Cemented carbide tools are used in applications demanding high hardness and toughness. However, fracture during operation of WC-Co tools is deleterious for tool performance and lifetimes. Interfaces in WC-Co, especially the WC grain boundary network are crucial for mechanical properties as crack path is mostly intergranular. Therefore, it is beneficial to gain knowledge about the WC boundaries. This will in turn allow for grain boundary engineering (GBE) of WC-Co tools with superior performance and lifetimes.
To achieve this, firstly, WC grain boundaries have been examined using wedge loaded double cantilever beam (DCB) fracture tests. Since the literature review of WC-Co revealed that Σ2 twist boundaries are the most abundant boundary type, these have been studied with DCB tests. Ideal Σ2 twist boundaries (perpendicular to the surface) for fracture testing have been identified from electron backscatter diffraction (EBSD) data acquired from a WC-10wt%Co sample and processed via a custom MATLAB script working in MTEX. The fracture energy of Σ2 twist boundaries has been compared to cleaving the lowest energy plane (type II {101 ̅0}) in a WC single crystal. Fracture energies of 3.59 ± 0.30 Jm-2 and 7.04 ± 0.36 Jm-2 were measured for cleaving Σ2 twist boundaries and type II {101 ̅0} planes respectively. This results in a few conclusions. Firstly, fracture energy measured for type II {101 ̅0} plane is in agreement with the values (7.25-7.33 Jm-2) reported from the literature using density functional theory calculations. Secondly, the fracture energy of Σ2 twist boundaries being half that of the type II {101 ̅0} planes is consistent with the observation that most of the fracture path in WC-Co is intergranular.
As a complementary study, grain boundary character distribution (GBCD) analysis of the most abundant boundaries (Σ2 boundaries and Σ13 boundaries) in WC-Co was performed. This analysis is based on Rohrer et al.’s five parameter GBCD analysis [1]. Furthermore, the effect of chromium doping on WC boundary populations of a chromium doped sample (WC-10wt%Co-1wt%Cr) was examined and compared with an undoped WC-Co (WC-10wt%Co) sample. It was found that the area fraction of Σ2 boundaries increased by a factor of 3 (from 5% to 15% area fraction) with chromium doping. Area fraction of Σ13 boundaries was not measured quantitively owing to low boundary counts. However, qualitative analysis indicated that Σ13 boundary population declined with doping. Finally, it was deduced that the fraction of Σ2 and Σ13 were more frequent when examining exclusively WC/WC boundaries compared to looking at all possible WC grain boundaries. This observation held true regardless of doping.
Interestingly, all further DCB tests performed at Σ2 twist boundaries on the chromium doped sample, WC-10wt%Co-1wt%Cr failed. This corresponds with the increased fraction of Σ2 twist boundaries on the chromium doped sample. It was thus hypothesised that chromium doping alters the boundary chemistry of Σ2 twist boundaries which in turn reduces the grain boundary energy (increasing the Σ2 twist boundary population) and thus increases the fracture toughness of the Σ2 twist boundary.