报告题目：Multiscale Modeling and Simulations–Application to Hydrogen
Embrittlement in Metals
地点：力学一楼二层多媒体会议室
Embrittlement in Metals
报告人：宋俊教授
时间：2011年12月30日上午10：00 地点：力学一楼二层多媒体会议室
2003年毕业于中国科学技术大学近代力学系，后赴美国普林斯顿大学大学深造，获得硕士和博士学位，并在布朗大学完成博士后研究。现为加拿大麦吉尔大学采矿与材料工程系助理教授。
报告摘要：
Physical phenomena at different scales necessitate different computational techniques to provide suitable and accurate descriptions. Concurrent multiscale modeling framework allows real-time exchange across different scales so as to directly weld them into a unified framework, and thus is a very powerful tool in computational materials science. In this seminar, the application of a multiscale, coupled atomistics and continuum computational framework, to the problem of hydrogen embrittlement in metals will be discussed. The embrittlement of metallic systems by hydrogen is a wide spread phenomenon but the precise role of hydrogen in this process is not well understood and predictive mechanisms are not available. Here, a new model is proposed wherein hydrogen accumulation around a microcrack tip prevents crack-tip dislocation emission/absorption, and thus suppresses crack tip blunting and ductile fracture while promoting cleavage fracture. Direct simulations are performed to examine the evolution of equilibrium hydrogen distributions around a crack tip under increasing applied load, followed by measurement of dislocation emission and/or cleavage. Our conceptual models are well demonstrated through these imulations.
Accompanying the atomistic simulations, a kinetic analysis is used to calculate the size of the crack-tip hydrogen cloud as a function of hydrogen chemical potential, temperature, hydrogen diffusion rate, load level, and loading rate. Combining the kinetic analysis with the deformation/fracture analysis generates a mechanism map that predicts a ductile to brittle transition as a function of material and loading parameters. The mechanism map is applied to predict hydrogen embrittlement of unnotched tensile metal specimens, with the predictions and experiments matching well. This suggests the underlying physics and mechanics are well captured by our model.
In this talk, I will also give a short tour of our programs in the Department of Materials Engineering and more broadly, Faculty of Engineering at McGill University. In addition the current research thrusts and strategic directions will also be briefly touched.
报告摘要：
Physical phenomena at different scales necessitate different computational techniques to provide suitable and accurate descriptions. Concurrent multiscale modeling framework allows real-time exchange across different scales so as to directly weld them into a unified framework, and thus is a very powerful tool in computational materials science. In this seminar, the application of a multiscale, coupled atomistics and continuum computational framework, to the problem of hydrogen embrittlement in metals will be discussed. The embrittlement of metallic systems by hydrogen is a wide spread phenomenon but the precise role of hydrogen in this process is not well understood and predictive mechanisms are not available. Here, a new model is proposed wherein hydrogen accumulation around a microcrack tip prevents crack-tip dislocation emission/absorption, and thus suppresses crack tip blunting and ductile fracture while promoting cleavage fracture. Direct simulations are performed to examine the evolution of equilibrium hydrogen distributions around a crack tip under increasing applied load, followed by measurement of dislocation emission and/or cleavage. Our conceptual models are well demonstrated through these imulations.
Accompanying the atomistic simulations, a kinetic analysis is used to calculate the size of the crack-tip hydrogen cloud as a function of hydrogen chemical potential, temperature, hydrogen diffusion rate, load level, and loading rate. Combining the kinetic analysis with the deformation/fracture analysis generates a mechanism map that predicts a ductile to brittle transition as a function of material and loading parameters. The mechanism map is applied to predict hydrogen embrittlement of unnotched tensile metal specimens, with the predictions and experiments matching well. This suggests the underlying physics and mechanics are well captured by our model.
In this talk, I will also give a short tour of our programs in the Department of Materials Engineering and more broadly, Faculty of Engineering at McGill University. In addition the current research thrusts and strategic directions will also be briefly touched.