A sintering-informed multiscale framework for predictive modeling of ductile-to-brittle transition in Cu-W composites

Read full text
Authors: Ruiping Luo, Bowen Dong, Kelei Wang, Liangyi Nie, Wenyan Luo, Wenting Guo, Jinchuan Jie, Zhenpeng Wu
Cited by: 0
References: 29

Abstract

The ductile-to-brittle transition (DBT) in Cu-W composites is governed by complex microstructural evolution during sintering, yet traditional modeling approaches often rely on idealized geometries that fail to capture process-induced features. In this work, a sintering-informed multiscale modeling framework is developed to predict the mechanical behavior of Cu-W composites across varying W mass fractions (60-90 wt.%). Molecular dynamics (MD) simulations are first employed to model the powder consolidation and Cu melt infiltration processes, generating physically realistic microstructures that reflect particle sintering, neck formation, and interfacial morphology. These atomistically informed configurations are then mapped onto continuum finite element models incorporating cohesive zone elements to simulate interfacial debonding, matrix plasticity, and particle fracture. The simulated stress-strain responses, strength trends, and fracture modes show excellent agreement with experimental data, capturing the transition from ductile necking in W60 to brittle-like failure in W90. Three key insights emerge: (1) the percolation of the W skeleton and emergence of bimodal particle size distributions constrain plastic flow and promote strain localization; (2) failure shifts from matrix-dominated to interface-dominated mechanisms with increasing W content; and (3) even without precise interfacial parameters, calibrating cohesive properties within physically reasonable ranges enables reliable prediction of untested compositions.

Fields of Study & Subjects

Powder Metallurgy Techniques and Materials Microstructure and mechanical properties Advanced materials and composites

Publication details

Journal / source
Journal of Materials Research and Technology
Volume
41
Pages
5288–5302
Publication date
12 February 2026
Language
EN