The rapidly growing demand for data storage and processing, driven by artificial intelligence (AI) and other data-intensive applications, is posing a serious challenge for current computing devices based on the von Neumann architecture. For every calculation, data sets need to be shuffled sequentially between the processor, and multiple memory and storage units through bandwidth-limited and energy-inefficient interconnects, typically causing 40% power wastage. Phase-change materials (PCMs) based on chalcogenides or antinomies show great promise to break this bottleneck by enabling non-volatile memory devices that can optimize the complex memory hierarchy, and neuro-inspired computing devices that can unify computing with storage in memory cells. The basic working principle is that PCMs can be switched between the amorphous and crystalline phase rapidly and reversibly by either electrical or optical pulses. The large contrast in electrical resistance and optical reflectivity between the two solid-state phases defines the logic state “0” and “1” for memory applications, while the continuous and non-linear change in resistance and reflectivity upon partial amorphization or gradual crystallization can be used to emulate neuronal dynamics for brain-like computing. In addition to traditional Ge-Sb-Te based alloys, metal oxides and two-dimensional materials, such as VO₂ and MoTe₂, have also been utilized for phase-change memory applications, where the switching between different logical states is achieved by transitions between different crystalline phases.

electrical properties nanostructure phase transformation physical vapor deposition (PVD) sputtering thin film