Threshold Switching Devices Research Articles

Investigation of recessed‐source/drain SOI feedback FET‐based integrate and fire neuron circuit with compact model of threshold switching devices

AbstractIn this article, the investigation of recessed‐source/drain (Re‐S/D) SOI feedback FET (FBFET)‐based integrate and fire (IF) neuron circuit parameters is presented using a threshold switching device compact model. FBFETs offer high ION and low SS with minimal power consumption, operating efficiently at lower voltages and currents than conventional MOSFETs. Utilizing ION/IOFF ratio and threshold voltage limits (Vt2/Vt1) of the device, a model is developed to mimic hysteresis characteristics, which is then used to implement an IF neuron circuit. Our findings show that altering the Re‐S/D thickness between 0 and 50 nm enhances the ION of the device under study while decreasing hysteresis width. We detected a significant increase in output spike frequency of 46.8% and 65.14% for input current pulse amplitudes of 5 and 20 nA, respectively. Furthermore, increasing the Re‐S/D thickness from 0 to 50 nm led to a significant 29.97% enhancement in spike amplitude. In addition, when using input current pulse amplitudes of 5 and 20 nA, we saw energy savings per spike of 3.36% and 12.7%, respectively. At the same time, there was an increase in power of 8.69% and 9.54%. These enhancements in performance metrics establish our proposed integrate and fire neuron circuit as a promising candidate for efficient neuromorphic system implementation.

Integration of Ag-based threshold switching devices in silicon microchips

Threshold-type resistive switching (RS) is an essential electronic behavior in many types of integrated circuits and can be exploited in multiple applications, such as leaky integrate-and-fire neurons for artificial neural networks (ANNs). Many research articles have shown that metal/insulator/metal (MIM) devices using Ag electrodes exhibit stable threshold-type RS, but all of them presented large devices (>1 µm2) fabricated on unfunctional SiO2/Si substrates. In this article, for the first time we integrate Ag-based threshold-type RS devices at the back-end-of-line interconnections of silicon microchips. The insulator used is multilayer hexagonal boron nitride (h-BN), and the size of the devices is ∼0.05 µm2. The devices can switch between a high resistive state (HRS) and a low resistive state (LRS) without the need of any forming process, and we observe a high endurance over 1 million cycles over multiple devices. By performing circuit simulation using SPICE software, we confirm that this electrical behavior is suitable for being used as leaky integrate-and-fire electronic neuron in spiking neural networks for image recognition, and the h-BN artificial neurons operate correctly for 94 % of the images presented. Our study represents a significant advancement towards the integration of Ag-based threshold-type RS devices in silicon microchips.

Local bandgap narrowing in the forming state of threshold switching materials

Threshold switching (TS) materials, such as amorphous chalcogenide, have received significant attention for their application in storage class memory and in-memory computing. These materials contribute to efficient data processing and reduced power consumption in data centers. The initial switching process after fabricating a TS device, known as “forming,” has a profound impact on its subsequent TS behavior. However, it remains unclear how TS materials undergo changes in their atomic and electronic structures during the forming process. Consequently, the key factors that govern TS behavior remain obscure, necessitating a deeper understanding of the underlying physics behind TS phenomena. In this Letter, we investigated the forming state of the TS material AlTeN by combining scanning internal photoemission microscopy (SIPM) and ab initio calculations. Thanks to nondestructive evaluation by SIPM measurements, we observed local bandgap narrowing of AlTeN after its forming process. This is an experimental demonstration showing the presence of nuclei of the conductive filament formed in its ON state. Moreover, we conducted an ab initio calculation to reveal the origin of bandgap narrowing. We applied strong electrothermal stresses to the AlTeN model by ab initio molecular dynamics simulation with high electronic and lattice temperatures. By quenching from the electrothermal stress conditions, we reproduced an experimentally observed forming state with a narrowed bandgap. Analysis of the electronic structures of the forming state revealed that the origin of bandgap narrowing is the generation of the valence band top and conduction band bottom stemming from the increased homopolar bonds.

Mechanism for Local-Atomic Structure Changes in Chalcogenide-based Threshold-Switching Devices.

Threshold-switching devices based on amorphous chalcogenides are considered for use as selector devices in 3D crossbar memories. However, the fundamental understanding of amorphous chalcogenide is hindered owing to the complexity of the local structures and difficulties in the trap analysis of multinary compounds. Furthermore, after threshold switching, the local structures gradually evolve to more stable energy states owing to the unstable homopolar bonds. Herein, based on trap analysis, DFT simulations, and operando XPS analysis, it is determined that the threshold switching mechanism is deeply related to the charged state of Se-Se homopolar defects. A threshold switching device is demonstrated with an excellent performance through the modification of the local structure via the addition of alloying elements and investigating the time-dependent trap evolution. The results concerning the trap dynamics of local atomic structures in threshold switching phenomena may be used to improve the design of amorphous chalcogenides.

Low-power perovskite-based threshold switching memristor for artificial nociceptor

Organic-inorganic halide perovskites (OHPs) with good ion diffusion paths are effective functional materials for simulating biological neurons. In this work, we utilize the Ag diffusion threshold switching phenomenon in FAPbI3 films to investigate the nociceptor function. The threshold switching device exhibits a low threshold voltage (0.075 V) and a large characteristic switching ratio of 107 in volatile resistance switching, along with a minimum switching slope (SS< 2.6 mV/dec), and the device achieves low power consumption of 150 pW. Importantly, the nociceptor functions of threshold firing and sensitization have been successfully simulated based on the volatile threshold switching characteristics, which demonstrates the potential in practical neuromorphic applications.

Volatile threshold switching devices for hardware security primitives: Exploiting intrinsic variability as an entropy source

In the age of the Internet of Things, the proliferation of edge devices has resulted in a significant increase in personal information that is susceptible to theft and counterfeiting at various stages of data communication. As a result, substantial attention has been focused on hardware (HW) security elements, such as the true random number generator and physical unclonable function. With the recent surge in research and development of emerging memristors, which exploit the inherent variability of these devices, there has been a notable increase in studies on HW security. Particularly, volatile threshold switch (TS) devices, which exhibit insulator/metal characteristics below/above a certain threshold voltage, show great promise as security devices due to their lower power consumption and higher cycling endurance compared to nonvolatile memory devices. Despite the promising attributes and increasing demand for TS devices for HW security, there remains a lack of a comprehensive overview covering various TS devices and their potential contributions to HW privacy. To address this gap, this review provides an encompassing analysis of different types of TS devices and their performance in HW security literature, providing insight into current limitations and the future prospects of HW security primitives based on TS devices.

Self‐Healing Magnetic Field‐Assisted Threshold Switching Device Utilizing Dual Field‐Driven Filamentary Physics

AbstractAdvanced filamentary devices are crucial for developing low‐power devices to implement high‐speed logic and neuromorphic devices. Among these, HfO2‐based filamentary devices have attracted attention as viable options due to their threshold‐switching characteristics and compatibility with complementary metal‐oxide‐semiconductor (CMOS) technology. However, the unpredictability of conventional filament formation/rupture driven by an electric field challenges consistency and reliability. A paradigm shift from conventional stochastic electric field‐driven ion migration to controllable ion‐based transportation is essential to devise functional low‐power devices capable of controlling the filament process. This work introduces a magnetic field‐assisted threshold switching (MA‐TS) device, which integrates a neodymium magnet and a nickel (Ni) barrier layer to enable controlled dual field‐driven ion transportation. The dual field‐driven process combining the conventional vertical electric field‐driven ion migration with lateral magnetic field‐driven ion transportation, reveals a distinctive aspect of ion movement. The MA‐TS device achieves superior performances characterized by an ultra‐low threshold voltage (≈0 V), minimized leakage current in the off‐state, a variation‐immune hysteresis‐free characteristic, enhanced yield, and revival‐ability (i.e., self‐healing) after a failed TS operation. By overcoming the limitations of conventional filamentary devices, the MA‐TS device opens up a promising avenue for efficient and stable low‐power applications.

First integration of Ni barrier layer for enhanced threshold switching characteristics in Ag/HfO2-based TS device

Utilizing Ag/HfO2 with nickel (Ni) as a barrier layer, a novel threshold switching (TS) device is devised to overcome challenges such as low reliability, high threshold voltage, and high leakage current. Compared against an Ag/Ti/HfO2-based TS device, the Ag/Ni/HfO2-based TS device exhibits improved electrical characteristics: yield enhancement from 31.7 % to 40.0 %, enhanced endurance from ∼10 cycles to ∼300 cycles, and suppression in off-state current (IOFF) from 1.2 × 10−7 A to 5.2 × 10−10 A under a high compliance current (e.g., 10−4 A). The results obtained through transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDS), and atomic force microscopy (AFM) support the evidence of those accomplishments. Reducing the effective area of the TS device improves control over erratically generated filaments and the electric field within the switching layer, resulting in enhanced performance such as a reduced threshold voltage (VTH ∼0.35 V), minimized VTH variability (∼0.01 V), decreased a threshold current (ITH, i.e., the leakage current in the off-state before activation, ∼5.2 × 10−10 A), and maximum conductance (∼5.0 × 10−5 S) of low-resistance state. These findings suggest that the optimized Ag/Ni/HfO2-based TS device can serve as a practical solution for low-power applications.

A ternary gate-connected threshold switching thin-film transistor

Multi-valued logic has been a significant focus of research in various fields with the advancement of information technology. One approach to realizing ternary logic is integrating of a threshold switching (TS) device with a transistor, but this method often entails a complex fabrication process. This work suggests a ternary gate-connected threshold switching thin-film transistor (TS-TFT) by serially connecting the TS device with a bottom-gate thin-film transistor (TFT). The fabrication process is simplified with a structure that shares electrodes and insulators. Different threshold voltages from TS and TFT devices produce stable multiple states. The Pt/HfO2/TiN TS device has an electronic trapping/detrapping switching mechanism that exhibits low power consumption and high reliability. With the superior electrical performance of an amorphous indium gallium zinc oxide TFT, the TS-TFT has stable endurance. Furthermore, pulse switching and ternary inverter are demonstrated from the practical point of view.

Steep slope threshold switching field‐effect transistors based on 2D heterostructure

AbstractIn dealing with the increasing power dissipation of electronic systems with increasing integration density, a field‐effect transistor (FET) with steep switching slope that overcomes the thermionic limit is vital to achieve low‐power operations. Here, we report two types of threshold switching (TS) FETs based on 2D Van der Waals heterostructures by virtue of the abrupt resistive switching of the hexagonal boron nitride (hBN) TS device. The common hBN dielectric layer functions as the switching medium for the TS device and the gate dielectric for the 2D FET enabling seamless integration of the hBN TS device and baseline 2D FET. TS FET in source configuration by connecting the TS device to the source terminal of the 2D FET offers an ultralow average subthreshold swing (SS) of 1.6 mV/dec over six decades of drain current at room temperature and suppressed leakage current. TS FET in gate configuration by connecting the TS device to the gate terminal of the 2D FET also exhibits steep switching slope with ultralow SS of 10.6 mV/dec. The proposed compact device structures integrating 2D FET and TS device provide a potential approach of monolithic integration toward next‐generation low‐power electronics.

GeSe ovonic threshold switch: the impact of functional layer thickness and device size

Three-dimensional phase change memory (3D PCM), possessing fast-speed, high-density and nonvolatility, has been successfully commercialized as storage class memory. A complete PCM device is composed of a memory cell and an associated ovonic threshold switch (OTS) device, which effectively resolves the leakage current issue in the crossbar array. The OTS materials are chalcogenide glasses consisting of chalcogens such as Te, Se and S as central elements, represented by GeTe6, GeSe and GeS. Among them, GeSe-based OTS materials are widely utilized in commercial 3D PCM, their scalability, however, has not been thoroughly investigated. Here, we explore the miniaturization of GeSe OTS selector, including functional layer thickness scalability and device size scalability. The threshold switching voltage of the GeSe OTS device almost lineally decreases with the thinning of the thickness, whereas it hardly changes with the device size. This indicates that the threshold switching behavior is triggered by the electric field, and the threshold switching field of the GeSe OTS selector is approximately 105 V/μm, regardless of the change in film thickness or device size. Systematically analyzing the threshold switching field of Ge–S and Ge–Te OTSs, we find that the threshold switching field of the OTS device is larger than 75 V/μm, significantly higher than PCM devices (8.1–56 V/μm), such as traditional Ge2Sb2Te5, Ag–In–Sb–Te, etc. Moreover, the required electric field is highly correlated with the optical bandgap. Our findings not only serve to optimize GeSe-based OTS device, but also may pave the approach for exploring OTS materials in chalcogenide alloys.

A self-selecting memory element based on a method of interconnected ovonic threshold switching device

A self-selecting memory element based on a method of interconnected ovonic threshold switching device

Ovonic threshold switching-based artificial afferent neurons for thermal in-sensor computing.

Artificial afferent neurons in the sensory nervous system inspired by biology have enormous potential for efficiently perceiving and processing environmental information. However, the previously reported artificial afferent neurons suffer from two prominent challenges: considerable power consumption and limited scalability efficiency. Herein, addressing these challenges, a bioinspired artificial thermal afferent neuron based on a N-doped SiTe ovonic threshold switching (OTS) device is presented for the first time. The engineered OTS device shows remarkable uniformity and robust endurance, ensuring the reliability and efficacy of the artificial afferent neurons. A substantially decreased leakage current of the SiTe OTS device by nitrogen doping results in ultra-low power consumption less than 0.3 nJ per spike for artificial afferent neurons. The inherent temperature response exhibited by N-doped SiTe OTS materials allows us to construct a highly compact artificial thermal afferent neuron over a wide temperature range. An edge detection task is performed to further verify its thermal perceptual computing function. Our work provides an insight into OTS-based artificial afferent neurons for electronic skin and sensory neurorobotics.

Improved Switching Speed and Current Density of Conductive Bridge Threshold Switch Devices by Ag$_{\text{2}}$Se Layer

Improved Switching Speed and Current Density of Conductive Bridge Threshold Switch Devices by Ag$_{\text{2}}$Se Layer

Threshold Switching of ALD-NbOx Films for Neuromorphic Applications

Neuromorphic architecture has been suggested as an alternative to the existing von Neumann counterpart. The neuromorphic approach allows massively parallel processing and asynchronous timing schemes with low power consumption. This work presents ALD- NbOx thin films as a potential material for neuromorphic computation. NbO2 shows metal-insulator transition which is the desired property for threshold switching (TS). However, direct forming of NbO2 is rather difficult and deposited NbOx tends predominantly to result in Nb2O5. To obtain NbO2 we used an oxygen scavenger layer of Ti to alter Nb2O5 to NbO2. After a proper electroforming process, the device with a Ti insert showed metal-insulator transition above a specific voltage, i.e., a threshold voltage, and a stable threshold switching characteristics with wide operation range, characteristics not observed in the absence of a Ti insert. The well-defined TS characteristics clearly indicated the Ti role in controlling the oxidation state of NbOx.To verify the oxidation state of NbOx, XPS depth profile analyses were used. The presence of a Ti insertion layer resulted in an increase of 10~30% in Nb4+ states, depending on the total thickness of the NbOx films, while Nb5+ states decreased. Bearing in mind practical applications, we used only CMOS-compatible materials and processes to fabricate TS devices. Our approach suggests a reliable method to fabricate NbO2 neurons.Acknowledgment: This work was supported by Electronics and Telecommunications Research Institute (ETRI) grant funded by the Korean government under grant " Non-CMOS Neuromorphic Device Basic Technology." (22BB1110).

Excellent Reliability Characteristics of Ovonic Threshold Switch Device with Higher‐Temperature Forming Technique

Amorphous chalcogenide‐based ovonic threshold switch (OTS) has significant potential as a selector device in high‐density memory arrays. However, to realize the ideal selector device, OTS still needs to enhance its reliability characteristics such as device uniformity and threshold voltage (Vth) drift. Herein, the effect of elevated temperature forming on the selector characteristics of an OTS device is investigated. The findings indicate that the reliability performance of the device significantly improves when the forming process is conducted at higher temperature. In particular, it shows excellent Vth uniformity (σ = 32 mV) and a notable enhancement in the Vth drift characteristics. Based on trap analysis at cryogenic temperature (77 K), it is confirmed that improved reliability characteristics can be attributed to the stability of the activated trap formed by the HT forming process.

Strain-induced degradation and recovery of flexible NbOx-based threshold switching device

We investigate the functionality of NbOx-based selector devices on a flexible substrate. It was observed that the failure mechanism of cyclic tensile strain is from the disruption of atom arrangements, which essentially led to the crack formation of the film. When under cyclic compressive strain, buckling delamination of the film occurs as the compressed films have debonded from their neighboring layers. By implementing an annealing process after the strain-induced degradation, recovery of the device is observed with reduced threshold and hold voltages. The physical mechanism of the device is investigated through Poole–Frenkel mechanism fitting, which provides insights into the switching behavior after mechanical strain and annealing process. The result demonstrates the potential of the NbOx device in flexible electronics applications with a high endurance of up to 105 cycles of cyclic bending strain and the recovery of the device after degradation.

Phase transition induced threshold resistive switching in two-dimensional VTe2 nanosheets for Boolean logic operations

Vanadium chalcogenides have been extensively studied owing to the diverse crystallographic structures with various stoichiometric ratios. The metal-to-insulator transition (MIT) widely reported in vanadium chalcogenides is a rapid reversible phase transition that requires small energy, demonstrating potential applications in memory devices. In this work, two-dimensional (2D) vanadium telluride (VTe2) nanosheets are prepared by the chemical vapor deposition method. The synthesized VTe2 nanosheets exhibit volatile threshold switching (TS) behaviors due to the MIT phase transition, which can be further confirmed by the temperature dependent TS behaviors. The TS memristor demonstrates good stability and high reliability with up to 1000 continuous and repeatable writing/erasing operations. Furthermore, based on the TS behaviors, the fabricated memristor can be utilized to implement basic Boolean logic operations of “OR,” “AND,” and “NOT.” This study not only demonstrates the TS behaviors in the 2D VTe2 nanosheets owing to the MIT phase transition but also shows the potential applications of the TS devices in Boolean logic operations.

Neuromorphic computing devices based on the asymmetric temperature gradient

Neuromorphic computing devices, which emulate biological neural networks, are crucial in realizing artificial intelligence for information processing and decision-making. Different types of neuromorphic computing devices with varying resistance levels have been developed, such as oxide-based memristors caused by ion diffusion, phase transition-based devices caused by threshold switching, progressive crystallization/amorphization, and spintronics-based devices caused by magnetic domain switching. However, these devices face significant challenges, including disruptions in the reading process, limited scalability in integrated circuits, and non-linearity in weight change. To address these challenges, alternative approaches are required. In this study, we introduce a multi-layer-multi-terminal neuromorphic computing device based on the asymmetric temperature gradient. Our device exhibits a wide range of synaptic functions, including potentiation, depression, and both anti-symmetric and symmetric spike-timing-dependent plasticity. The thermal driving strategy offers an energy-efficient platform for future neuromorphic computing devices to achieve artificial intelligence.

Robust Threshold-Switching Behavior Assisted by Cu Migration in a Ferroionic CuInP2S6 Heterostructure.

The two-dimensional layered material CuInP2S6 (CIPS) has attracted significant research attention due to its nontrivial physical properties, including room-temperature ferroelectricity at the ultrathin limit and substantial ionic conductivity. Despite many efforts to control its ionic conductance and develop electronic devices, such as memristors, improving the stability of these devices remains a challenge. This work presents a highly stable threshold-switching device based on the Cu/CIPS/graphene heterostructure, achieved after a comprehensive investigation of the activation of Cu's ionic conductivity. The device exhibits exceptional threshold-switching performance, including good cycling endurance, a high on/off ratio of up to 104, low operation voltages, and an ultrasmall subthreshold swing of less than 1.8 mV/decade for the resistance-switching process. Through temperature-dependent electrical and Raman spectroscopy measurements, the stable resistive-switching mechanism is interpreted with a drifting and diffusion model of Cu ions under the electric field, rather than the conventional conducting filament mechanism. These results make the layered ferroionic CIPS material a promising candidate for information storage devices, demonstrating a compelling approach to achieving high-performance threshold-switching memristor devices.