A computational study of the high-pressure high-temperature liquid phase sintering of polycrystalline diamond

Abstract

In light of mounting experimental evidence, the mechanism behind polycrystalline diamond (PCD) sintering has come under scrutiny. In this body of work, it was hypothesised that sintering requires the presence of non-diamond carbon in the reaction volume to achieve the level of sintering and the microstructures observed in experiment. Based around this hypothesis, there were two objectives. The primary objective was to use theories and simulation to investigate and challenge the currently accepted mechanisms behind the liquid phase sintering of PCD, and the secondary objective was to develop a usable computational model capable of predicting microstructure evolution. A variety of computational methods ranging from the nano- to the meso-scale were used to investigate the sintering process. At the smallest investigated scale, molecular dynamics was used to perform nano-scale high-pressure high-temperatrue (HPHT) di usion and sintering experiments. The factors in uencing carbon di usion and sintering on the molecular level were explored and the learning was applied to the meso-scale model. Two meso-scale methods were investigated. Due to the limitations encountered with the Monte Carlo approach, a new phase eld model was developed with the novel incorporation of elastic stresses in the inter-granular contact regions. It was found that the externally applied pressure resulted in densi cation rates correlating well with a newly developed theory. Upon validation with experimental work, it was found that elastic energy alone could not account for the discrepancy in diamond density between simulated and experimental microstructures. The subsequent implementation of a super-saturation algorithm to simulate the presence of non-diamond carbon helped bridge the gap. Qualitative and quantitative image analysis of experimental and simulated microstructures suggested that the hypothesis remains valid and thus challenges the long-standing theory that PCD sinters via ordinary liquid phase sintering. This seminal work has shown that further research in this eld is required to not only improve the computational models, but also to directly investigate the in uence of non-diamond carbon on PCD sintering both computationally and experimentally.