© Department of Materials Science and Technology, IIT Delhi

A Three-scale Concurrent Method for Micro-crack Growth in Polycrystalline Alloys

Prof. Pritam Chakraborty

IIT Kanpur

Abstract

A significant portion of service life of components under high cycle fatigue condition is spent in crack nucleation, and micro-crack growth to a detectable size. In polycrystalline alloys, these phenomena are strongly dependent on microstructural features such as inclusion, grain size, orientation, etc. and their distribution, which cause scatter in total life obtained from experiments. Thus, accurate prediction of fatigue life of polycrystalline alloys for materials design and life extension requires development of microstructure dependent models for crack nucleation and micro-crack growth. In this study, a three-scale concurrent method is developed that can allow incorporation of a micro-crack of sub-grain size in standard specimens and model its growth in the polycrystal. In the method, a small region surrounding the micro-crack (region 1) is discretized with a very fine Finite Element (FE) mesh and Crystal Plasticity coupled Phase-field Fracture (CPPFF) model is developed to capture micro-crack growth across few grains. The domain surrounding region 1 is represented by Voronoi grains (region 2) and rationally quadratic polygonal elements with Crystal Plasticity (CP) constitutive law are utilized to capture the grain-level anisotropic elasto-plastic behavior. Such a representation of region 2 allows use of reasonably large polycrystalline microstructure around the micro-crack with significantly less computational cost. Finally, region 3 is represented by von Mises plasticity with flow stress obtained from homogenization of Representative Volume Element (RVE) response evaluated using CPFEM. The interfaces of the different regions are connected using constraint equations and the various models/methods such as CPPFF, CP with polygonal elements, etc. are verified to ensure accuracy. Finally, the global and local responses between concurrent two-scale and three-scale methods are compared to demonstrate the computational advantage and accuracy of the proposed approach.


Bio

Pritam Chakraborty is an Associate Professor in the Department of Aerospace Engineering and Adjunct Faculty in the Materials Science Program, at IIT Kanpur. Before joining IIT Kanpur as an Assistant Professor, he was a staff scientist at Idaho National Lab. USA, following his PhD from The Ohio State University, USA. Pritam’s interests are in computational solid mechanics, multi-scale modeling, integrated computational materials engineering, FEM, plasticity, crystal plasticity, fracture and fatigue. Recently he has also delved into combining simulations and digital image correlation for model development and calibration. Pritam has nearly forty publications in reputed journals related to solid mechanics and method development. He has also graduated several Masters and PhD students with contributions in different areas of mechanics, materials modeling and multi-scale method.


Abstract

A significant portion of service life of components under high cycle fatigue condition is spent in crack nucleation, and micro-crack growth to a detectable size. In polycrystalline alloys, these phenomena are strongly dependent on microstructural features such as inclusion, grain size, orientation, etc. and their distribution, which cause scatter in total life obtained from experiments. Thus, accurate prediction of fatigue life of polycrystalline alloys for materials design and life extension requires development of microstructure dependent models for crack nucleation and micro-crack growth. In this study, a three-scale concurrent method is developed that can allow incorporation of a micro-crack of sub-grain size in standard specimens and model its growth in the polycrystal. In the method, a small region surrounding the micro-crack (region 1) is discretized with a very fine Finite Element (FE) mesh and Crystal Plasticity coupled Phase-field Fracture (CPPFF) model is developed to capture micro-crack growth across few grains. The domain surrounding region 1 is represented by Voronoi grains (region 2) and rationally quadratic polygonal elements with Crystal Plasticity (CP) constitutive law are utilized to capture the grain-level anisotropic elasto-plastic behavior. Such a representation of region 2 allows use of reasonably large polycrystalline microstructure around the micro-crack with significantly less computational cost. Finally, region 3 is represented by von Mises plasticity with flow stress obtained from homogenization of Representative Volume Element (RVE) response evaluated using CPFEM. The interfaces of the different regions are connected using constraint equations and the various models/methods such as CPPFF, CP with polygonal elements, etc. are verified to ensure accuracy. Finally, the global and local responses between concurrent two-scale and three-scale methods are compared to demonstrate the computational advantage and accuracy of the proposed approach.


Bio

Pritam Chakraborty is an Associate Professor in the Department of Aerospace Engineering and Adjunct Faculty in the Materials Science Program, at IIT Kanpur. Before joining IIT Kanpur as an Assistant Professor, he was a staff scientist at Idaho National Lab. USA, following his PhD from The Ohio State University, USA. Pritam’s interests are in computational solid mechanics, multi-scale modeling, integrated computational materials engineering, FEM, plasticity, crystal plasticity, fracture and fatigue. Recently he has also delved into combining simulations and digital image correlation for model development and calibration. Pritam has nearly forty publications in reputed journals related to solid mechanics and method development. He has also graduated several Masters and PhD students with contributions in different areas of mechanics, materials modeling and multi-scale method.


Abstract

A significant portion of service life of components under high cycle fatigue condition is spent in crack nucleation, and micro-crack growth to a detectable size. In polycrystalline alloys, these phenomena are strongly dependent on microstructural features such as inclusion, grain size, orientation, etc. and their distribution, which cause scatter in total life obtained from experiments. Thus, accurate prediction of fatigue life of polycrystalline alloys for materials design and life extension requires development of microstructure dependent models for crack nucleation and micro-crack growth. In this study, a three-scale concurrent method is developed that can allow incorporation of a micro-crack of sub-grain size in standard specimens and model its growth in the polycrystal. In the method, a small region surrounding the micro-crack (region 1) is discretized with a very fine Finite Element (FE) mesh and Crystal Plasticity coupled Phase-field Fracture (CPPFF) model is developed to capture micro-crack growth across few grains. The domain surrounding region 1 is represented by Voronoi grains (region 2) and rationally quadratic polygonal elements with Crystal Plasticity (CP) constitutive law are utilized to capture the grain-level anisotropic elasto-plastic behavior. Such a representation of region 2 allows use of reasonably large polycrystalline microstructure around the micro-crack with significantly less computational cost. Finally, region 3 is represented by von Mises plasticity with flow stress obtained from homogenization of Representative Volume Element (RVE) response evaluated using CPFEM. The interfaces of the different regions are connected using constraint equations and the various models/methods such as CPPFF, CP with polygonal elements, etc. are verified to ensure accuracy. Finally, the global and local responses between concurrent two-scale and three-scale methods are compared to demonstrate the computational advantage and accuracy of the proposed approach.


Bio

Pritam Chakraborty is an Associate Professor in the Department of Aerospace Engineering and Adjunct Faculty in the Materials Science Program, at IIT Kanpur. Before joining IIT Kanpur as an Assistant Professor, he was a staff scientist at Idaho National Lab. USA, following his PhD from The Ohio State University, USA. Pritam’s interests are in computational solid mechanics, multi-scale modeling, integrated computational materials engineering, FEM, plasticity, crystal plasticity, fracture and fatigue. Recently he has also delved into combining simulations and digital image correlation for model development and calibration. Pritam has nearly forty publications in reputed journals related to solid mechanics and method development. He has also graduated several Masters and PhD students with contributions in different areas of mechanics, materials modeling and multi-scale method.


© Department of Materials Science and Engineering, IIT Delhi