Quantifying stress and strain in diamond

Birefringence in diamond is an anomalous optical property for a nominally isotropic material. The occurrence of birefringence in diamond is a result of the photoelastic effect; the change of refractive index caused by stress (Nye, 1957). With the development of the MetriPol system (Glazer et al., 1996), a new technique is now available that allows rapid and accurate measurement of birefringence. Detailed knowledge of the photoelastic effect in diamond is vital to be able to use this new quantitative birefringence analysis with confidence. Initial investigation of the current literature on the photoelastic constants of diamond showed some confusion between the results of different groups. So to confirm the values believed to be the most accurate, a new technique was developed during the course of this study that allows a quantifiable stress to be applied to a crystal sample while it is being viewed under a microscope. This has allowed for an investigation of the photoelastic constants of diamond with the MetriPol system. The data produced is within ± 15% of the results of Grimsditch et al. (1979) and has confirmed the signs of the photoelastic constants, as well as the fact that the refractive index of diamond decreases with increasing hydrostatic pressure. Lang (1967) postulated five causes of stress and strain within natural diamond. These being lattice parameter variations, dislocations, fractures, inclusions and plastic deformation. Preliminary investigation of each of these five causes in this study highlighted the importance of understanding the geometry of the stress field when using quantitative birefringence analysis. This lead to further investigation of the birefringence halos that form around mineral inclusions that remain under remnant pressure. Application of the MetriPol analysis to measuring the remnant pressure required a theoretical model to be produced to understand the relationship between the measured value of retardation and the peak level of anisotropy. The birefringence analysis and theoretical model were initially applied to garnet inclusions in diamonds from Udachnaya, Siberia. While this gave a suitable result for the remnant pressure and calculated source pressure and temperature conditions, it could not be verified by another other analytical technique. Verification of the model was obtained by measurements on a coesite inclusion in a diamond plate from Finsch, South Africa, using various Raman techniques (including 2D mapping and point analysis). However, when the source pressure and temperature conditions were calculated from the remnant pressure, the results were well below the diamond and coesite stability fields. This calls into question the suitability of the coesite-in-diamond geobarometer, first proposed by Sobolev et al. (2000). A series of high pressure high temperature (HPHT) deformation experiments were performed on a set of diamonds, to investigate the association of brown colour with plastic deformation. With the use of a deformation DIA type press, a uniaxial stress was applied to diamonds that were already under HPHT conditions. In two tests where the uniaxial stress was not introduced there was no colour change in the sample. In four tests where the uniaxial stress was introduced, the colour of the samples changed from either colourless or yellow, to brown