases, a large portion of the effort in this dissertation is devoted to developing a suite of hardness models, which can be divided into three categories: hardness model for both brittle and ductile materials, temperature-dependent hardness model and hardness model for layered structures. In turn, the hardness of the LPSO phases is obtained/modeled, based on these hardness models that were developed. Hardness, defined as the resistance of a material to deformation, is a quick and efficient measure of mechanical performance of materials. However, to date no comprehensive predictive models exist for both metals and ceramics. We present a physics-based model that is capable of predicting Vickers indentation hardness of both brittle and ductile materials with model inputs from either first-principles calculations or experiments. Particularly, we go beyond the elastic properties of materials commonly used in the literature and introduce the plastic properties of materials in terms of active slip systems, including the Peierls-Nabarro flow stress, Burgers vector and slip plane spacing into the model. It is demonstrated that this model can predict hardness values from below 0.1 GPa of pure aluminum to above 100 GPa of diamond. The predictive power of the new model has the potential to significantly advance the computational discovery and design of new materials with enhanced performance. Furthermore, a new temperature dependent hardness model is also proposed based on the thermally activated dislocation width in combination with our previous Vickers hardness model. The thermally activated dislocation width, a basic building block for the temperature dependent Peierls-Nabarro flow stress in the hardness model, captures dislocation-diffusion mechanisms during the materials’ deformation. In the proposed model, the material hardness is determined by (a) diffusion mechanisms, (b) slip systems, (c) diffusing species, and (d) phase transformations. The model has been calibrated for and agrees well with experimental hot hardness results of 16 materials, which were available from the public domain, including metals and ceramics. The hardness model for layered structures is also modeled in order to investigate the origin of the Hall-Petch relation in structures with twinned, tilt and twist boundaries, especially, hardness enhancement of these structures based on material’s active slip systems of the structure as well as the elastic properties since the slip systems are crucial to understanding the deformation of materials. The active slip systems in this model are modulated by the relaxation of atomic positions near the boundaries. This proposed model explains the flow stress and the hardness changes as the twin or grain size in the structure changes, that is previously considered as an outcome of the Hall-Petch relation.