Mechanism of thermal gradients in spark plasma sintering of titanium and the resultant graded microstructure: A novel strategy of incorporating master sintering curve into finite element modeling

Abstract

Spark plasma sintering (SPS) is an efficient consolidation process for metals and ceramics because of its high heating rates, lower sintering temperatures and shorter holding time than other sintering processes. To understand and utilize the thermal gradients that occur in SPS, the present work developed a strategy to incorporate master sintering curve (MSC) into a thermal-electric (TE) finite element modeling (FEM) to enable continuous nodal-specific simulation of the evolution of thermal and electrical properties of the powder compact with respect to both relative density and temperature during SPS. This combined model is proposed as an alternative to the thermal-electric-mechanical fully coupled model that is more computationally expensive and difficult to parameterize experimentally. Pure Ti was selected as the material for experimental verification because of its technological importance for aerospace and defense applications. While lab-scale Ti samples were sintered for construction and validation of MSC, a scaled-up Ti sample with diameter of 70 mm was made via SPS, and its microstructure were characterized to verify the accuracy of the MSC - TE FEM combined model. Both modeling and experimental results showed that the local relative density of the pure Ti decreased as the location moved radially from center to surface. The location-specific relative densities predicted by the combined model were within ±1.5% deviation from the experimental results, which verified the ability and accuracy of the MSC-TE FEM combined model to guide experimental design and parameter selection for control and utilization of gradients in SPS to achieve graded microstructure. • SPS of pure Ti has an apparent sintering activation energy of of 156.5 kJ/mol • Master sintering curve is incorporated into FEM for improved prediction of gradients • The accuracy of MSC-FEM combined model is experimentally validated • Deviation between the experimental and simulated results are <1.5%