Binder-Free Synthesis of Nanostructured Amorphous Cobalt Phosphate for Resistive Memory and Artificial Synaptic Device Applications

Abstract

The rise of artificial intelligence and machine learning demands versatile electronic devices for memory and braininspired computing applications. The electronic materials are the backbones of these applications. Considering this, a functional Co x (PO 4 ) 2 nanomaterial was synthesized for resistive memory and neuromorphic computing applications. The synthesized nanomaterial was well characterized by using X-ray diffraction, Fourier transform infrared spectroscopy, field emission-scanning electron microscopy, and X-ray photoelectron spectroscopy. The fabricated Ag/Co x (PO 4 ) 2 /ITO device shows bipolar resistive switching and memristive properties. The SET and RESET voltages were analyzed by using different statistical measures, and their distribution was studied by using the Weibull technique. The results suggested that the SET voltages were more uniformly distributed than the RESET voltage. The switching nonlinearity was modeled and predicted by using Holt's exponential smoothingbased statistical time series analysis method. In the case of nonvolatile memory tests, the device shows good endurance (10 3 cycles) and memory retention (3 × 10 4 s) with excellent memory window (1.7 × 10 3 ) properties. Moreover, the device can mimic the potentiation−depression and spike-timing-dependent plasticity-based Hebbian learning rules, suggesting Co x (PO 4 ) 2 is a potential nanomaterial for the fabrication of artificial synapse. The detailed analysis of electrical results suggested that the space-charge-limited current-based charge transport was responsible for the device conduction, whereas the formation and rupture of conductive filament(s) were responsible for the resistive switching in the Ag/Co x (PO 4 ) 2 /ITO memristive device. The results of the present investigation suggested that the Co x (PO 4 ) 2 nanomaterial is a potential candidate for resistive memory and brain-inspired computing applications

Surface Modification of a Titanium Carbide MXene Memristor to Enhance Memory Window and Low‐Power Operation

“…[43] TSA is a critical statistical technique used in modeling and predicting time/cyclic series data. [46] In this study, we modeled and predicted the V SET and V RESET of the Ag/MXene/ Pt device. At the outset, we verified the stationarity properties of the switching voltages using the Dickey-Fuller (ADF) test.…”

“…In the neurobiological computing system, the pre-synaptic neuron transmits the signals to the post-synaptic neuron through the synaptic junctions. When the external stimuli are applied to the pre-synaptic neuron, ions will enter into the post-synaptic neuron to form the conducting filaments and this phenomenon facilitates memory and learning by generating action potentials [12,46,57,58] We emulated biological synapses by fabricating a two-terminal Ag/MXene/Pt memristor device. Herein, additional bio-synaptic characteristics such as EPSC, IPSC, PPF, and PPD were mimicked to evaluate the suitability of fabricated Ag/ MXene/Pt artificial synapses for neuromorphic computing applications.…”

Promising Materials and Synthesis Methods for Resistive Switching Memory Devices: A Status Review

“…Overall device performance is decided by many factors; however, the active switching layer is one of the major aspects of all. In recent years, our research group actively engaged in a synthesizing variety of switching layer materials by using different synthesis techniques for nonvolatile memory and neuromorphic computing applications. ,,− ,− There are three main categories of nanomaterial/thin film synthesis viz. physical, chemical, and biological methods.…”

Ultrafine Nickel Sulfide-Based Bipolar Resistive Switching Device as Artificial Synapses for Neuromorphic Application

“…Nickel-based materials, particularly nickel oxides, have been widely investigated as a resistive switching material for high-density memory applications. − The resistive switching devices are considered as artificial synapses upon the application of low-power electrical stimuli . Reports have been published on boron nitride, cobalt phosphate, copper oxide, zinc oxide, tantalum oxide, and titanium dioxide nanotube-based devices for high-density memory and/or neuromorphic computing applications. − The nickel-based sulfide materials have wide applications in dye-sensitized solar cells, supercapacitors, rechargeable batteries, electrochemical sensors, oxygen evolution reactions, and photocatalysis. − However, the neuromorphic performance study of nickel sulfide-based materials is still unexplored. For this reason, we have considered nickel sulfide nanomaterials for neuromorphic computing applications by fabricating a metal–insulator–metal type of device.…”