Material development has played a crucial role in modern civilization and IT. The importance of high-density and high-performance memory in modern computer systems and IT is ever increasing. This trend will be more obvious as computational architectures shift from being processing-centric to memory- (or data-) centric. The need for emerging and new memory technologies with nonvolatility and low power-consuming performance is rapidly increasing, while improvements in current dynamic random-access memory and NAND flash are being pursued. In both new and current memories, material innovation is of central importance. In this issue of MRS Bulletin, recent improvements in these two critical fields are reviewed with a focus on emerging and novel materials for the disruptive memory concept. Recent progress in scanning probe-based memory devices is also described.

Dynamic random-access memory (DRAM) is the main memory in most current computers. The excellent scalability of DRAM has significantly contributed to the development of modern computers. However, DRAM technology now faces critical challenges associated with further scaling toward the ∼10-nm technology node. This scaling will likely end soon because of the inherent limitations of charge-based memory. Much effort has been dedicated to delaying this. Novel cell architectures have been designed to reduce the cell area, and new materials and process technologies have been extensively investigated, especially for dielectrics and electrodes related to charge storage. In this article, the current issues, recent progress in and the future of DRAM materials, and fabrication technologies are discussed.

Ferroelectrics are promising for nonvolatile memories. However, the difficulty of fabricating ferroelectric layers and integrating them into complementary metal oxide semiconductor (CMOS) devices has hindered rapid scaling. Hafnium oxide is a standard material available in CMOS processes. Ferroelectricity in Si-doped hafnia was first reported in 2011, and this has revived interest in using ferroelectric memories for various applications. Ferroelectric hafnia with matured atomic layer deposition techniques is compatible with three-dimensional capacitors and can solve the scaling limitations in 1-transistor-1-capacitor (1T-1C) ferroelectric random-access memories (FeRAMs). For ferroelectric field-effect-transistors (FeFETs), the low permittivity and high coercive field Ec of hafnia ferroelectrics are beneficial. The much higher Ec of ferroelectric hafnia, however, makes high endurance a challenge. This article summarizes the current status of ferroelectricity in hafnia and explains how major issues of 1T-1C FeRAMs and FeFETs can be solved using this material system.

Nonvolatile memories (NVMs) are key devices in computers to save a user’s information. Besides flash memory, several types of NVMs that use magnetoresistance, resistance change of metal oxides, and phase change of chalcogenide alloys have been studied. Among these, phase-change random-access memory (PC-RAM) is competitive from the viewpoint of switching speed, high durability, and scalability. In 2017, Intel and Micron Technology shipped commercial devices named Optane that use a phase-change material as storage class memories. Condensed-matter physicists have recently been attracted to phase-change materials because of their functionality as topological insulators. If the topological phase state is controllable and applied to PC-RAM, electron spin transfer and storage effects will be further available in addition to electrical resistance switching.

Spin-transfer-torque magnetoresistive random-access memory (STT-MRAM) is an emerging nonvolatile memory that uses magnetic tunnel junctions (MTJs) to store information. MTJs with a crystalline MgO(001) tunnel barrier sandwiched between ferromagnetic layers, such as CoFeB, exhibit giant tunnel magnetoresistance, which is used to readout the STT-MRAM. Writing of STT-MRAM is based on current-induced magnetization reversal, called STT switching. STT-MRAM with perpendicular magnetization is especially important for high-density and low-power-consuming memory applications such as embedded memory for large-scale integrated circuit. For STT-MRAM to replace ultrahigh-density dynamic random-access memory, however, there are still technological challenges concerning the materials and fabrication processes of MTJs. This article reviews the physics and materials science of MTJs for STT-MRAM. We also discuss the importance of new MTJ materials and processes for next-generation ultrahigh-density MRAM.

Resistance switching random-access memory (ReRAM), with the ability to repeatedly modulate electrical resistance, has been highlighted as a feasible high-density memory with the potential to replace negative-AND flash memory. Such resistance modulation usually involves ion migration and filament formation, which usually lead to relatively low device reliability and yield. Resistance switching can also come from an entirely electronic origin, as in nanometallic memory, by electron trapping and detrapping. Recent research has revealed additional merits of its mechanism, which entails smart, atomic-sized floating gates that can be easily engineered in amorphous Si, oxides, and nitrides. This article addresses the basic ideas of nanometallic ReRAM, which may also be a contender for analogue computing and non-von Neumann-type computation.

High-density storage technology beyond hard disk drives and flash memory is required. Efforts are underway to develop new high-density storage technology based on scanning probe-based data storage. One of the candidates for scanning probe-type storage is thermomechanical data storage (also known as millipede, developed by IBM Zürich), and another is ferroelectric data storage. In this article, probe data-storage technologies are overviewed. Thermomechanical data storage and ferroelectric data storage are described in detail for next-generation high-density data-storage technology based on scanning probe microscopy. Ferroelectric data storage and scanning nonlinear dielectric microscopy-based and field-effect transistor-type probe-based probe data storage are also described.

It is curious that first-principles quantum simulations for establishing the electronic structure and bonding patterns of molecules and materials are conducted using fields, yet the standard theoretical approach to understanding their thermal behavior, phase transitions, and self-assembly on larger length- and time scales relies on classical force fields acting on particle degrees of freedom. This article discusses how equilibrium models of classical particle assemblies can be exactly reframed as statistical field theories, and how these theories can be numerically simulated. Today, such field-theoretic simulations have emerged as a highly efficient way to study phase transitions and self-assembly behavior in broad classes of soft materials, including block polymers, polyelectrolyte complexes, and polymeric emulsions.