Nonvolatile Resistive Random Access Memory Research Articles

Investigation on self-rectifying properties of Pt/HfO2/Ti with rivet-like structure based on ALD conformal technology

This work explores an architecture for nonvolatile resistive random-access memory (RRAM) systems. The study proposes a self-rectifying RRAM device utilizing a two-terminal 1R selector that harmonizes the gate control efficacy of transistors with the inherent simplicity of diode structures. A rivet-like HfO2-based RRAM array is meticulously constructed through atomic layer deposition (ALD), aiming to enhance device performance and retention stability. The conformal fabrication technique of ALD method is critical in achieving uniform coverage of isolation layers and precise electrode placement, which is instrumental in the fabrication of high-performance memory cells. Empirical analyses indicate significant improvements in rectification and ON/OFF ratios compared to existing RRAM models, bolstered by compatibility with established CMOS processes. It reveals that these advances are conducive to scalable, high-density memory integration, positioning the RRAM as a viable contender for future computational applications that require high efficiency and neuromorphic computing capabilities.

Random telegraph noise characteristic of nonvolatile resistive random access memories based on optical interference principle

The influence of random telegraph noise (RTN) could reduce the reading margin, which would cause computational errors in data recognition. This paper proposes a current sensor based on the principle of optical fiber interference, which can avoid the interference generated during the RTN testing process and improve the accuracy due to its passive characteristics. In this study, a hafnium oxide based memristor was fabricated, the switching voltages of Cu and TiN as the top electrodes are 0.2 V and 0.15 V, respectively. In addition, the RTN spectral density of the two device structures in LRS increases from 10−5 to 10−1 A2 Hz−1 and from 10−5 to 101 A2 Hz−1 with increasing applied voltage. While the RTN in high resistance state is independent of the applied voltage. Furthermore, based on the analysis of the experimental data, the generation mechanism of the RTN is attributed to local defects and the capture or emission of carriers by traps.

Memory device based on a nano-granulars and nano-worms structured MoS2 active layer: The origin of resistive switching characteristics

This study systematically demonstrates resistive switching (RS) functionality of MoS2 nanostructured [nano-granulars (NGs) and nano-worms (NWs)] based non-volatile resistive random access memory (ReRAM) device. Interestingly, the SET and RESET voltage sharply changes from NGs to NWs device. The RS behavior of devices is interpreted by proposing and discussing a model. In the Ag/MoS2(NGs)/ITO device, a high resistance state (HRS) and low resistance state (LRS) could be ascribed due to the formation/disruption of conducting filament via trapping-detrapping of electrons across the NGs active layer. In contrast, the RS behavior of Ag/MoS2(NWs)/ITO device is caused by forming of an Ag metallic filamentary route between the electrodes. The NWs structured device show substantially better endurance and homogeneity than the device with an NGs structure. This nanostructural-dependent device functionality allows extended RS properties and establishes a relationship between the material's nano-level structural and electrical features.

Near room temperature multilevel resistive switching memory with thin film ionic liquid crystals

Multilevel operation of nonvolatile resistive random-access memory devices was demonstrated using thin films of an ionic liquid crystal, 1-dodecyl-3-methylimidazolium tetrafluoroborate ([C12mim][BF4]), as a resistive switching layer.

Growth and characterization of sputter-deposited Ga2O3-based memristive devices

In the last few years, there has been significant interest in gallium oxide devices for resistive switching technologies due to its remarkable sensitivity to oxygen. In this study, we present the growth and resistive switching of a (2¯01) oriented (75 ± 3) nmβ-Ga2O3 thin film on a Ru/Al2O3 substrate using magnetron radio frequency sputtering. The observed resistive switching was attributed to the formation and rupture of conductive filaments constituted by oxygen vacancies in the β-Ga2O3 film as confirmed by x-ray photoelectron spectroscopy and energy-dispersive x-ray spectroscopy. The electrical conduction was found to be of Ohmic nature in the low-resistance ON state, while the high-resistance OFF state was governed by the Poole–Frenkel transport mechanism. Exhibiting stable endurance cycles, long retention times, and ON/OFF ratios of up to 104, the devices can be considered as promising prototypes for future nonvolatile resistive switching random access memory with respect to both switching performance and device stability.

Ionic Liquid Crystal Thin Film as Switching Layer in Nonvolatile Resistive Memory.

In this study, we propose the use of an ionic liquid crystal (ILC) as a new resistive switching layer in nonvolatile resistive random-access memory (ReRAM) devices. The high-quality vacuum-deposited ILC films of 1-hexadecyl-3-methylimidazolium hexafluorophosphate ([C16mim][PF6]) enabled to demonstrate the first operation of ReRAM devices with a low set voltage of ∼1 V and stable switching behavior for up to ∼44 cycles. The key to the successful operation is that the ILC layer is in the liquid crystal phase (smectic A), where the electric double layers formed at the electrode-ILC interfaces play a significant role. The results of basic electrical properties and I-V curve fittings suggested the following operation principle: the formation and rupture of charge-composed filaments within the ILC film, where the current conduction is primarily governed by the trap charge limited current (TCLC) mechanism. These achievements will pave the way for advanced studies of ILC-based electronic devices.

Resistive switching characteristics of MnO2-based thin film for transparent non-volatile ReRAM

The present study covers the fabrication of indium tin oxide (ITO)/manganese dioxide (MnO2)/ITO capacitor-based transparent resistive random access memory (ReRAM) device. The ITO/MnO2/ITO device exhibits 80 % transparency in the visible region along with steady bipolar resistive switching characteristics. The conduction mechanism investigation reveals that the current in low resistance state (LRS) is governed by Ohmic conduction. However, the dominance of trap-controlled space charge limited conduction (SCLC) mechanism in high resistance state (HRS) was observed. The resistive switching phenomenon in ITO/MnO2/ITO device was caused due to the formation and rupture of conduction filament via electron trapping and de-trapping in oxide-related traps (oxygen vacancies). In terms of reliability, the MnO2-based device exhibits good endurance up to 104 cycles and long retention for 104 s at room-temperature as well as at 85 °C. Cumulative probability distribution shows resistance ratio of 103 between LRS and HRS, which is sufficient to read memory states. These outcomes suggest that the ITO/MnO2/ITO-based transparent ReRAM could be a futuristic see-through electronic device for artificial intelligence (AI) application.

Peculiarities of electron transport and resistive switching in point contacts on TiSe2, TiSeS, and CuxTiSe2

Compound TiSe2 has received much attention among the transition metal chalcogenides because of its thrilling physical properties concerning atypical resistivity behavior, the emergence of charge density wave (CDW) state, induced superconductivity, etc. Here, we report the discovery of a new feature of TiSe2, namely, the observation of resistive switching in voltage biased point contacts (PCs) based on TiSe2 and its derivatives doped by S and Cu (TiSeS, CuxTiSe2). The switching occurs between a low resistive mainly “metallic-type” state and a high resistive “semiconducting-type” state when a bias voltage is applied (usually < 0.5 V), and reverse switching occurs when a voltage of opposite polarity is applied (usually < 0.5 V). The difference in resistance between these two states can reach up to two orders of magnitude at room temperature. The origin of this effect can be attributed to the variation of stoichiometry in the PC core due to the drift/displacement of Se/Ti vacancies under a high electric field. Additionally, we demonstrated that heating occurs in the PC core, which can facilitate the electric field-induced effect. At the same time, we did not find any evidence for CDW spectral features in our PC spectra for TiSe2. The observed resistive switching allows proposing TiSe2 and their derivatives as promising materials, e.g., for non-volatile resistive random access memory (ReRAM) engineering.

Magnetic field control and swift heavy ion beam assisted tuning of resistive switching properties of BSFO/CFO/LNO heterostructures

Resistive switching (RS) behavior in mixed oxide insulators has shown a great promise as memristors or non-volatile resistive random-access memory (RRAM) applications. For dilute magnetic oxide multilayers, a novel approach of controlled defects induced and the magnetic field control of RS behavior is proposed. Resistive switching in Bi0.6Sr0.4FeO3 /CoFe2O4 /LaNiO3 (BSFO/CFO/LNO) multilayer heterostructures has been investigated as a case study. All oxide junctions consisting of conducting LaNiO3 (LNO) bottom electrode and BSFO-CFO active layers were fabricated by using chemical solution deposition. A set of samples were irradiated with 150 MeV Ag11+ ions for three different ion fluence of ∼1 × 10+11 ions cm−2, 1 × 10+12 ions cm−2 and 5 × 10+12 ions cm−2. Polycrystalline phase pure films with smooth, crack free surfaces were observed for pristine and irradiated samples. Optical spectroscopy revealed a decrease in the transmittance upon increasing ion fluence due to increase in the light scattering centres. The optical band gap showed a systematic decrease from 2.09 eV to 1.65 eV with increasing ion fluence. Room temperature I-V characteristics showed consistent and pronounced bipolar switching for all samples below ± 5 V. Upon applied magnetic fields of 0.58 T, the resistive switching ratios were found to increase significantly and were further tuned by 150 MeV Ag11+ ion beam irradiations. The magnetic field control of electrical transport properties in the controlled defect assisted oxide heterojunctions offers new insights to the existing understanding of oxide-based RS mechanism.

Design and synthesis of a solution-processed redox-active bis(formazanate) zinc complex for resistive switching applications.

In this paper, we report the synthesis and characterization of a mononuclear zinc complex (1) containing a redox-active bis(4-antipyrinyl) derivative of the 3-cyanoformazanate ligand. Complex 1 was readily obtained by refluxing zinc acetate with 3-cyano-1,5-(4-antipyrinyl)formazan (LH) in a methanolic solution. Single-crystal X-ray diffraction analysis of complex 1 shows that the formazanate ligands bind to the zinc center in a five-member chelate "open" form via the 1- and 4-positions of the 1,2,4,5-tetraazapentadienyl formazanate backbone leading to the formation of the mononuclear bis(formazanate) zinc complex exhibiting a distorted octahedral geometry. We also report the study of resistive-switching random access memory application of this solution-processable bis(formazanate) Zn(II) complex to facilitate the practical implementation of transition metal complex-based molecular memory devices. The complex demonstrated high conductance switching with a large ON-OFF ratio, good stability, and a long retention time. A trap-controlled space charge limited current mechanism is proposed for the observed resistive switching behavior of the device, wherein the role played by the [ZnIIL2] complex that comprises the extended redox-active conjugated ligand backbone is revealed by corroborating electrochemical studies, spectrochemical experiments, and DFT calculations. In addition to providing significant insights into the molecular design of transition metal complexes for memory applications, this study also presents the utilization of ZnIIL2 towards the realization of non-volatile resistive random access memory (RRAM) devices with inorganic/organic hybrid active layers that are highly cost-effective and sustainable. These devices exhibited multilevel switching and low current operation, both of which are desirable for advanced memory applications.

Charge trapped CdS quantum dot embedded polymer matrix for a high speed and low power memristor.

The data storage requirement in the digital world is increasing day by day with the advancement of the internet of things. In this respect, nonvolatile resistive random-access memory is an option that provides high density and low power data storage capabilities. In this work, zero-dimensional colloidal CdS quantum dots and a polymer composite at an appropriate ratio were used to fabricate a memristive device. Comparison with a pristine CdS quantum dot-based device reveals that a surrounding matrix around the quantum dots is needed for observing memristive behavior. The quantum dots embedded in the polymer matrix device showed extremely stable electrical switching behavior that can be operated for more than 300 cycles and 60 000 seconds. Moreover, the device needs extremely low power to operate at a very high speed. The smooth surface morphology dictates a charge trapping mechanism for the switching phenomenon; however, an interplay between different charge transport mechanisms leads to the fast switching and high on-off ratio of the device.

A compute-in-memory chip based on resistive random-access memory

Realizing increasingly complex artificial intelligence (AI) functionalities directly on edge devices calls for unprecedented energy efficiency of edge hardware. Compute-in-memory (CIM) based on resistive random-access memory (RRAM)1 promises to meet such demand by storing AI model weights in dense, analogue and non-volatile RRAM devices, and by performing AI computation directly within RRAM, thus eliminating power-hungry data movement between separate compute and memory2–5. Although recent studies have demonstrated in-memory matrix-vector multiplication on fully integrated RRAM-CIM hardware6–17, it remains a goal for a RRAM-CIM chip to simultaneously deliver high energy efficiency, versatility to support diverse models and software-comparable accuracy. Although efficiency, versatility and accuracy are all indispensable for broad adoption of the technology, the inter-related trade-offs among them cannot be addressed by isolated improvements on any single abstraction level of the design. Here, by co-optimizing across all hierarchies of the design from algorithms and architecture to circuits and devices, we present NeuRRAM—a RRAM-based CIM chip that simultaneously delivers versatility in reconfiguring CIM cores for diverse model architectures, energy efficiency that is two-times better than previous state-of-the-art RRAM-CIM chips across various computational bit-precisions, and inference accuracy comparable to software models quantized to four-bit weights across various AI tasks, including accuracy of 99.0 percent on MNIST18 and 85.7 percent on CIFAR-1019 image classification, 84.7-percent accuracy on Google speech command recognition20, and a 70-percent reduction in image-reconstruction error on a Bayesian image-recovery task.

Water-Soluble Polythiophene-Conjugated Polyelectrolyte-Based Memristors for Transient Electronics.

The key to protect sensitive information stored in electronic memory devices from disclosure is to develop transient electronic devices that are capable of being destroyed quickly in an emergency. By using a highly water-soluble polythiophene-conjugated polyelectrolyte PTT-NMI+Br- as an active material, which was synthesized by the reaction of poly[thiophene-alt-4,4-bis(6-bromohexyl)-4H-cyclopenta(1,2-b:5,4-b')dithiophene] with N-methylimidazole, a flexible electronic device, Al/PTT-NMI+Br-/ITO-coated PET (ITO: indium tin oxide; PET: polyethylene terephthalate), is successfully fabricated. This device shows a typical nonvolatile rewritable resistive random access memory (RRAM) effect at a sweep voltage range of ±3 V and a history-dependent memristive switching performance at a small sweep voltage range of ±1 V. Both the learning/memorizing functions and the synaptic potentiation/depression of biological systems have been emulated. The switching mechanism for the PTT-NMI+Br--based electronic device may be highly associated with ion migration under bias. Once water is added to this device, it will be destructed rapidly within 20 s due to the dissolution of the active layer. This device is not only a typical transient device but can also be used for constructing conventional memristors with long-term stability after electronic packaging. Furthermore, the soluble active layer in the device can be easily recycled from its aqueous solution and reused for fabricating new transient memristors. This work offers a train of new thoughts for designing and constructing a neuromorphic computing system that can be quickly destroyed with water in the near future.

Frequency-Dependent Synapse Weight Tuning in 1S1R with a Short-Term Plasticity TiOx-Based Exponential Selector.

Short-term plasticity (STP) is an important synaptic characteristic in the hardware implementation of artificial neural networks (ANN), as it enables the temporal information processing (TIP) capability. However, the STP feature is rather challenging to reproduce from a single nonvolatile resistive random-access memory (RRAM) element, as it requires a certain degree of volatility. In this work, a Pt/TiOx/Pt exponential selector is introduced not only to suppress the sneak current but also to enable the TIP feature in a one selector-one RRAM (1S1R) synaptic device. Our measurements reveal that the exponential selector exhibits the STP characteristic, while a Pt/HfOx/Ti RRAM enables the long-term memory capability of the synapse. Thereafter, we experimentally demonstrated pulse frequency-dependent multilevel switching in the 1S1R device, exhibiting the TIP capability of the developed 1S1R synapse. The observed STP of the selector is strongly influenced by the bottom metal-oxide interface, in which Ar plasma treatment on the bottom Pt electrode resulted in the annihilation of the STP feature in the selector. A mechanism is thus proposed to explain the observed STP, using the local electric field enhancement induced at the metal-oxide interface coupled with the drift-diffusion model of mobile O2- and Ti3+ ions. This work therefore provides a reliable means of producing the STP feature in a 1S1R device, which demonstrates the TIP capability sought after in hardware-based ANN.

Fully solution-processed organic RRAM device with highly stable butterfly-shaped hysteresis

Resistive random access memory (RRAM) is considered a promising next-generation memory device owing to its low energy consumption and ultra-high speed. In this study, we demonstrated a vacuum-free and low-cost solution to fabricate a highly stable nonvolatile RRAM device. Organic materials such P3HT and PEDOT:PSS were used to form the active layer and top electrode, respectively. The fabricated organic RRAM has a structure of ITO/P3HT/PEDOT:PSS, and exhibits well-controlled and highly stabilized nonvolatile butterfly shape resistive switching behavior, with an endurance of up to 1000 repeated cycling tests. Schottky emission, hopping, and direct tunneling were evaluated for switching mechanism. The geometrical structure was analyzed using the field emission scanning electron microscopy image, and the switching mechanisms were confirmed through the ultraviolet photoelectron spectroscopy measurement and energy band diagram analysis. Our results indicate that the fabricated solution-processed organic RRAM devices have the potential for use in nonvolatile memory devices.

Optimizing the thickness of Ta2O5 interfacial barrier layer to limit the oxidization of Ta ohmic interface and ZrO2 switching layer for multilevel data storage

• Optimization of the thickness (0, ∼ 2, ∼ 4, and ∼ 6 nm) of the Ta 2 O 5 interfacial barrier layer has been utilized to limit the oxidization of the Ta ohmic interface and ZrO 2 RS layer. • Lower forming/SET-voltages high pulse endurance (10 6 cycles), long retention time (10 4 s) at 100 °C, and reliable multilevel resistance states were obtained. • Multilevel resistance states have been scientifically investigated via modulating the compliance current (CC) and RESET-stop voltages. • I-V characteristics of HRS are found to be a good linear fit with the Schottky equation. • Schottky barrier height rises by increasing the stop-voltage during RESET-operation, resulting in enhancing the data storage memory window (on/off ratio). The multilevel storage capability of nonvolatile resistive random access memory (ReRAM) is greatly desired to accomplish high functioning memory density. In this study, Ta 2 O 5 thin film with different thicknesses (2, 4, and 6 nm) was exploited as an appropriate interfacial barrier layer for limiting the formation of the interfacial layer between the 10 nm thick sputtering deposited resistive switching (RS) layer and Ta ohmic electrode to improve the switching cycle endurance and uniformity. Results show that lower forming voltage, narrow distribution of SET-voltages, good dc switching cycles (10 3 ), high pulse endurance (10 6 cycles), long retention time (10 4 s at room temperature and 100°C), and reliable multilevel resistance states were obtained at an appropriate thickness of ∼2 nm Ta 2 O 5 interfacial barrier layer instead of without Ta 2 O 5 and with ∼4 nm, and ∼6 nm Ta 2 O 5 barrier layer, ZrO 2 -based memristive devices. Besides, multilevel resistance states have been scientifically investigated via modulating the compliance current (CC) and RESET-stop voltages, which displays that all of the resistance states were distinct and stayed stable without any considerable deprivation over 10 4 s retention time and 10 4 pulse endurance cycles. The I - V characteristics of RESET-stop voltage (from −1.7 to −2.3 V) of HRS are found to be a good linear fit with the Schottky equation. It can be seen that Schottky barrier height rises by increasing the stop-voltage during RESET-operation, resulting in enhancing the data storage memory window (on/off ratio). Moreover, RESET-voltage and CC control of HRS and LRS revealed the physical origin of the RS mechanism, which entails the formation and rupture of conducting nanofilaments. It is thoroughly investigated that proper optimization of the barrier layer at the ohmic interface and the switching layer is essential in memristive devices. These results demonstrate that the ZrO 2 -based memristive device with an optimized ∼2 nm Ta 2 O 5 barrier layer is a promising candidate for multilevel data storage memory applications.

Bipolar resistive switching in Ag/VO2(B)/SiOx/n++Si RRAM

Non-volatile resistive random-access memory (RRAM) is being promoted as a possible alternative to flash memory, however the optimal material system and sophisticated fabrication techniques hinder its utilization in practical routes. Here, we demonstrate the direct fabrication of metal/oxides/semiconductor (MOS) structured Ag/VO2(B)/SiOx/n++Si RRAM via drop-coating process, in which bipolar resistive switching behavior was obtained and investigated systematically. The RRAM devices exhibit good cycle-to-cycle endurance (>30 cycles) and high on/off ratio (>60). The switching mechanism is proposed to form Ag conducting filaments via VO2(B) nanorods’ guide by comparing the resistive switching behavior of Ag/SiOx/n++Si, Ag/VO2(B)/n++Si, Ag/VO2(B)/SiOx/n++Si devices and the corresponding SEM images before and after the application of electric field, which is confirmed by introducing NaCl barrier layer in Ag/VO2(B)-NaCl/SiOx/n++Si devices. The present study may pave a convenient route for fabricating the ultrahigh density resistive memory devices without the aid of complex fabrication techniques, as well as provide a new potential material system for RRAM.

Resistive switching effect in RE-Doped cobalt ferrite nanoparticles

Spinel ferrite along with numerous important properties, also exhibit resistive switching effect. Cerium (Ce) doped Co-ferrite nanoparticles having composition CoFe2-xCexO4 (where x = 0.0, 0.05, 0.15, 0.20) were synthesized by co-precipitation synthesis route. X-ray Diffraction analysis confirmed the formation of cubic spinel phase with minor amount of impurities. Frequency dependent dielectric and electrical properties were studied at room temperature within frequency range of 20 Hz to 3 MHz. Dielectric constant, dielectric losses and AC conductivity decreases with cerium substitutions which is usual trend observed in ferrite materials. Modified Debye function was employed to fit Dielectric constant (ε′)- frequency (f) data. Jonscher's power law was employed to study AC conductivity. DC resistivity decreased with increasing temperature. DC resistivity and drift mobility decreased with cerium substitution while activation energy increased. Current–Voltage (I–V) curves of the synthesized composition were studied at room temperature. Current–Voltage (I–V) curves showed nonlinear hysteresis loop like behavior. Current follow different values for forward and reversed applied DC voltage which confirmed that resistive switching effect exit in studied samples. Sample with Ce composition x = 0.25 exhibit larger hysteresis loop than that of other compositions. This shows that sample for Ce composition x = 0.25, can be employed for nonvolatile Resistive Random-Access Memory (ReRAM) applications. Resistive switching effect was explained by utilizing Space-charge-limited current (SCLC) conduction model and by formation of a Schottky barrier at the metal-semiconductor interface.

Non-volatile memory based in-memory computing technology

By integrating the storage and computing functions on the fundamental elements, computing in-memory (CIM) technology is widely considered as a novel computational paradigm that can break the bottleneck of Von Neumann architecture. Nonvolatile memory device is an appropriate hardware implementation approach of CIM, which possess significantly advantages, such as excellent scalability, low consumption, and versatility. In this paper, first we introduce the basic concept of CIM, including the technical background and technical characteristics. Then, we review the traditional and novel nonvolatile memory devices, flash and resistive random access memory (RRAM), used in non-volatile based computing in-memory (nvCIM) system. After that, we explain the operation modes of nvCIM: in-memory analog computing and in-memory digital computing. In addition, the applications of nvCIM are also discussed, including deep learning accelerator, neuromorphic computing, and stateful logic. Finally, we summarize the current research advances in nvCIM and provide an outlook on possible research directions in the future.

Oscillatory Circuits With a Real Non-Volatile Stanford Memristor Model

Stanford memristor model is a widely used model that accurately characterizes real non-volatile metal-oxide resistive random access memory (RRAM) devices with bipolar switching characteristics. The paper studies for the first time the dynamics and bifurcations in a class of nonlinear oscillators with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">real non-volatile memristor devices</i> obeying Stanford model. This is in contrast with papers in the literature considering oscillators with ideal, abstract, or artificial memristor models, that are unable to describe physical memristors implemented in nanotechnology. One main new idea in the paper is to use the memristor as a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">programmable nonlinear resistor</i> . Namely, two principal modes of operation are considered. 1) Analogue transient phase: the oscillator is designed so that in the transient oscillations the voltage on the memristor is below threshold, hence the main memristor state variable, i.e., the gap of the insulating material, is almost constant and the memristor behaves as a static nonlinear resistor. 2) Programming phase: the nonlinear characteristic of the memristor, which depends on the gap, can be changed via the application of voltages above threshold. The paper studies nonlinear oscillations in the transient phase for a fixed gap as well as the bifurcations phenomena displayed when the gap is varied. The paper also discusses the differences between the approach in the paper and those to design other memristor oscillators with non-volatile memristors.