Squeezing Out Voids in Phase-Change Materials
The emerging non-volatile phase-change memory holds the potential for the next-generation data storage. As a prototype of the phase-change materials, the chalcogenide Ge2Sb2Te5 (GST225, abbreviated as GST hereafter) alloy has been successfully used in optical discs and random-access memories, exploiting the fast phase transition and the large optical (reflectivity) or electrical (resistivity) contrast between its amorphous (a-GST) and crystalline (c-GST) polymorphs. However, the physics underlying the large property contrast (especially the resistivity contrast) is still unknown. In our recent study, we have demonstrated such a large resistivity difference in the a-GST amorphous state, generated and observed using the thermodynamic variable pressure as a “knob to turn.”
In situ experiments applying increasing pressure (0 to ~6 GPa) demonstrate a dramatic electrical resistivity decrease of approximately four orders of magnitude within the a-GST. The resulting property contrast in the a-GST itself, in the absence of a phase change in this case, is in fact comparable in magnitude to that found upon crystallization of a-GST to c-GST. Ab initio molecular dynamics computations reveal that in this pressure regime, the most prominent and representative structural evolution accompanying the density increase is the decrease of the volume fraction of low-electron-density (LED) regions. This structural signature is much easier to identify than other subtle structural and bonding changes, and thus serves as a useful indicator of all the enhanced chemical interactions inside the amorphous alloy that change the resistivity. The variation of electrical conductivity arising from the pressure-induced removal of LED regions was rationalized using ab initio calculations, which reveal a reduction in both the energy band gap and electron localization. Our findings offer new perspectives for monitoring and explaining the large resistivity changes in phase change alloys.