Micro-Raman Mapping and Analysis of Indentation-Induced Phase Transformations in Germanium

Abstract

Although it has been confirmed by diamond anvil cell experiments that germanium transforms under hydrostatic pressure from the normal diamond cubic phase (Ge-I) to the metallic β-tin phase (Ge-II) and re-transforms to Ge-III (ST12 structure) or Ge-IV (BC8 structure) during release of the pressure, there are still controversies about whether the same transformations occur during nanoindentation. Here, we present new evidence of indentation-induced phase transformations in germanium. Nanoindentation experiments were performed on a (100) Ge single crystal using two triangular pyramidal indenters with different tip angles - the common Berkovich and the sharper cube-corner. Although the indentation load-displacement curves do not show any of the characteristics of phase transformation that are well-known for silicon, micro-Raman spectroscopy in conjunction with scanning electron microscopy reveals that phase transformations to amorphous and metastable crystalline phases do indeed occur. However, the transformations are observed reproducibly only for the cube-corner indenter.

Phase transformation pathways in amorphous germanium under indentation pressure

“…Using Raman mapping they showed that the extruded material 22 contained a-Ge, but there was also strong evidence for transformation of some of this phase to nanocrystalline dc-Ge. 32 Gogotsi and co-workers 12,21,23 and Oliver et al 30 also observed a-Ge around the residual indent area (extruded material) when phase transformation of dc-Ge took place under indentation. However, since the starting material was dc-Ge in all of these studies, from Raman spectra it was extremely difficult to distinguish the presence of (nanocrystalline) dc-Ge in the end phase from the starting dc-Ge phase.…”

“…Indeed, sharp indenters, high loads and/or fast loading rates were observed to favor deformation by phase transformation. [20][21][22][23][24] When using sharp indenter tips, Jang et al 22,32 noted that extensive extrusion of material outside of the contact area always accompanied a phase transformation and they suggested that this observation in itself can be used to infer that a phase transformation had occurred. Using Raman mapping they showed that the extruded material 22 contained a-Ge, but there was also strong evidence for transformation of some of this phase to nanocrystalline dc-Ge.…”

“…Indeed, the studies of Jang et al support this proposal since, from Raman mapping, they found crystalline Raman peaks corresponding to high pressure Ge phases in the center of the indented zone. 22,32 They tentatively interpreted one of the phases as bc8-Ge although they noted that the Raman peaks were not well assigned in the literature. They showed that this phase was unstable and appeared to transform to heavily strained dc-Ge since the residual Raman peak was significantly shifted to lower wave number compared to the starting dc-Ge peak at 301 cm À1 .…”