Fact-checked by Grok 2 weeks ago

Non-volatile random-access memory

Non-volatile random-access memory (NVRAM) is a class of that allows to locations while retaining stored information without the need for continuous electrical power, in contrast to volatile memories like (DRAM) that lose upon power loss. This persistence makes NVRAM suitable for applications requiring both speed and durability, such as system configuration storage, embedded devices, and boot acceleration. Key technologies underpinning NVRAM include ferroelectric random-access memory (FeRAM), which uses ferroelectric materials for fast read/write operations and high up to 10^14 cycles; magnetoresistive random-access memory (MRAM), leveraging magnetic states for non-destructive reads, low power consumption, and immunity to radiation; phase-change random-access memory (PCRAM or PCM), based on phase transitions in chalcogenide materials for high density and scalability; and (RRAM or ReRAM), employing resistance changes in oxide layers for simple structures, fast switching, and potential integration. These technologies address limitations of traditional non-volatile storage like NAND flash, which suffers from slower write speeds and issues, by approaching DRAM-like performance while maintaining data retention. Earlier forms, such as battery-backed static (SRAM) and electrically erasable programmable read-only memory (), laid the groundwork but are limited by size, cost, or power dependencies. NVRAM plays a critical role in modern architectures, enabling storage-class memory (SCM) that bridges the performance gap between volatile main memory and slower non-volatile , with latencies closer to (e.g., MRAM at sub-100 ns) compared to NAND flash (around 80-120 µs). Its adoption is driven by demands for , higher densities, and reliability in sectors like mobile devices, automotive systems, and data centers. Challenges persist in areas like fabrication costs, thermal stability, and variability, but ongoing advancements in materials and integration continue to enhance its viability. As of 2025, emerging applications in highlight NVRAM's potential for brain-inspired hardware, with resistive and magnetic variants supporting synaptic weights and low-power inference.

Fundamentals

Definition and Characteristics

Non-volatile random-access memory (NVRAM) is a type of that retains stored data even in the absence of power, while providing random byte-level read and write access similar to traditional . This persistence distinguishes it from volatile memories like and , which lose data upon power loss, though NVRAM serves as a counterpart offering at the cost of potentially slower operations. Key characteristics of NVRAM include its non-volatility, enabling typical exceeding 10 years without power. It supports without sequential constraints, allowing direct addressing of bytes or words. Read speeds generally range from 10 to 100 depending on the , while write operations can be slower due to the physics of . Endurance varies widely across technologies, from about 10^3 write cycles in some forms to up to 10^14 in advanced ferroelectric or magnetic types, reflecting trade-offs in . Modern NVRAM achieves densities up to around 64 Gb/cm², with projections exceeding 1 Tb/cm² through advanced scaling techniques such as 3D integration; as of 2025, ongoing scaling has pushed production densities toward 100 Gb/cm² in select MRAM and ReRAM devices, per industry roadmaps. Power consumption features low standby requirements since no energy is needed for retention, but write operations demand higher energy compared to reads. NVRAM finds general applications in embedded systems for reliable configuration storage, as write caches to accelerate data handling, and within storage hierarchies of computing devices to bridge the gap between fast volatile memory and slower persistent storage. Unlike read-only memory (ROM), which is mask-programmed during manufacturing and cannot be altered post-fabrication, NVRAM permits electrical reprogramming for flexible data updates.

Comparison to Volatile Memory

Volatile memory, such as (SRAM) and (DRAM), serves as the primary working memory in systems due to its rapid access times, but it requires continuous power to retain data. SRAM operates without the need for periodic refreshing, offering access latencies as low as 1 ns for reads and writes, making it ideal for high-speed applications like CPU caches, though it consumes relatively high static power and achieves lower densities (cell sizes of 40–500 F²). In contrast, DRAM provides higher density (cell sizes of 6–10 F²) and lower cost per bit but necessitates regular refresh cycles every few milliseconds to prevent , resulting in ongoing power draw of approximately 440 mW/TB and complete data volatility upon power interruption. Non-volatile random-access memory (NVRAM) differs fundamentally by retaining data without power, eliminating refresh overhead and enabling near-zero consumption, which enhances for long-term roles. However, NVRAM typically exhibits slower write speeds compared to , with latencies ranging from tens of nanoseconds (e.g., in MRAM) to microseconds or milliseconds in other types, versus the sub-10 ns writes of and ; read speeds in NVRAM can approach volatile levels (3–50 ns) but often lag in write-intensive scenarios due to constraints and material physics. This performance trade-off positions NVRAM as slower for frequent random accesses but advantageous in density potential, where it can rival or exceed in scalable architectures for storage-class applications, avoiding the heat and power limits that constrain volatile scaling. In terms of cost and scalability, remains more economical for high-speed needs, with at around $10/GB (as of 2023), while SRAM's complexity drives higher costs; NVRAM, though initially higher—often 5–10 times more expensive per bit than due to fabrication challenges—offers long-term savings through that reduces requirements and enables denser in structures. Scalability favors NVRAM for future high-capacity systems, as its non-volatility supports endurance cycles from 10⁶ to 10¹⁴ (far exceeding 's unlimited but power-dependent cycles) and allows stacking beyond the thermal limits of volatile alternatives. Use cases highlight these trade-offs: and dominate as main system memory and caches in processors for their speed in active computing, where data loss on power-off is tolerable with backups. NVRAM excels in persistent scenarios, such as storing boot code, configuration data, or serving as hybrid persistent memory in data centers, where its retention (often years without power) and low idle power reduce system complexity and enable instant-on capabilities without reloading from secondary storage. Energy efficiency further underscores NVRAM's niche, with standby power near zero compared to volatile memory's constant draw for retention—DRAM's refresh alone accounts for significant idle consumption, while NVRAM's non-volatility yields up to 30 times lower power in storage hierarchies (e.g., 14 mW/TB for comparable non-volatile systems).
AspectVolatile Memory (SRAM/DRAM)NVRAM
Access SpeedSRAM: <1–100 ns read/write; DRAM: 1–30 ns read/write3–50 ns read; 10 ns–ms write (varies by type)
Power (Standby)High (DRAM refresh: ~440 mW/TB; SRAM static draw)Near zero (no refresh needed)
Cost per BitDRAM: ~$10/GB (as of 2023); SRAM: higherInitially higher than DRAM (often 5–10x), but decreasing
DensityDRAM: 6–10 F²/cell; SRAM: 40–500 F²/cellComparable to DRAM, scalable to higher
Use CasesCPU caches, main memoryPersistent storage, hybrid memory
Sources: Metrics derived from IRDS 2023 MDS report. Trade-offs informed by surveys on NVMM technologies.

Historical Development

Early Concepts

The earliest form of non-volatile random-access memory (NVRAM) emerged in the late 1940s and early 1950s with the development of , which utilized tiny ferrite rings to store data magnetically and retain it without power. This technology was pioneered as part of the at MIT, marking the first practical implementation of non-volatile RAM in 1953 for flight simulation systems. Core memory operated by magnetizing the rings to represent binary states, allowing random access through wired threading, and it became the dominant memory type in computers from the mid-1950s to the early 1970s due to its reliability in harsh environments. A notable application of core memory occurred in the 1960s with the , where it served as both erasable RAM (2,048 words) and read-only core rope memory (up to 36,864 words), with the latter hand-woven by technicians to encode fixed programs for space missions. The manual wire-threading process, often performed by skilled women assemblers, ensured data persistence critical for aerospace operations where power interruptions could not risk data loss. This mechanical construction highlighted core memory's role in high-stakes reliability, as seen in the in 1969. Despite its advantages, core memory suffered from significant limitations, including low storage density—typically limited to kilobits per module—and high production costs, initially around $1 per bit due to labor-intensive weaving. These drawbacks, combined with the bulky mechanical nature requiring precise manual assembly, drove the shift toward semiconductor-based alternatives in the 1960s. Early proposals for non-volatile semiconductor memory focused on charge storage mechanisms inspired by , aiming to integrate persistence directly into silicon structures. A pivotal advancement came in 1967 when Dawon Kahng and Simon M. Sze at Bell Labs proposed the , a theoretical device that trapped charge on an isolated gate within the transistor to enable non-volatile data retention. This concept, detailed in their seminal paper, laid the groundwork for future semiconductor by addressing the volatility of standard MOS memory. Initial applications of these early non-volatile technologies were predominantly in aerospace and military sectors, where the need for power-independent data retention outweighed the high costs and low densities.

Key Milestones

The 1970s marked the beginning of practical non-volatile semiconductor memory with the introduction of ultraviolet-erasable in 1971 by Intel, developed by Dov Frohman using floating-gate avalanche injection for data storage and UV light for erasure. This was followed in 1978 by Intel's demonstration of the first , such as the 2816 device, which enabled byte-level electrical erasure and reprogramming without external equipment, significantly improving usability over . In the 1980s, breakthroughs in flash memory transformed non-volatile storage density and speed. Fujio Masuoka at invented flash memory in 1984, proposing the NOR flash architecture that allowed block-level electrical erasure while maintaining random access. By 1987, Masuoka introduced NAND flash at , designed for higher-density sequential access, enabling cost-effective mass storage applications. The 1990s saw commercialization of alternative non-volatile RAM types. Ramtron achieved the first commercial FRAM products in 1993, leveraging ferroelectric materials for fast read/write speeds comparable to with non-volatility. Earlier, in 1984, researchers at Honeywell demonstrated the first prototypes using magnetoresistive elements for non-volatile storage. In 1996, IBM researcher John Slonczewski invented spin-transfer-torque switching, a foundational mechanism for modern technologies. Additionally, scaling advancements included the adoption of multi-bit cells in , such as SanDisk's early multi-level cell (MLC) demonstrations for in 1996 and Intel's 2-bit-per-cell StrataFlash NOR in 1997, with the first 2-bit MLC introduced by Toshiba and SanDisk in 2001. Entering the early 2000s, phase-change RAM (PCRAM) experienced a revival through a 2000 collaboration between and , which licensed the technology for scalable, high-endurance non-volatile applications using chalcogenide materials. JEDEC began standardizing NVRAM interfaces in the 2000s, including specifications for embedded and serial non-volatile memories that supported broader system integration. These developments drove a shift from niche military and industrial uses to ubiquitous adoption in consumer electronics, powering devices like digital cameras, USB drives, and mobile phones by the mid-2000s.

Core Operating Principles

Floating-Gate Mechanism

The floating-gate mechanism relies on a modified metal-oxide-semiconductor field-effect transistor (MOSFET) structure, where an additional conductive layer, known as the floating gate, is inserted between the control gate and the channel, isolated by thin oxide layers. This floating gate, typically made of polysilicon, is electrically isolated and surrounded by silicon dioxide, allowing it to store charge without a direct electrical connection. The stored charge on the floating gate modulates the threshold voltage V_{th} of the underlying MOSFET, enabling non-volatile data retention by shifting the transistor's turn-on characteristics. Programming and erasing operations in the floating-gate MOSFET involve Fowler-Nordheim tunneling, a quantum mechanical process where electrons are injected into or removed from the floating gate through the thin oxide barrier under high electric fields (typically 8-10 MV/cm). During writing, a positive voltage applied to the control gate creates a strong field across the tunnel oxide, causing electrons to tunnel from the substrate or channel into the floating gate, increasing its negative charge and raising V_{th}. Erasing reverses this by applying a negative voltage, tunneling electrons out of the floating gate, which lowers V_{th}. The tunneling current density J is described by the Fowler-Nordheim equation: J = \frac{q^2 E^2}{8 \pi h \phi} \exp\left( -\frac{8 \pi \sqrt{2 m \phi^3}}{3 q h E} \right) where q is the electron charge, E is the electric field across the oxide, h is Planck's constant, \phi is the barrier height (approximately 3.1 eV for Si-SiO₂), and m is the electron effective mass. Reading the stored state occurs non-destructively by applying a standard gate voltage and sensing the resulting drain current, which reflects the V_{th} shift caused by the trapped charge; a higher V_{th} (charged state) results in lower current, representing one logic level, while a lower V_{th} (erased state) allows higher current for the opposite level. Charge retention is achieved through the isolation of the floating gate, with the oxide preventing leakage over extended periods (often >10 years at room temperature), as the trapped electrons remain stable without power. Despite its effectiveness, the floating-gate mechanism faces limitations from oxide degradation due to high during repeated tunneling, leading to trap generation and reduced , typically around $10^5 program/erase cycles before significant V_{th} window closure. Scaling below 10 nm exacerbates issues, as thinner oxides increase stress-induced leakage (SILC), compromising retention and reliability. To programming and size, variants like the split-gate structure incorporate a divided control gate, where one section overlaps the floating gate more selectively, reducing required voltages and oxide stress while improving coupling .

Alternative Storage Mechanisms

In contrast to charge-trapping mechanisms like floating-gate transistors, alternative storage principles in non-volatile random-access memory (NVRAM) rely on diverse physical phenomena to achieve bistable states without relying on stored electrical charge. These approaches leverage material properties such as , , phase transitions, or to retain data, offering potential advantages in , speed, or endurance while addressing limitations in charge leakage or . Ferroelectric storage exploits the spontaneous in materials like (PZT), where domains of aligned dipoles create bistable "up" and "down" states representing . Switching between these states occurs by applying an that reorients the dipoles through motion or , enabling non-destructive readout via the resulting charge. These devices demonstrate exceptional endurance, often exceeding 10^{12} cycles, due to the robust reversible nature of reversal. Magnetic storage, as in spin-transfer torque (STT-MRAM), encodes data through the relative orientation of ferromagnetic layers separated by a thin , altering via the tunnel (TMR) effect. In the parallel state, low allows current flow, while the antiparallel state yields high ; data writing involves spin-polarized current injecting torque to flip the free layer's . High-performance devices achieve TMR ratios greater than 100%, enabling reliable state discrimination even at nanoscale dimensions. Phase-change storage utilizes the reversible amorphous-to-crystalline in chalcogenide alloys such as GeSbTe, where the amorphous exhibits high resistivity and the crystalline low resistivity, defining the two states. Writing to the amorphous () state employs a melt-quench process: rapid heating to melting followed by fast cooling to suppress , while the crystalline (set) state forms via controlled annealing below the . This structural reconfiguration provides thermal stability for over decades at elevated temperatures. Resistive storage mechanisms in devices like resistive RAM (ReRAM) involve or conductive formation within layers, such as HfO2, to toggle between high- and low- states. In the low-resistance state, oxygen vacancies or metal s migrate under an applied field to form a localized conductive bridging electrodes; reversing the field dissolves the , restoring high . This change or electrochemical process enables compact, multilayer stacking but requires precise control to maintain state uniformity. Despite their promise, these alternative mechanisms face common challenges, including cycle-to-cycle variability in switching thresholds due to filament formation or nucleation, and stability issues where unintended changes or can degrade retention over time. Addressing these requires to minimize defects and enhance uniformity.

Established Commercial Types

Flash Memory

Flash memory is the most widely adopted form of non-volatile random-access memory (NVRAM), leveraging floating-gate transistors to store charge for without power. It dominates commercial applications due to its scalability, cost-effectiveness, and integration into devices like solid-state drives (SSDs) and embedded . Developed initially at , flash memory evolved into two primary architectures—NOR and —that balance access speed, density, and use cases, with emerging as the high-volume leader for data-centric . NOR flash, introduced by Fujio Masuoka and colleagues at in 1984, enables random access to individual bytes or words, making it suitable for code execution directly from (XIP) in applications like and embedded systems. Its architecture connects cells in parallel to bit lines, allowing fast reads comparable to but requiring block-level erases for reprogramming. In contrast, NAND flash, proposed by the same team in 1987, organizes cells in series within strings (typically 32–128 cells per string), enabling higher density through compact array structures ideal for bulk in SSDs and memory cards. NAND prioritizes block-oriented access, where is read or programmed in pages (e.g., 4–16 KB) but erased in larger blocks (e.g., 128–512 pages), trading random write flexibility for superior areal density. Operationally, flash memory relies on Fowler-Nordheim tunneling to inject or remove electrons from the floating gate, with programming at the page level and erasure at the block level to reset cells to a uniform state. To boost density without proportional area increases, (MLC) variants store 2 bits per cell using four voltage thresholds, while triple-level cell () and quad-level cell (QLC) encode 3 and 4 bits, respectively, via finer voltage distinctions that enhance capacity but raise error rates. Further scaling occurs through 3D NAND stacking, where memory layers are vertically integrated; by late , commercial implementations reach over 400 layers in Samsung's V-NAND and 276 layers in Micron's offerings, enabling terabit-scale dies while mitigating planar scaling limits. Performance characteristics include typical page read latencies of around 50 μs, program times of 500 μs to 1 ms per page, and block erase durations of 1–3 ms, with these values remaining stable across generations despite density gains. Endurance varies by cell type: SLC withstands up to 100,000 program/erase (P/E) cycles, 3,000–10,000, TLC about 1,000–3,000, and QLC 100–1,000, necessitating error correction codes () like BCH or LDPC to maintain reliability amid read disturbs and charge leakage. In the market, flash accounts for over 90% of NVRAM deployment in and SSDs as well as UFS/eMMC modules in smartphones, powering data-intensive ecosystems from to . As of Q3 2025, leading manufacturers include (approximately 35% share), followed by (around 25%) and Micron (about 18%), with single-die capacities exceeding 2 terabits in QLC configurations. Key challenges include limited P/E endurance, addressed via wear-leveling algorithms that redistribute writes across blocks to prevent premature failure, and degradation, where stored charge leaks over time—typically rated for 10 years at 55°C but diminishing with cumulative cycles and elevated temperatures.

Ferroelectric RAM

(FRAM), also known as FeRAM, is a technology that utilizes the ferroelectric effect in materials to store data as states in a . The fundamental storage employs a 1T-1C , consisting of one access and one ferroelectric , where the typically incorporates (PZT) or similar materials to enable bistable . These states—representing logic '0' and '1'—retain data without power due to the material's hysteresis loop, allowing non-volatility akin to while mimicking speed. In operation, writing occurs by applying an across the ferroelectric to reverse its direction, directly overwriting data without a separate erase step. This process requires low voltages of 1-3 V and completes in less than 100 ns, enabling rapid state changes comparable to dynamic . Reading is destructive, as sensing the polarization induces a charge that flips the state; however, a detects the output and immediately restores the original polarization, making the operation effectively non-destructive from the user's perspective with no net . FRAM exhibits high performance suited for data-intensive applications, with read and write speeds around 50-90 ns, far exceeding memory's millisecond-scale operations. Endurance surpasses 10^{12} cycles, limited primarily by material fatigue rather than wear mechanisms in other non-volatile types, while operating at low voltages (1-3 V) contributes to power efficiency below 5 mW during active cycles. Commercially available densities reach several megabits, though scaling to gigabit levels remains challenging as of due to the physical size requirements of ferroelectric capacitors, which hinder aggressive planar shrinkage. Commercial products emerged in the late , pioneered by Ramtron International (acquired by in 2012 and subsequently by Infineon in 2020), with devices like the FM series offering serial interfaces for easy integration. These are widely deployed in niche applications requiring frequent, low-power writes, such as smart cards for secure data logging, utility meters (e.g., and ) for tamper-resistant records, and automotive systems like controllers. Compared to , provides superior advantages including unlimited read cycles without degradation (due to the restore mechanism), exceptional radiation hardness up to 300 krad() total ionizing dose with minimal margin loss, and no block-level erases for instantaneous updates. However, its limitations include lower density scalability from bulky capacitors and potential imprint effects at elevated temperatures above 85°C, restricting it to rather than high-capacity roles.

Magnetoresistive RAM

Magnetoresistive random-access memory (MRAM) utilizes magnetic states to store data non-volatily, leveraging the effect in layered structures. Early MRAM concepts in the employed field-written mechanisms, where external magnetic fields generated by on-chip conductors switched the of ferromagnetic bits, as pioneered by and researchers. These designs, however, faced issues due to high power consumption and from adjacent field lines, limiting densities to prototypes like IBM's 16 Mb demonstrator in 2004. In the 2000s, toggle MRAM emerged as an advancement, developed by (later ), which used a bidirectional current pulse to induce a toggling for switching, enabling the first commercial product: a 4 Mb chip released by Everspin Technologies in 2006. This toggle mode improved reliability and endurance over field-written approaches but still required larger cell sizes and higher currents, capping densities at around 32 Mb for standalone devices. The transition to spin-transfer torque (STT) MRAM in the late 2000s, building on 's 1996 theoretical work on current-driven torque in magnetic multilayers, addressed these limitations by using spin-polarized currents to directly torque the magnetization without external fields. Everspin introduced the first commercial STT-MRAM product in 2013, starting with 4 Mb densities and scaling to 256 Mb by 2016, followed by a 1 Gb device entering pilot production in 2019. By 2025, Everspin's STT-MRAM portfolio includes densities up to 1 Gb, with ongoing expansions to support higher capacities planned for 2026-2027. At the core of STT-MRAM is the magnetic tunnel junction (MTJ), consisting of a fixed ferromagnetic layer, a thin insulating barrier (typically MgO), and a free ferromagnetic layer with . Data is encoded by the relative magnetization orientation: parallel alignment yields low resistance (logic '0'), while antiparallel yields high resistance (logic '1'), with the difference quantified by tunnel magnetoresistance (TMR) ratios exceeding 150% in modern devices. Operation involves passing a spin-polarized through the MTJ to generate on the layer's , flipping it in approximately 10 ns for writes, as demonstrated in 22 nm STT-MRAM macros. Reads occur by sensing the resistance via a small , exploiting the TMR without disturbing the state. Endurance surpasses 10^15 cycles, effectively unlimited for most applications, due to the absence of material degradation mechanisms like charge trapping. Commercially, Everspin dominates STT-MRAM production, offering chips from 1 Mb to 1 Gb with interfaces like xSPI and DDR4 for and uses. By 2025, adoption has accelerated in automotive and sectors for instant-on functionality and radiation-hardened storage, replacing or NOR in safety-critical systems. STT-MRAM's key advantages include near-unlimited endurance, sub-μJ write energies for low power operation, and compatibility with processes for high-density integration. However, challenges persist in scaling write currents as cell sizes shrink below 10 nm, requiring material innovations to maintain thermal stability and efficiency.

Advanced and Emerging Types

Phase-Change Memory

Phase-change memory (PCM) relies on the reversible phase transitions in chalcogenide glasses, such as the prototypical alloy Ge₂Sb₂Te₅ (GST), to store data non-volatily. In this material, the amorphous state exhibits high electrical resistance due to its disordered atomic structure, while the crystalline state demonstrates low resistance from ordered, conductive pathways. These distinct resistivity differences—often spanning two to three orders of magnitude—enable binary or multi-level data encoding by toggling between the states. The operation of PCM cells centers on thermal switching induced by Joule heating from applied electrical pulses. To write data into the amorphous (reset) state, a high-amplitude pulse rapidly heats the GST to its melting point (around 600–700°C), followed by a melt-quench process that cools the material faster than its crystallization rate, typically in about 50 ns. The crystalline (set) state is achieved with a lower-amplitude pulse that anneals the material to 200–300°C, allowing atomic rearrangement without melting, also in nanoseconds. Reading is performed non-destructively by sensing the cell's resistance via a small bias current, while multi-bit storage leverages intermediate phase states for higher density, such as partial crystallization levels. Performance metrics of PCM highlight its potential as a bridge between and , with endurance exceeding 10⁹ cycles per cell due to the robust phase stability of . Data retention reaches 10 years at 85°C, supported by high activation energies for (around 2.5–3 eV), though elevated temperatures accelerate drift. Switching speeds approach levels, with set operations at ~50 ns and reset at sub-10 ns, enabling latencies under 100 ns in arrays. Commercial development of PCM has seen prototypes from major manufacturers, including Samsung's 46 nm cells demonstrating scalable integration and SK Hynix's collaborations on chalcogenide-based selectors for denser arrays. Intel's , launched in 2017 as a PCM-derived technology, offered up to 1000× faster access than with stacking for terabyte-scale modules, influencing storage-class architectures despite its discontinuation in 2022 due to cost challenges. PCM finds applications in embedded non-volatile storage for automotive and devices, where its CMOS compatibility supports integration at 28 nm nodes and beyond, and as a storage-class memory (SCM) for data-centric by blending DRAM speed with flash density. Key challenges include high write power from thermal requirements (often >1 µJ per bit) and resistance drift in amorphous states, which can shift read margins over time and necessitate error-correction schemes.

Resistive RAM

Resistive random-access memory (ReRAM), also known as resistive RAM (RRAM), is a non-volatile memory technology that stores data by altering the resistance of a material between two electrodes, typically through the formation and dissolution of conductive filaments. This resistive switching mechanism relies on ion migration within a thin insulating layer, enabling high-speed, low-power operation suitable for embedded applications. ReRAM encompasses several variants, with oxygen-based resistive RAM (OxRAM) and conductive-bridge RAM (CBRAM) being prominent. OxRAM operates via oxygen vacancy filaments in metal oxide layers such as hafnium dioxide (HfO₂), where applied voltage modulates vacancy migration to form localized conductive paths. In contrast, CBRAM uses metal cations (e.g., Ag⁺ or Cu²⁺) that dissolve from an active electrode into a solid electrolyte, forming metallic bridges under bias. The core operation of ReRAM involves applying a voltage to form or rupture a conductive within the switching layer, transitioning the device between high-resistance state (HRS) and low-resistance state (LRS). In the set , positive bias drives ion or vacancy migration to create the filament, enabling current flow; the applies opposite (bipolar mode) or elevated (unipolar mode) voltage to dissolve it, often aided by in unipolar variants. Switching speeds are typically below 10 ns, with operating voltages under 3 V, allowing compatibility with processes. Bipolar switching offers precise control but requires alternating polarities, while unipolar simplifies circuitry at the cost of higher currents. ReRAM demonstrates robust performance metrics, including endurance exceeding 10¹⁰ cycles in optimized HfO₂-based OxRAM devices, far surpassing traditional limits. Data retention exceeds 10 years at elevated temperatures, supporting reliable non-volatile storage. Its crossbar array architecture enables stacking for high-density scaling, potentially reaching terabit capacities in multi-layer configurations. Commercial adoption of ReRAM has accelerated in the 2020s through pilots by Weebit Nano in collaboration with CEA-Leti, demonstrating embedded ReRAM macros in advanced nodes. By 2025, Weebit Nano achieved foundry integration with ' 22FDX platform, receiving initial wafers and validating 8 Mb arrays for low-power use, with volume production slated for 2026. These developments position ReRAM for applications in , where its analog resistance states mimic synaptic weights for efficient brain-inspired processing, and in (IoT) devices requiring ultra-low power and high endurance. ReRAM's advantages stem from its simple two-terminal structure, which integrates seamlessly with processes using standard materials, reducing fabrication costs compared to multi-layer alternatives. However, challenges include device-to-device and cycle-to-cycle variability arising from filament formation, which can affect uniformity in arrays, and the need for a forming voltage (typically 3-5 V) to initialize the , complicating low-voltage designs. Ongoing research mitigates these through material engineering and forming-free variants.

Novel Research Approaches

Ferroelectric field-effect transistors () represent a promising approach for non-volatile random-access memory by integrating ferroelectric materials into the of a , enabling shifts for with non-destructive readout. This mechanism leverages the reversible of the ferroelectric layer to store states ('0' or '1') without requiring separate capacitors, unlike traditional FeRAM, allowing seamless embedding into logic circuits for in-memory computing. Recent prototypes in 2025 have demonstrated FeFETs suitable for embedded logic applications, such as reconfigurable sequential logic-in-memory using van der Waals ferroelectrics, achieving compact non-volatile operation with low power consumption. Additionally, advancements in non-hysteric FeFET designs with higher-κ dielectrics have shown reversible for enhanced endurance and speed in logic-memory integration. IBM's project introduced a nanomechanical probe array for thermomechanical in the early 2000s, using arrays of (AFM) cantilevers to indent a surface, achieving ultrahigh densities up to 1 Tb/in² through operation of thousands of probes. Although initially prototyped for devices, recent has revived interest in similar probe-based systems, focusing on scanning probe (m-SPL) with advanced polymers like to enable nanoscale changes for writing and reading. These efforts aim to push beyond current limits with terabit-scale capacities in lab settings, though scalability remains a hurdle. Atomic-scale storage using scanning tunneling microscopy (STM) probes offers extreme density by manipulating individual atoms or molecules on surfaces to encode bits, far exceeding conventional NVRAM limits. In a seminal , researchers at created a rewritable atomic memory by positioning chlorine atoms on a surface, achieving densities 500 times higher than hard drives at the time, with bits stored as vacancies in a self-assembled layer. Ongoing STM-based techniques continue to explore hydrogen depassivation lithography on for robust atomic-scale devices, enabling precise at the single-atom level. Optical non-volatile random-access memory leverages phase-change materials responsive to light for all-optical switching, bypassing electrical contacts to reduce power and enable photonic integration. A 2025 breakthrough demonstrated an all-silicon optical memory using photon avalanche upconversion in erbium-doped nanoparticles, allowing non-volatile state retention with light-induced phase transitions for sub-nanosecond switching. Similarly, polarization-encodable cells based on tin selenide (SnSe) have shown ultrafast optical reconfiguration, storing data via changes in phase-change films. These approaches promise low-loss, reconfigurable for beyond-Moore . In 2025, researchers at developed the PoX (Phase-change ) hybrid device, a two-dimensional graphene-channel that achieves picosecond-level programming speeds of 400 ps by exploiting transitions in layers for charge trapping, enabling ultra-low non-volatile . Building on this, in October 2025, the team demonstrated the world's first full-featured 2D NOR chip using ATOM2CHIP integration of materials with processes, achieving 94% yield and positioning it as a candidate for accelerators. Despite these advances, novel NVRAM approaches face significant challenges, including fabrication complexity from nanoscale precision requirements and low yields in integrating exotic materials like ferroelectrics or oxides into processes. However, their potential lies in enabling beyond-Moore densities—exceeding 1 Tb/cm² in and methods—and energy-efficient systems for . As of 2025, these technologies remain at the lab demonstration stage, with prototypes showing promising metrics but not yet scaled to due to reliability and cost barriers.

References

  1. [1]
    What is NVRAM (non-volatile random access memory)? - TechTarget
    Mar 11, 2024 · NVRAM (non-volatile random access memory) refers to computer memory that can hold data even when power to the memory chips has been turned off.
  2. [2]
    Overview of emerging nonvolatile memory technologies
    Sep 25, 2014 · Overview of emerging nonvolatile memory technologies. Nano Review; Open access; Published: 25 September 2014. Volume 9, article number 526 ...
  3. [3]
    Non Volatile Memory | Western Digital
    Non-volatile memory (NVM) retains data when power is lost, unlike volatile memory. Examples include MRAM, PCM, and ReRAM.
  4. [4]
  5. [5]
    Emerging Nonvolatile Memory Technologies in the Future of ...
    Jun 30, 2025 · This review examines a range of emerging memory materials and device architectures, including resistive random-access memories (ReRAMs), ...
  6. [6]
    Non-Volatile Memory - an overview | ScienceDirect Topics
    Nonvolatile memory is defined as a type of memory that retains stored information even when power is turned off, as it does not require power to maintain the ...Types and Technologies of... · Integration of Non-Volatile...
  7. [7]
  8. [8]
    Radiation Effects in Advanced and Emerging Nonvolatile Memories
    DRAM is slower than SRAM, with an access time, or latency of about 10-50 ns. The memory state is physically stored in an array of capacitors, which are each ...
  9. [9]
    Progress of emerging non-volatile memory technologies in industry
    Nov 7, 2024 · For FRAM, Fujitsu provide a variety of products for serial and parallel memory with write endurances varying between 1010 and 1014 cycles, ...
  10. [10]
    Optimizing RRAM Memory Capacity for High-Density Storage
    Current technological goals for RRAM capacity optimization center on achieving ultra-high density storage exceeding 1 Tb/cm² while maintaining acceptable ...
  11. [11]
    NVRAM-Assisted Optimization Techniques for Flash Memory ...
    Due to the high density, byte-addressability, and low standby power consumption ... higher write latency, and has much higher write power than DRAM. This ...<|separator|>
  12. [12]
    What is NVRAM and How Does It Work
    Jul 18, 2025 · NVRAM stands for non-volatile random access memory. This type of memory keeps important information safe even when a device loses power.
  13. [13]
    Types of Memory in Embedded Systems - Barr Group
    May 1, 2001 · The third member of the hybrid memory class is NVRAM (non-volatile RAM). Nonvolatility is also a characteristic of the ROM and hybrid memories ...
  14. [14]
    2023 IRDS Mass Data Storage
    NVMe was designed to take advantage of the shorter latency associated with solid state non-volatile memory (not just NAND flash, but also any future non- ...
  15. [15]
    [PDF] A Survey of Non-Volatile Main Memory Technologies - JCST
    Non-Volatile Main Memories (NVMMs) have high density, non-volatility, low cost, and energy efficiency, but high write latency and limited endurance. They are ...
  16. [16]
    Memory that never forgets: emerging nonvolatile ... - Oxford Academic
    Compared to traditional memory technologies, they have the advantages of near-zero standby power, high storage density and nonvolatility, which make them ...
  17. [17]
    Pioneers of Semiconductor Non-Volatile Memory (NVM): The First ...
    May 18, 2020 · in 1967, Dawon Khang and Simon Sze came up with the idea of adding a fourth, floating, layer to store charge an MOS transistor. They ...
  18. [18]
    The History And Timeline Of Flash Memory - SemiAnalysis
    Aug 5, 2022 · Ramtron introduces first commercial FRAM non-volatile memory. 1983 ... Freescale (later Everspin) ships first commercial MRAM non-volatile memory.Missing: PCRAM | Show results with:PCRAM
  19. [19]
    Ferroelectric field effect transistors for electronics and optoelectronics
    Feb 23, 2023 · In 1993, the company RAMTRON successfully developed the first FRAM products.153 Thereafter, some major semiconductor companies, such as ...Iii. Ferroelectrics · B. Ferroelectric Materials · V. Fefets For Electronic...<|separator|>
  20. [20]
    Advances in Spin-Transfer-Torque MRAM for MRS Fall Meeting 2022
    Nov 27, 2022 · Spin-transfer-torque switching was invented at IBM in 1996 by John Slonczewski [2], who also invented the magnetic tunnel junction [3] (which ...Missing: prototype | Show results with:prototype
  21. [21]
    The evolution of phase-change memory - EDN Network
    Jul 26, 2010 · In February 2000, Intel and Ovonyx announced collaboration and licensing agreement that spawned the modern age of research and development in ...Missing: revival | Show results with:revival
  22. [22]
    Standards & Documents Search - JEDEC
    Published JEDEC documents on this website are self-service and searchable directly from the homepage by keyword or document number.Why JEDEC Standards Matter · Type Registration, Data Sheets · Ddr5 sdramMissing: NVRAM | Show results with:NVRAM
  23. [23]
    A floating gate and its application to memory devices
    A floating gate and its application to memory devices · K. Kahng, S. M. Sze · Published 8 July 1967 · Physics, Engineering · IEEE Transactions on Electron Devices.Missing: seminal | Show results with:seminal
  24. [24]
    Electron emission in intense electric fields - Journals
    The main features of the phenomenon of the extraction of electrons from cold metals by intense electric fields are well known.
  25. [25]
    [PDF] A Comprehensive Simulation Model for Floating Gate Transistors
    The objective of this work was to develop a complete simulation model for a floating-gate (FG) MOSFET using both standard SPICE primitives and also MOSFET.
  26. [26]
  27. [27]
    Floating Gate - an overview | ScienceDirect Topics
    Tunnel oxide scaling has been limited by the occurrence of stress induced leakage current (SILC) [15–18] limiting the minimum thickness to about 7 nm. The ...
  28. [28]
  29. [29]
    Spin-transfer torque RAM technology: Review and prospect
    Development of MTJs with high tunnel magneto-resistance (TMR) ratio. An MTJ consists of two ferromagnetic thin films separated by a tunnel barrier, as shown ...
  30. [30]
    Vacancy Structures and Melting Behavior in Rock-Salt GeSbTe
    May 3, 2016 · In particular, amorphization (reset operation in PCM) is a key step for PCM devices, which is usually achieved through the melt-quench process ...
  31. [31]
    Evolution of the conductive filament system in HfO 2 - Nature
    Dec 13, 2021 · The formation and rupture of CFs yield the low resistance state (LRS) and high resistance state (HRS) of the oxide memristor, respectively.<|control11|><|separator|>
  32. [32]
    Variability and power enhancement of current controlled resistive ...
    May 1, 2023 · Experimental measurements indicate a clear improvement on resistance states stability when using current sweeps to induce both set and reset processes.
  33. [33]
    Phase change materials in non-volatile storage - ScienceDirect.com
    Amorphous phase stability. An ideal non-volatile memory would require that both crystalline and amorphous phases are thermodynamically stable in the range of ...Missing: variability | Show results with:variability
  34. [34]
    Where is Memory Technology From, and Where is it Headed - KIOXIA
    Oct 1, 2019 · Toshiba created the world's first NOR flash memory in 1984, and in 1987, it invented NAND flash memory, which is widely used still today. ──Not ...
  35. [35]
    New ultra high density EPROM and flash EEPROM with NAND ...
    Abstract: In order to realize ultra high density EPROM and Flash EEPROM, a NAND structure cell is proposed. This new structure is able to shrink cell size ...
  36. [36]
    Storage 101: Understanding the NAND Flash Solid State Drive
    Mar 11, 2020 · This article introduces you to several important SSD concepts so you have a clearer picture of the components that go into an SSD and how they work together.
  37. [37]
    What is TLC Flash (Triple-Level Cell Flash)? - TechTarget
    May 23, 2024 · TLC flash (triple-level cell flash) is a type of NAND flash memory that stores three bits of data per cell. TLC is also known as multi-level cell-3 (MLC-3), 3- ...
  38. [38]
    Samsung unveils 10th Gen V-NAND: 400+ layers, 5.6 GT/s and ...
    Feb 26, 2025 · Samsung has introduced its upcoming 10th-generation V-NAND flash memory with over 400 active layers and a 5.6 GT/s interface speed.
  39. [39]
  40. [40]
    [PDF] Characterizing Flash Memory: Anomalies, Observations, and ...
    Despite con- tinued density scaling, the basic performance (read, program, and erase latencies) of flash devices has been roughly constant for over a decade.
  41. [41]
    SLC, MLC, TLC & 3D NAND Differences: USB, SSD, Cards
    Common types of NAND flash storage are SLC, MLC, TLC and 3D NAND. This article discusses the different characteristics of each type of NAND.Missing: operation block
  42. [42]
    Global NAND Memory Market Share: Quarterly
    Oct 16, 2025 · We expect the NAND market to see further growth in 2H 2025, driven by AI-related demand that has led to an ongoing HDD shortage and stronger ...
  43. [43]
    SK hynix unveils AI NAND strategy, including gargantuan petabyte ...
    Oct 28, 2025 · The company expects AIN D to be designed to reach petabyte-level density, far beyond today's terabyte-scale SSDs, which retain key benefits of ...
  44. [44]
    From endurance to intelligence: Flash memory's battle against ...
    Expect to see AI-driven, context-aware flash operations, smarter error correction, and wear leveling that responds to changing conditions. Security ...Missing: degradation | Show results with:degradation
  45. [45]
    Why Flash Wears Out and How to Make it Last Longer
    Mar 6, 2012 · In this article we'll look at the endurance of flash devices and how advancements in SSD controller technology can increase a drive's lifespan while improving ...
  46. [46]
    Demonstration of 1T1C FeRAM Arrays for Nonvolatile Memory ...
    Information storage is achieved by switching the polarization direction of ferroelectric capacitors, typically using materials such as lead zirconate titanate ( ...
  47. [47]
    None
    ### Summary of FRAM Sections from https://www.mouser.com/pdfDocs/mram-vs-fram.pdf
  48. [48]
    [PDF] Low-Power FRAM Microcontrollers and Their Applications (Rev. A)
    In contrast, FRAM requires only ~100 ns to program one data word. Additionally, no pre-erase is required and, because of the fast write, there is virtually ...Missing: reversal <5V
  49. [49]
    The New Indelible Memories - IEEE Spectrum
    Jan 31, 2003 · They are almost all flash memory chips, a type of electrically erasable and programmable read-only memory. Nonvolatility, flash's property of ...
  50. [50]
    [PDF] Radiation Evaluation of Ferroelectric Random Access Memory ...
    It provides non-volatile storage like Flash, but offers faster writes, high read-write cycle endurance (>1015 cycles), and very low power consumption [1].
  51. [51]
    A Full Spectrum of 3D Ferroelectric Memory Architectures Shaped ...
    Apr 13, 2025 · The 1T-1C FeRAM consists of a single access transistor and a ferroelectric metal-ferroelectric-metal (MFM) capacitor stacked on the transistor's ...
  52. [52]
    NVM in Data Storage: A Post-Optane Future - ACM Digital Library
    The first commercial FeRAM products were introduced in the late 1990s. Ramtron International Corporation, now part of Cypress Semiconductor, was one of the ...
  53. [53]
    F-RAM™ for Smart E Meters - DigiKey
    Jan 26, 2015 · This presentation will show how to replace EEPROMs with a superior solution in F-RAM. It will also review common design problems with ...
  54. [54]
  55. [55]
    Heavy-ion Radiation Impact on a 4Mb FRAM under Different Test ...
    The impact of heavy-ions on commercial Ferroelectric Memories (FRAMs) is analyzed. The influence of dynamic and static test modes as well as several stimuli ...
  56. [56]
    Spin-Transfer-Torque MRAM: The Next Revolution in Memory
    Mar 5, 2023 · Spin-Transfer-Torque MRAM was invented at IBM by John Slonczewski in the early 1990s. By using a spin-polarized current, ...
  57. [57]
    (PDF) Development of the magnetic tunnel junction MRAM at IBM
    This paper reviews the remarkable developments of the magnetic tunnel junction over the last decade and in particular, work aimed at demonstrating its ...
  58. [58]
    [PDF] Status and Outlook Torque MRAM - Everspin Technologies
    The first MRAM product used toggle-mode writing as a form of magnetic-field induced switching and became commercially available in. 2006 (the 4Mb Everspin ...
  59. [59]
    Everspin officially announces the world's first ST-MRAM chip, will be ...
    Nov 13, 2012 · Everspin officially announces the world's first ST-MRAM chip, will be available in 2013 | MRAM-Info.Missing: date | Show results with:date
  60. [60]
    Everspin Enters Pilot Production Phase for the World's First 28 nm 1 ...
    Jun 11, 2019 · Everspin has been in volume production of its 256 Mb STT-MRAM product for more than a year and now adds the ground-breaking 1 Gb capacity ...Missing: date | Show results with:date
  61. [61]
    EMD4E001G - 1Gb Spin-transfer Torque MRAM - Everspin
    The EMD4E001G is the newest and highest capacity STT-MRAM device from Everspin, targeted at enterprise and computing applications that need higher capacity.
  62. [62]
    Everspin Technologies at Q3 Investor Summit: Expanding MRAM ...
    Sep 16, 2025 · These are the two X5 products that we brought to production recently. We have a high-density product that is planned later in 2026, 2027.
  63. [63]
    Spin-transfer Torque MRAM Technology - Everspin
    STT-MRAM uses spin-transfer torque to manipulate electron spin, establishing magnetic states in a MTJ, with reduced switching energy and high scalability.
  64. [64]
    Recent progress in spin-orbit torque magnetic random-access memory
    Oct 1, 2024 · We review recent advancements in perpendicular SOT-MRAM devices, emphasizing on material developments to enhance charge-spin conversion efficiency.
  65. [65]
  66. [66]
    Everspin Technologies: Positioning for 2025 Growth as MRAM ...
    Aug 6, 2025 · - Everspin Technologies leads MRAM growth in 2025, driven by aerospace, automotive, and defense demand for non-volatile memory. - Recent $14.55M ...
  67. [67]
    Everspin Expands High-Reliability MRAM Portfolio with New ...
    "Our latest high-reliability PERSYST MRAM provides designers with a robust, fast, and scalable alternative to Static RAM or NOR flash, ensuring ...Missing: densities | Show results with:densities
  68. [68]
    [PDF] Challenges and Process Solutions for STT MRAM Scaling
    SOT MRAM. Fast write of < 1 ns possible for SOT MRAM, potentially with infinite endurance (> 1E12). Difference & Similarities between STT and SOT MRAM. STT MRAM.
  69. [69]
    Phase-change properties of GeSbTe thin films deposited by plasma ...
    PCM uses the reversible phase change between the crystalline and amorphous states of chalcogenide materials brought about by Joule heating. Ternary Ge2Sb2Te5 ( ...
  70. [70]
    Abstract - arXiv
    In a phase change memory device, switching between a crystalline and amorphous material state occurs in nanoseconds and is thermally induced by Joule heating.
  71. [71]
    Phase change memory programming circuit with improved speed
    Jan 15, 2025 · To program the PCM cell into the amorphous state, the phase change material is first heated to the melting temperature and then rapidly quenched ...
  72. [72]
    Phase-Change Memory for In-Memory Computing | Chemical Reviews
    Phase-change memory (PCM) is a leading memory technology being explored for IMC. In this perspective, we review the current state of phase-change materials.Missing: PCRAM revival Ovonyx
  73. [73]
    An overview of phase-change memory device physics - IOPscience
    Mar 26, 2020 · In this article, we provide an overview of the current understanding of the main PCM device physics that underlie the read and write operations.Abstract · Essentials of phase-change... · Write operation · Read operationMissing: PCRAM RAM
  74. [74]
    Atomic scale insight into the effects of Aluminum doped Sb 2 Te for ...
    Oct 11, 2018 · However, when apply GST to PCM, the mediocre performance of Set speed (~50 ns) and retention (85 °C, 10 years) has made it unsatisfactory for ...
  75. [75]
    [PDF] Phase-change Memory - Towards a Storage-class ... - IEEE Xplore
    A Ga-Sb-Ge material was studied and showed significantly improved 10-year data retention up to 220◦C, while still exhibiting strong 80 ns set speeds and 109 ...
  76. [76]
    IBM, SK Hynix sign phase-change memory deal - EE Times
    Jun 12, 2012 · Memory IC maker SK Hynix Inc. has agreed to develop non-volatile phase-change memory technology and products with IBM.
  77. [77]
    Intel to Shut Down Optane Memory Business, Retire 3D XPoint ...
    Jul 28, 2022 · Intel plans to wind up both its Optane Memory business, as well as further development of 3D XPoint.Missing: 2017-2023 discontinued
  78. [78]
    Subnanosecond flash memory enabled by 2D-enhanced hot-carrier ...
    Apr 16, 2025 · The Dirac channel flash shows a program speed of 400 picoseconds, non-volatile storage and robust endurance over 5.5 × 106 cycles. Our results ...
  79. [79]
    Phase change memory for automotive grade embedded NVM ...
    This paper illustrates the enormous potential of PCM in the field of eNVM. The evolution of embedded PCM (ePCM) is reviewed, showing evidence for the effort ...
  80. [80]
    Extended switching endurance of phase change memory ... - NIH
    Jul 1, 2025 · Phase change memory has been regarded as a promising candidate for storage class memory application. However, the high switching current and ...
  81. [81]
    Drift-tolerant Coding to Enhance the Energy Efficiency of Multi-Level ...
    This work proposes a drift-tolerant coding scheme to enable the fast write operation on MLC PCM without sacrificing any data accuracy.
  82. [82]
    Resistive Switching Random-Access Memory (RRAM): Applications ...
    This work addresses the RRAM concept from materials, device, circuit, and application viewpoints, focusing on the physical device properties and the ...Missing: variants OxRAM
  83. [83]
    Resistive Random Access Memory (RRAM): an Overview of ...
    Apr 22, 2020 · Endurance is thus defined as the number of times a RRAM device can be switched between the HRS and the LRS while ensuring a reliably ...
  84. [84]
    Resistive RAM Performance: A Comparative Analysis of Materials
    Their RRAM technology achieves switching speeds of approximately 10ns, with demonstrated endurance of 10^7 cycles and data retention exceeding 10 years at ...
  85. [85]
  86. [86]
    [PDF] ReRAM: Now in Commercialisation - AFR
    “The work Weebit and CEA-Leti are doing to make Weebit ReRAM available on. GlobalFoundries' 22FDX is a welcome development as we continue to expand the ...Missing: pilots 2020s
  87. [87]
    GlobalFoundries announced that its 22FDX+ with RRAM technology ...
    Aug 29, 2025 · Weebit received first 22-nm GlobalFoundries wafers with its embedded RRAM memory ... Weebit Nano licenses its ReRAM technology to DB HiTek foundry.
  88. [88]
    Towards Memristor-Based Neuromorphic RFID Circuits and Architectures
    **Summary of ReRAM Applications in Neuromorphic Computing:**
  89. [89]
    ReRAM Next-Generation Memory Technology | Weebit
    Our ReRAM (or RRAM) NVM memory technology is ideal for IoT, MCU, Analog, Power & Secure ICs, providing higher performance with lower costs.Guide to ReRAM / RRAM · ReRAM Advantage | Technology · Embedded ReRAM IPMissing: commercial pilots 2020s
  90. [90]
    [PDF] RRAM Variability and its Mitigation Schemes - TU Delft Repository
    The RRAM has multiple advantages over STT-MRAM and PCRAM such as: 1) It can be better scaled (down to few nm) because of its ion based switching mechanism [3]; ...
  91. [91]
    (Invited) Resistive Memories (RRAM) Variability: Challenges and ...
    Aug 8, 2025 · In this work, we address Resistive RAM (RRAM) variability. To this aim, we investigate various RRAM technologies (Oxide RAM and Conductive ...
  92. [92]
    Reconfigurable Sequential-Logic-in-Memory Implementation ...
    Jan 2, 2025 · This study proposes a compact, nonvolatile, and reconfigurable van der Waals (vdW) ferroelectric field-effect transistor (FeFET)-based sequential logic-in- ...Missing: prototypes embedded
  93. [93]
    Logic and Memory Ferroelectric Field-Effect-Transistor Using ...
    Sep 4, 2025 · We propose a higher- κ \kappa non-hysteric ferroelectric field-effect transistor (FEFET) using reversible domain wall displacement.
  94. [94]
    Millipede - A MEMS-based scanning-probe data-storage system
    Mar 1, 2003 · Ultrahigh storage densities of up to 1 Tb/in.2 or more can be achieved by using local-probe techniques to write, read back, and erase data ...Missing: nanomechanical revival 2025
  95. [95]
    Probe‐Based Mechanical Data Storage on Polymers Made by ... - NIH
    In this study, we report the first demonstration of polysulfide polymers as storage media for probe‐based data systems. Mechanical information storage was ...
  96. [96]
    Atomic memory holds 500 times more data per square inch than ...
    Jul 18, 2016 · Rewritable system moves atomic-scale data storage closer to commercial reality.
  97. [97]
    Lithography for robust and editable atomic-scale silicon devices and ...
    Jul 23, 2018 · Here, we report scanning tunneling microscope (STM) techniques to substantially improve automated hydrogen lithography (HL) on silicon.Missing: NVRAM | Show results with:NVRAM
  98. [98]
    All-silicon non-volatile optical memory based on photon avalanche ...
    Jan 25, 2025 · They can be achieved by altering the material properties interacting with light, including phase-change memories (PCMs), ferroelectric memories, ...
  99. [99]
    Polarization-encodable photonic memory cells using next ... - SciOpen
    Jan 22, 2025 · The study uses tin selenide (SnSe) to create polarization-encodable photonic memory cells, enabling ultrafast switching by changing the optical ...
  100. [100]
    Researchers develop flash memory device
    Apr 17, 2025 · Researchers from Shanghai-based Fudan University have developed a picosecond-level flash memory device with an unprecedented program speed of 400 picoseconds.
  101. [101]
    World's fastest Flash memory developed: writes in just 400 ...
    Apr 18, 2025 · The newly developed chip, named “PoX” (Phase-change Oxide), is capable of switching at 400-picoseconds, substantially surpassing the previous world record.
  102. [102]
  103. [103]
    Optical Phase Change Materials - Annual Reviews
    This review on Optical Phase Change Materials is by Yi-Siou Huang, Chih-Yu Lee, Ichiro Takeuchi, and Carlos A. Ríos Ocampo, published in July 2025.<|separator|>