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) show great promise to break this bottleneck by enabling nonvolatile memory devices that can optimize the complex memory hierarchy, and neuro-inspired computing devices that can unify computing with storage in memory cells. The articles in this issue of MRS Bulletin highlight recent breakthroughs in the fundamental materials science, as well as electronic and photonic implementations of these novel devices based on PCMs.

While fast-switching rewritable nonvolatile memory units based on phase-change materials (PCMs) are already in production at major technology companies such as Intel (16–64 GB chips are currently available), an in-depth understanding of the physical factors that determine their success is still lacking. Recently, we have argued for a liquid-phase metal-to-semiconductor transition (M-SC), located not far below the melting point, Tm, as essential. The M-SC is itself a consequence of atomic rearrangements that are involved in a fragile-to-strong viscosity transition that controls both the speed of crystallization and the stabilization of the semiconducting state. Here, we review past work and introduce a new parameter, the “metallicity” (inverse of the average Pauling electronegativity of a multicomponent alloy). When Tm-scaled temperatures of known M-SCs of Group IV, V, and VI alloys are plotted against their metallicities, the curvilinear plot leads directly to the composition zone of all known PCMs and the temperature interval below Tm, where the transition should occur. The metallicity concept could provide guidance for tailoring PCMs.

Phase-change materials (PCMs) have demonstrated a wide range of potential applications ranging from electronic memories to photonic devices. These applications are enabled by the unconventional portfolio of properties that characterizes crystalline PCMs. Here, we address the origin of these unusual properties and how they are related to the application potential of these materials. Evidence will be presented that the properties are related to an unconventional bonding mechanism. Employing a novel map, which separates solids according to the number of electrons transferred and shared between adjacent atoms, it is shown that PCMs occupy a well-defined region. Depicting physical properties such as the optical dielectric constant as the third dimension in the map reveals systematic property trends. Such trends can be utilized to unravel the origins of the unconventional materials properties or alternatively, as a means to optimize them.

The exploitation of phase-change materials (PCMs) in diverse technological applications can be greatly aided by a better understanding of the microscopic origins of their functional properties. Over the last decade, simulations based on electronic-structure calculations within density functional theory (DFT) have provided useful insights into the properties of PCMs. However, large simulation cells and long simulation times beyond the reach of DFT simulations are needed to address several key issues of relevance for the performance of devices. One way to overcome the limitations of DFT methods is to use machine learning (ML) techniques to build interatomic potentials for fast molecular dynamics simulations that still retain a quasi-ab initio accuracy. Here, we review the insights gained on the functional properties of the prototypical PCM GeTe by harnessing such interatomic potentials. Applications and future challenges of the ML techniques in the study of PCMs are also outlined.

The cycling endurance of phase-change memory is one of the last hurdles to overcome to enable its adoption in the larger market for persistent memory products. Phase-change memory cycling endurance failures, whether they are stuck-SET (caused by elemental segregation) or stuck-RESET (caused by void formation), are caused by atomic migration. Various driving forces responsible for the atomic migration have been identified, such as hole-wind force, electrostatic force, and crystallization-induced segregation. We introduce several strategies to improve cycling endurance based on an understanding of driving forces and interactions among them. Utilizing some of these endurance-improving techniques, record-high phase-change memory cycling endurance at around 1012 cycles has been recently reported using a confined phase-change memory cell with a metallic liner.

High-current switching performance of ovonic threshold switching (OTS) selectors have successfully enabled the commercialization of high-density three-dimensional (3D) stackable phase-change memory in Intel’s 3D Xpoint technology. This bridges the huge performance gap between dynamic random access memory (DRAM) and Flash. Similar to phase-change memory, OTS uses chalcogenide-based materials, but whereas phase-change memory reversibly switches between a high-resistance amorphous phase and a low-resistance crystalline phase, OTS freezes in the amorphous phase. In this article, we review recent developments in OTS materials and their performance in devices, especially current density and selectivity. Advantages and challenges of OTS devices in the integration with the phase-change memory are discussed. We introduce the evolution of theoretical models for explaining the OTS behavior, including thermal runaway, field-induced nucleation, and generation/recombination of charge carriers.

Driven by the rapid rise of silicon photonics, optical signaling is moving from the realm of long-distance communications to chip-to-chip, and even on-chip domains. If on-chip signaling becomes optical, we should consider what more we might do with light than just communicate. We might, for example, set goals for the storing and processing of information directly in the optical domain. Doing this might enable us to supplement, or even surpass, the performance of electronic processors, by exploiting the ultrahigh bandwidth and wavelength division multiplexing capabilities offered by optics. In this article, we show how, by using an integrated photonics platform that embeds chalcogenide phase-change materials into standard silicon photonics circuits, we can achieve some of these goals. Specifically, we show that a phase-change integrated photonics platform can deliver binary and multilevel memory, arithmetic and logic processing, as well as synaptic and neuronal mimics for use in neuromorphic, or brain-like, computing—all working directly in the optical domain.