Investigation of the effects of extreme pressure and temperature on carbon materials

Carbon is a very interesting element due to its ability to form different types of bonds with both itself and other elements to form a wide variety of structures. The two most common crystalline phases of pure carbon are graphite and cubic diamond. However, there are several less common forms of carbon with their existence being controversial as well as the conditions required to form them. For example, evidence for a hexagonal form of diamond known as Lonsdaleite was published in the 1960s, however later evidence suggested that this is not a distinct phase of carbon but is just faulted diamond. In addition, the pressure-temperature diagram of carbon has been the subject of many studies, however, there are still many open questions concerning the exact pressure and temperature conditions required to form different phases of carbon. The goal of this dissertation was to help answer these unresolved questions by characterising the microstructure of pure carbon materials following treatment to high pressures and in some cases high temperatures.

In the first part of this thesis, the effects of high pressure treatment at room temperature on the structure and properties of the non-crystalline form of carbon known as glassy carbon (GC) were investigated. It was found that the characteristic isotropic graphitic microstructure of GC was resistant to pressures up to ~40 GPa. This is consistent with GC having superelastic properties in which it can recover from large deformations. At pressures of ~40 GPa, the structure was permanently changed into one that contained an anisotropic oriented graphitic microstructure and therefore this pressure represents the limit of GC's superelastic and non-graphitising properties. A small fraction of diamond-like bonds within the GC microstructure were found to be removed at this pressure, providing evidence that these bonds are responsible for helping maintain the GC microstructure. Following compressions above ~60 GPa at room temperature, the GC transformed into a transparent phase, providing evidence that diamond-like phases may have formed. However, after recovering the samples at ambient, the majority of the sample had reverted to the oriented graphitic phase. Detailed examination using electron microscopy based techniques revealed that the recovered material had bands or veins that contained hexagonal and cubic diamond. This was a remarkable finding as it showed that both forms of diamond could be made at room temperature, in contrast to the conventional wisdom that diamonds can only be formed at high temperatures. In addition, this finding does not support the proposition that hexagonal diamond is not a distinct phase of carbon. The likely cause for the formation of diamond phases at room temperature was found to be shear strain within the sample during compression.

The effects of both high pressures and high temperatures on GC were also investigated in this thesis using modulated laser heating.  By studying site specific regions of the microstructure of GC compressed to 16 GPa, it was found that ~2350 K was the minimum temperature required to form diamond, with small crystallite sizes of ~10 nm.  As the temperature increased up to ~3450 K, the size of the diamond crystals increased.  In contrast to the synthesis of diamond from a graphitic precursor using conventional heating in which there is usually a crystallographic orientation relationship between phases, the diamond crystals formed using laser heating were found to be randomly orientated. The random orientation of diamonds indicated that the diamond formed directly from the isotropic GC structure rather than in a two-step process in which GC transformed first into orientated graphite and then to diamond. At temperatures of ~3450 K and above, evidence was found that the GC had melted with a swelling of the surface and voids within the microstructure.

Further laser heating experiments on GC and other experiments on diamond indirectly heated using rhenium were used to provide information on the pressure-temperature (P-T) conditions required to melt carbon. Despite much conjecture in the literature, the results from this thesis do not support a positive melt slope for the P-T melting curve above the triple point at ~12 GPa. The implications for this finding are that the diamond stability field is limited with increasing pressure and other high density phases of carbon are likely.