Jinfeng Kang

Characteristics and mechanism of conduction/set process in TiN∕ZnO∕Pt resistance switching random-access memories

The characteristics and mechanism of conduction/set process in TiN∕ZnO∕Pt-based resistance random access memory devices with stable and reproducible nanosecond bipolar switching behavior were studied. The dependencies of memory behavior on cell area, operating temperature, and frequency indicate that the conduction mechanism in low-resistance states is due to electrons hopping through filament paths. We also identify that the set process is essentially equivalent to a soft dielectric breakdown associated with a polarization effect caused by the migration of space charges under a low electric field stress. The generation/recovery of oxygen vacancies and nonlattice oxygen ions play a critical role in resistance switching.

A Low Energy Oxide-Based Electronic Synaptic Device for Neuromorphic Visual Systems with Tolerance to Device Variation

Neuromorphic computing is an emerging computing paradigm beyond the conventional digital von Neumann computation. An oxide-based resistive switching memory is engineered to emulate synaptic devices. At the device level, the gradual resistance modulation is characterized by hundreds of identical pulses, achieving a low energy consumption of less than 1 pJ per spike. Furthermore, a stochastic compact model is developed to quantify the device switching dynamics and variation. At system level, the performance of an artificial visual system on the image orientation or edge detection with 16 348 oxide-based synaptic devices is simulated, successfully demonstrating a key feature of neuromorphic computing: tolerance to device variation.

HfOx-Based Vertical Resistive Switching Random Access Memory Suitable for Bit-Cost-Effective Three-Dimensional Cross-Point Architecture

The three-dimensional (3D) cross-point array architecture is attractive for future ultra-high-density nonvolatile memory application. A bit-cost-effective technology path toward the 3D integration that requires only one critical lithography step or mask for reducing the bit-cost is demonstrated in this work. A double-layer HfOx-based vertical resistive switching random access memory (RRAM) is fabricated and characterized. The HfOx thin film is deposited at the sidewall of the predefined trench by atomic layer deposition, forming a vertical memory structure. Electrode/oxide interface engineering with a TiON interfacial layer results in nonlinear I-V suitable for the selectorless array. The fabricated HfOx vertical RRAM shows excellent performances such as reset current (<50 μA), switching speed (<100 ns), switching endurance (>10(8) cycles), read disturbance immunity (>10(9) cycles), and data retention time (>10(5) s @ 125 °C).

Unified Physical Model of Bipolar Oxide-Based Resistive Switching Memory

A unified model is proposed to elucidate the resistive switching behavior of metal-oxide-based resistive random access memory devices using the concept of electron hopping transport along filamentary conducting paths in dielectric layer. The transport calculation shows that a low-electron-occupied region along the conductive filament (CF) is formed when a critical electric field is applied. The oxygen vacancies in this region are recombined with oxygen ions, resulting in rupture of the CFs. The proposed mechanism was verified by experiments and theoretical calculations. In this physical model, the observed resistive switching behaviors in the oxide-based systems can be quantified and predicted.

Gd-doping effect on performance of HfO2 based resistive switching memory devices using implantation approach

An implantation doping approach is implemented to fabricate Gd-doped HfO2 resistive random access memory (RRAM) devices. The significantly enhanced performances are achieved in the Gd-doped HfO2 RRAM devices including improved uniformity of switching parameters, enlarged ON/OFF ratio, and increased switching speed without obvious reliability degradation. This performance improvement in the Gd-doped HfO2 RRAM devices is clarified to the suppressed randomicity of oxygen vacancy filaments’ formation and the reduced oxygen ion migration barrier induced by trivalent Gd-doping effect. The achieved results also demonstrate the validity of implantation doping approach for the fabrication of RRAM devices.

Ionic doping effect in ZrO2 resistive switching memory

Oxygen vacancy (VO) plays the critical role for resistive switching in transition metal oxide resistive random access memory (RRAM). First principles calculation is performed to study the impact of metallic ion (Al, Ti, or La) doping in ZrO2 on the behaviors of VO, including defect energy level and formation energy (Evf). Trivalent dopant (Al or La) significantly reduces Evf. Based on the calculated results, ZrO2-based RRAM devices are designed to control the formation of VO, and improved resistive switching uniformity is demonstrated in experiments.

Oxide-based RRAM switching mechanism: A new ion-transport-recombination model

This paper presents a unified physical model to elucidate the resistive switching behavior of metal-oxide-based resistive random access memory (RRAM) devices using the ion-transport-recombination model. In this model, the rupture of conductive filaments is caused by recombination of oxygen ions and electron-low-occupied oxygen vacancies. The transport equations of interstitial oxygen ions in the oxide matrix are introduced to quantitatively investigate the reset speed and other properties such as uniformity, endurance, and reset current. The proposed mechanism was verified by experiments.

HfOx based vertical resistive random access memory for cost-effective 3D cross-point architecture without cell selector

Double-layer stacked HfOx vertical RRAM is demonstrated for 3D cross-point architecture using a cost-effective fabrication process. Electrode/oxide interface engineering using TiON layer results in non-linear I-V suitable for the selector-less array. The fabricated HfOx vertical RRAM shows excellent performances such as reset current (<;50μA), switching speed (~50ns), switching endurance (>108 cycles), half-selected read disturbance immunity (>109 cycles), retention (>105s @125oC). Moreover, a unique write/read scheme is proposed for 3D cross-point architecture. Analysis shows that for such 3D selector-less array, a large Ron (~100kΩ) from the non-linear I-V helps reduce the sneak path current, and a low interconnect resistance using metal planes as word lines reduces the undesirable voltage drop on the interconnect. As a conservative estimate, simulation shows that Mb-scale array without cell selector is achievable.

Bipolar switching behavior in TiN/ZnO/Pt resistive nonvolatile memory with fast switching and long retention

Highly stable bipolar resistive switching behaviors of TiN/ZnO/Pt devices were demonstrated for the first time. The excellent memory characteristics including fast switching speed (<20 ns for set and <60 ns for reset), long retention (in the order of 105 s) and non-electroforming process were demonstrated. The bipolar switching behaviors can be explained by formation and rupture of the filamentary conductive path consisting of oxygen vacancies. The excellent bipolar switching behavior can be attributed to the significant amount of oxygen vacancies in ZnO film and the effect of TiN layer serving as an oxygen reservoir.

A neuromorphic visual system using RRAM synaptic devices with Sub-pJ energy and tolerance to variability: Experimental characterization and large-scale modeling

We report the use of metal oxide resistive switching memory (RRAM) as synaptic devices for a neuromorphic visual system. At the device level, we experimentally characterized the gradual resistance modulation of RRAM by hundreds of identical pulses. As compared with phase change memory (PCM) reported recently in [1,2], >100×-1000× energy consumption reduction was achieved in RRAM as synaptic devices (<;1 pJ per spike). Based on the experimental results, we developed a stochastic model to quantify the device switching dynamics. At the system level, we simulated the performance of image orientation selectivity on a neuromorphic visual system which consists of 1,024 CMOS neuron circuits and 16,348 RRAM synaptic devices. It was found that the system can tolerate the temporal and spatial variability which are commonly present in RRAM devices, suggesting the feasibility of large-scale hardware implementation of neuromorphic system using RRAM synaptic devices.