Modeling the stress-strain response and microstructure development of porous sintered steels

QC250507

Abstract

The work in this thesis deals with developing a modeling framework for the prediction of stress-strain behavior and microstructural changes in sintered metals containing porosity. Sintered steels are used in certain structural applications, for example, in the automotive industry. Mechanical behavior then becomes of importance, where compact density plays a major role in controlling properties. In addition, microscale features related to the shape and size of the pores also affect the stress-strain response. These features are, in turn, influenced by the sintering cycle. Thus, a mean-field diffusion-based sinteringmodel is employed together with a representative volume element (RVE)micromechanical model in an attempt to predict experimentally measured properties of a bainitic sintered steel.

Paper A presents a detailed characterization of uniaxial tensile and compressive behavior, based on compact density. Micro and macro hardness testing is carried out and compared with tensile results. Tests on samples of the same green density, subjected to different hold times and temperatures during the sinter cycle, are also performed. Effects of carbon content, pore structure and density on tensile behavior are discussed.

The sintering model is described and presented in Paper B. A new computational framework is introduced for the “two-particle” model, incorporating five different transport mechanisms. Density-dependence is introduced by relating particle overlap to the solid volume in a close-packed structure. Predicted microscopic shrinkage is compared to experimentally measured dimensional change of sintered tensile specimens for two different sinter cycles. A parametric study investigates the influence of different transportIImechanisms and particle size. The quality of fit for the model and reasons for experimentally observed differences between two cycles are discussed.

Micromechanical modeling is addressed in Papers C and D. In Paper C, the RVE is introduced in the form of close-packed overlapping spherical particles. Detailed motivation is given for how the model relates to compact density,microstructural features of the pores and particle size. Matrix parameters are reported, obtained by fitting the experimental tensile curve at one density using small-strain theory and von Mises plasticity. Results are then presented for simulations at five densities, where good agreement is shown between simulated and experimental curves. Lastly, a parametric study investigating the effects of sinter neck curvature is presented.

In Paper D, the RVE is augmented with the introduction of cohesive zones between particles to account for fracture behavior. Cohesive parameters are identified for a bi-linear traction-separation law that give good qualitative agreement between experimental and model results. The effect of varying the number of cohesive zones on the fracture response is investigated. Discussion focuses on further improvements to the model based on in-situ microstructural observations found in literature. In the introduction and conclusion section of the thesis, the proposed framework is discussed in the context of the integrated computational materialsengineering (ICME) approach and state of the art in sintering and porosity modeling. Avenues for further development to improve the predictive ability or extend the utility of the model are suggested in the outlook.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-363171