Spark plasma sintering of ceramics: understanding temperature distribution enables more realistic comparison with conventional processing

Abstract

Spark plasma sintering (SPS) is a consolidation technique that can combine high heating and cooling rates with a uniaxial applied pressure resulting in short processing time. SPS has been successful in producing ceramics with novel microstructures, which are often reported to have been produced at temperatures lower than would be possible using conventional densification techniques. This has led many authors to infer the presence of additional densification mechanisms intrinsic to the SPS process. The present study considers the relationship between temperature measured in SPS and that experienced by the sample. A simple analytical model enables calculation of the equilibrium temperature distribution in a cylinder of known thermal conductivity and emissivity with constant heat generation per unit volume. Application of this model to the die assembly used in SPS allows estimation of the equilibrium sample temperature. Optimised processing conditions were identified for a fully dense submicrometre grain size alumina, and a SiN–40 vol.-%TiB2 composite. Both materials can only be processed to full density with optimised microstructures in a narrow temperature window. Using the analytical model, the temperature seen by the sample was calculated, and compared with the temperature required using conventional processing techniques. Observed microstructures were consistent with those obtained using conventional processing techniques when processed at the calculated temperatures, suggesting that when SPS is applied to ceramics, additional densification mechanisms intrinsic to the SPS process are not required to explain the observed microstructures. The dynamics of the temperature distribution and problems associated with producing large samples using SPS are also considered.