A review of memristor: material and structure design, device performance, applications and prospects

Abstract

With the booming growth of artificial intelligence (AI), the traditional von Neumann computing architecture based on complementary metal oxide semiconductor devices are facing memory wall and power wall. Memristor based in-memory computing can potentially overcome the current bottleneck of computer and achieve hardware breakthrough. In this review, the recent progress of memory devices in material and structure design, device performance and applications are summarized. Various resistive switching materials, including electrodes, binary oxides, perovskites, organics, and two-dimensional materials, are presented and their role in the memristor are discussed. Subsequently, the construction of shaped electrodes, the design of functional layer and other factors influencing the device performance are analyzed. We focus on the modulation of the resistances and the effective methods to enhance the performance. Furthermore, synaptic plasticity, optical-electrical properties, the fashionable applications in logic operation and analog calculation are introduced. Finally, some critical issues such as the resistive switching mechanism, multi-sensory fusion, system-level optimization are discussed.

Perspective: Zinc‐Tin Oxide Based Memristors for Sustainable and Flexible In‐Memory Computing Edge Devices

“…It is in these cases where large amounts of data movement are required to perform simple arithmetic operations, that memristive structures can show the best performance gain when compared to typical computing architectures. [ 8 ]…”

“…It is in these cases where large amounts of data movement are required to perform simple arithmetic operations, that memristive structures can show the best performance gain when compared to typical computing architectures. [8] Another area where RRAM devices can show great potential is in low-cost flexible/ transparent system on panel (SOP) applications. Amorphous oxide semiconductors (AOS), processed at moderate temperatures (<200 °C), are used in pixel driver thin-film transistors (TFTs) in commercially available large-area active-matrix flat-panel displays.…”

In Situ Unveiling of the Resistive Switching Mechanism of Halide Perovskite-Based Memristors

“…metal−insulator−metal, MIM). 1,2 The full-hardware artificial neural networks (e.g., convolutional neural networks, CNNs) and low-power artificial dendrites with computing function have been realized by memristors. 3,4 In the past decades, various kinds of materials have been explored as the functional layer of memristors, such as the twodimensional (2D) materials, 5−7 binary oxides, 8−10 binary nitrides, 11,12 organic materials, 13,14 halide perovskites, 15−18 etc.…”

“…Among them, the halide perovskite has been considered as a promising candidate for memristor demonstrations owing to the favorable advantages in outstanding hysteresis effect, fast ion migration, and solution processability. 1,19,20 To date, perovskite memristors have achieved large resistance switching ratio (ON/OFF ratio, the ratio of high resistance to low resistance at the same voltage) of 10 9 , 21 good cycle endurance of 10 6 , 22 long retention time of 4.2 × 10 7 seconds, 22 low power consumption of <5 × 10 −8 mW, 23 and multilevel storage capabilities, 24 which are comparable to or even better than rivals. 6,8,9 In addition, perovskite-based memristors have been applied in demonstrations of a logic gate circuit, 25−27 high density cross matrix memory, 22,28 and artificial synapse with neuromorphic computing.…”

“…The memristor (short for memory resistor) has been spotlighted because it has the potential to break the bottleneck due to its advantages in non-von Neumann architectures, large storage capacity, nonvolatile property, and simple structure (i.e. metal–insulator–metal, MIM). , The full-hardware artificial neural networks (e.g., convolutional neural networks, CNNs) and low-power artificial dendrites with computing function have been realized by memristors. , In the past decades, various kinds of materials have been explored as the functional layer of memristors, such as the two-dimensional (2D) materials, − binary oxides, − binary nitrides, , organic materials, , halide perovskites, − etc . Among them, the halide perovskite has been considered as a promising candidate for memristor demonstrations owing to the favorable advantages in outstanding hysteresis effect, fast ion migration, and solution processability. ,, To date, perovskite memristors have achieved large resistance switching ratio (ON/OFF ratio, the ratio of high resistance to low resistance at the same voltage) of 10 9 , good cycle endurance of 10 6 , long retention time of 4.2 × 10 7 seconds, low power consumption of <5 × 10 –8 mW, and multilevel storage capabilities, which are comparable to or even better than rivals. ,, In addition, perovskite-based memristors have been applied in demonstrations of a logic gate circuit, − high density cross matrix memory, , and artificial synapse with neuromorphic computing. − …”