Formation Of Metallic Filaments Research Articles

Atomistic description of conductive bridge formation in two-dimensional material based memristor

In-memory computing technology built on 2D material-based nonvolatile resistive switches (aka memristors) has made great progress in recent years. It has however been debated whether such remarkable resistive switching is an inherent property of the 2D materials or if the metal electrode plays any role? Can the metal atoms penetrate through the crystalline 2D materials to form conductive filaments as observed in amorphous oxide-based memristors? To find answers, here we investigate MoS2 and h-BN-based devices with electrochemically passive and active (metal) electrodes using reactive molecular dynamics with a charge equilibration approach. We find that the SET and RESET processes in active electrode-based multilayer devices involve the formation and disruption of metal filaments linking the two electrodes exclusively through the grain boundaries, the configuration of which affects the volatility of the resistive switching. Whereas the switching mechanisms in passive electrode-based devices require the formation of interlayer B-N bonds and popping of the S atom to the Mo plane at the point defects. We also show that metal atom adsorption at the point defects causes resistive switching in monolayer MoS2. Our atomic-level understanding provides explanations to the apparently contradictory experimental findings and enables defect-engineering guidelines in 2D materials for such disruptive technology.

Multi-factors-regulated multi-level down-scalable and robust memristors

The memristor based on resistive switching (RS) behavior has attracted extensive attention due to its wide application prospects in artificial intelligence (AI) and information memory. In this work, a multi-factor responsive memristor based on Ag/MnO2/F-doped tin oxide (FTO) sandwich structure was fabricated by magnetic sputtering method. It was found that the RS behavior of the device was significantly improved when the diameter of the top electrode decreased from 2.0 mm to 0.5 mm, indicating that the device with Ag/MnO2/FTO structure has the down-scalable potential. Through comprehensive research, it can be concluded that the bipolar RS behavior of the memristor can be adjusted by multi-factors such as active layer thickness, top electrode size, voltage scanning rate, bias voltage amplitude and temperature. Specifically, the resistance-temperature dependence and temperature coefficient of the device in the low resistance state (LRS) indicate that the formation and disconnection of metallic Ag conductive filaments (CFs) play an important role in the RS behavior of the memristive device. Therefore, this work will lay an experimental and theoretical foundation for comprehensively understanding the mechanism of RS behavior and developing high-performance memristor with multi-level memory capacity.

Formation of Cluster-Structured Metallic Filaments in Organic Memristors for Wearable Neuromorphic Systems with Bio-Mimetic Synaptic Weight Distributions.

With increasing demand for wearable electronics capable of computing huge data, flexible neuromorphic systems mimicking brain functions have been receiving much attention. Despite considerable efforts in developing practical neural networks utilizing several types of flexible artificial synapses, it is still challenging to develop wearable systems for complex computations due to the difficulties in emulating continuous memory states in a synaptic component. In this study, polymer conductivity is analyzed as a crucial factor in determining the growth dynamics of metallic filaments in organic memristors. Moreover, flexible memristors with bio-mimetic synaptic functions such as linearly tunable weights are demonstrated by engineering the polymer conductivity. In the organic memristor, the cluster-structured filaments are grown within the polymer medium in response to electric stimuli, resulting in gradual resistive switching and stable synaptic plasticity. Additionally, the device exhibits the continuous and numerous non-volatile memory states due to its low leakage current. Furthermore, complex hardware neural networks including ternary logic operators and a noisy image recognitions system are successfully implemented utilizing the developed memristor arrays. This promising concept of creating flexible neural networks with bio-mimetic weight distributions will contribute to the development of a new computing architecture for energy-efficient wearable smart electronics.

Highly Reliable 3D Channel Memory and Its Application in a Neuromorphic Sensory System for Hand Gesture Recognition.

Brain-inspired neuromorphic computing systems, based on a crossbar array of two-terminal multilevel resistive random-access memory (RRAM), have attracted attention as promising technologies for processing large amounts of unstructured data. However, the low reliability and inferior conductance tunability of RRAM, caused by uncontrollable metal filament formation in the uneven switching medium, result in lower accuracy compared to the software neural network (SW-NN). In this work, we present a highly reliable CoOx-based multilevel RRAM with an optimized crystal size and density in the switching medium, providing a three-dimensional (3D) grain boundary (GB) network. This design enhances the reliability of the RRAM by improving the cycle-to-cycle endurance and device-to-device stability of the I-V characteristics with minimal variation. Furthermore, the designed 3D GB-channel RRAM (3D GB-RRAM) exhibits excellent conductance tunability, demonstrating high symmetricity (624), low nonlinearity (βLTP/βLTD ∼ 0.20/0.39), and a large dynamic range (Gmax/Gmin ∼ 31.1). The cyclic stability of long-term potentiation and depression also exceeds 100 cycles (105 voltage pulses), and the relative standard deviation of Gmax/Gmin is only 2.9%. Leveraging these superior reliability and performance attributes, we propose a neuromorphic sensory system for finger motion tracking and hand gesture recognition as a potential elemental technology for the metaverse. This system consists of a stretchable double-layered photoacoustic strain sensor and a crossbar array neural network. We perform training and recognition tasks on ultrasonic patterns associated with finger motion and hand gestures, attaining a recognition accuracy of 97.9% and 97.4%, comparable to that of SW-NN (99.8% and 98.7%).

(Invited) Design and Modeling of Resistive Switching Devices for Neuromorphic Applications

Resistive switching devices are a critical component for the realization of computational systems that mimic biological neural systems. In biological systems, connection weights between the neurons are dependent on the time lapse between the neurons’ action potentials. Resistive switching devices (memristors) with multiple conductance states can mimic this biological spike-time-dependent plasticity (STDP) behaviour, with multiple conductance states being analogous to different synaptic weights. Additionally, memristors are one of the contenders as storage elements in non-volatile memories. The change in conductance in memristors can be broadly attributed to vacancy migration inside the switching layer (valence change memory) or the formation of metallic filaments inside the switching layer (electrochemical metallization).Electrochemical metallization-based memristors with vertical transport and small inter-electrode distances have been reported recently. Their device characteristics exhibit multiple conductance states with relatively low switching voltages, which makes them well-suited for low-power neuromorphic applications. Our work models the transport in these memristors, focusing on explaining and capturing their current-voltage characteristics. Our model also captures the switching dynamics (with an emphasis on estimating switching energies and delays) and explains the experimentally observed STDP behaviour in these devices. We have proposed models for filament growth and dissolution along one dimension (axial) and two dimensions (axial and radial). The simulation results obtained using our model (and implemented in Verilog-A) have been validated with experimental data from multiple sources. Our work demonstrates the flexibility of including different transport mechanisms in a unified framework.We have characterized valence change memory (VCM) memristors with HfO2 and HfZrO2 as switching layers. They exhibit bipolar switching, negative differential resistance, multiple conductance states, and STDP behaviour, which have been analyzed. We have also modeled the transport in these VCM-based memristors, and simulated the I-V and dynamic switching characteristics, and STDP behaviour in these devices. We have simulated arrays of, both, ECM and VCM devices using the proposed models, and tried to demonstrate a typical pattern classification problem using in-memory computation. We believe that our models provide useful insight into the design of memristors for neuromorphic and memory applications.

Enhanced resistive switching of silver copper iodide thin films prepared by interfacial phase formation

Resistive random-access memory (ReRAM), an alternative to conventional charge storage memory, employs the switching of a resistive material between high-resistance and low-resistance states, which can be generated by the formation/dissolution of metal filaments bridging the two electrodes in the cell. Silver iodide (AgI) is a promising solid electrolyte for ReRAM cells, but its application is hindered by poor photostability and low electrical conductivity. Herein, a Cu-substituted β-AgI system, β′-Ag0.7Cu0.3I, was successfully prepared via sequential deposition of a Cu metal layer onto a AgI thin film by a galvanic reaction, which induced the spontaneous incorporation of Cu+ cations into the AgI lattice. Cu substitution at the Cu–AgI heterojunction interface is a novel way to modulate the chemical composition and improve the resistive switching properties of intrinsic polycrystalline AgI thin films. We compared the resistive switching properties of a ReRAM thin film devices based on β′-Ag0.7Cu0.3I with those based on pristine β-AgI. This device exhibited enhanced performance with high ON/OFF ratio (∼104) and low working voltage compared with the β-AgI-based device. These results strongly suggest that binary metal halide materials can serve as simple model systems for the efficient formation of metal filaments and have potential low-cost, low-power memory applications.

Simulating the filament morphology in electrochemical metallization cells

Electrochemical metallization (ECM) cells are based on the principle of voltage controlled formation or dissolution of a nanometer-thin metallic conductive filament (CF) between two electrodes separated by an insulating material, e.g. an oxide. The lifetime of the CF depends on factors such as materials and biasing. Depending on the lifetime of the CF—from microseconds to years—ECM cells show promising properties for use in neuromorphic circuits, for in-memory computing, or as selectors and memory cells in storage applications. For enabling those technologies with ECM cells, the lifetime of the CF has to be controlled. As various authors connect the lifetime with the morphology of the CF, the key parameters for CF formation have to be identified. In this work, we present a 2D axisymmetric physical continuum model that describes the kinetics of volatile and non-volatile ECM cells, as well as the morphology of the CF. It is shown that the morphology depends on both the amplitude of the applied voltage signal and CF-growth induced mechanical stress within the oxide layer. The model is validated with previously published kinetic measurements of non-volatile Ag/SiO2/Pt and volatile Ag/HfO2/Pt cells and the simulated CF morphologies are consistent with previous experimental CF observations.

Resistive Switching and Current Conduction Mechanisms in Hexagonal Boron Nitride Threshold Memristors with Nickel Electrodes

AbstractThe 2D insulating material hexagonal boron nitride (h‐BN) has attracted much attention as the active medium in memristive devices due to its favorable physical properties, among others, a wide bandgap that enables a large switching window. Metal filament formation is frequently suggested for h‐BN devices as the resistive switching (RS) mechanism, usually supported by highly specialized methods like conductive atomic force microscopy (C‐AFM) or transmission electron microscopy (TEM). Here, the switching of multilayer hexagonal boron nitride (h‐BN) threshold memristors with two nickel (Ni) electrodes is investigated through their current conduction mechanisms. Both the high and the low resistance states are analyzed through temperature‐dependent current–voltage measurements. The formation and retraction of nickel filaments along boron defects in the h‐BN film as the resistive switching mechanism is proposed. The electrical data are corroborated with TEM analyses to establish temperature‐dependent current–voltage measurements as a valuable tool for the analysis of resistive switching phenomena in memristors made of 2D materials. The memristors exhibit a wide and tunable current operation range and low stand‐by currents, in line with the state of the art in h‐BN‐based threshold switches, a low cycle‐to‐cycle variability of 5%, and a large On/Off ratio of 107.

Temperature-modulated switching behaviors of diffusive memristor for biorealistic emulation of synaptic plasticity

Temperature is known as an important factor in biological synaptic transmission. In this study, temperature-modulated switching behaviors are reported in an amorphous carbon (a-C) diffusive memristor device to emulate biorealistic synaptic plasticity. The devices exhibit memory switching and threshold switching behaviors depending on the compliance current and ambient temperature. As confirmed by conducting atomic force microscopy, the thermal effect can promote the electrochemical formation of a stable metallic conductive filament. A series of timing-controlled pulse experiments are carried out to study the temperature effect on the switching characteristics, and the device shows second-order memristive behaviors. Frequency-dependent synaptic plasticity and timing-controlled spike-time-dependent plasticity are demonstrated in the device, which are analogous to the synaptic strength in a biological synapse at elevated temperatures. As a proof of concept, the forgetting behavior of numerical images learned at different temperatures and different pulse durations is conceptually emulated with synaptic device arrays. It is expected the present device with second order memristive behaviors provides alternatives for biorealistic synaptic applications.

Random networks of disconnected nanoparticles in dielectric layers as a source of electric responsivity

Many efforts are currently focused on materials with responsive features for neural-inspired devices. Different approaches are followed, based on various mechanisms – from ferroelectric switching to structural phase changes, from magnetic tunnel junctions to metal filament formation. Here we analyze an alternative strategy based on an unconventional electrical response arising from percolative charge transport and charge trapping in discrete random networks of oxide nanostructures in a dielectric matrix. After an analysis of the mechanisms which can potentially be a source of a plastic response in this class of systems, we report evidence of this behavior in a system comprising an alkali-germanosilicate amorphous matrix with incorporated Ga-oxide nanostructures. The active material – consisting in a film 70 nm thick interfaced to p-type Si and Au electrodes – gives a responsive behavior to pulsed bias which is accompanied by bias dependent electric conduction, with resistivity changes of an order of magnitude by applying 2 V, as well as a dielectric response with hysteretic features, as expected by the model. The results represent a first proof of concept of an unexplored strategy for the design of responsive systems.

Electric-Field-Induced Metal Filament Formation in Cobalt-Based CBRAM Observed by TEM

Conductive-bridging random access memory (CBRAM) using a cobalt (Co) electrode has recently featured a CMOS-compatible process, excellent data retention, and a sub-μA operating current level, which are difficult to achieve by conventional CBRAM. However, the resistive switching (RS) mechanism of Co CBRAM has not been extensively explored compared to that of the conventional CBRAM cells using Ag, Cu, or Ni as active metal electrodes. Because only implicit inferences based on electrical measurements have been made, the formation of Co filaments is not yet clearly understood. This study presents evidence of Co filament formation in Co/10 nm SiOx/TiN resistive random access memory (RRAM) using direct transmission electron microscopy (TEM) observations. Co protrusions larger than 5 nm are observed, and their exact atomic composition is investigated by spectroscopy. We explain the RS operation of Co/SiOx/TiN cells based on the Co electromigration-mediated filament development process through Co electrode deformation, protrusion, and subsequent Co conductive bridge formation. An asymmetric RS operation and negative temperature correlation of the resistance are found in low resistance state (LRS) cells, which further supports the Co-involved RS operation in Co/SiOx/TiN devices.

Metal doped polyaniline as neuromorphic circuit elements for in-materia computing

ABSTRACT Polyaniline-based atomic switches are material building blocks whose nanoscale structure and resultant neuromorphic character provide a new physical substrate for the development next-generation, nanoarchitectonic-enabled computing systems. Metal ion-doped devices consisting of a Ag/metal ion doped polyaniline/Pt sandwich structure were fabricated using an in situ wet process. The devices exhibited repeatable resistive switching between high (ON) and low (OFF) conductance states in both Ag+ and Cu2+ ion-doped devices. The threshold voltage for switching was>0.8 V and average ON/OFF conductance ratios (30 cycles for 3 samples) were 13 and 16 for Ag+ and Cu2+ devices, respectively. The ON state duration was determined by the decay to an OFF state after pulsed voltages of differing amplitude and frequency. The switching behaviour is analagous to short-term (STM) and long-term (LTM) memories of biological synapses. Memristive behaviour and evidence of quantized conductance were also observed and interpreted in terms of metal filament formation bridging the metal doped polymer layer. The successful realization of these properties within physical material systems indicate polyaniline frameworks as suitable neuromorphic substrates for in materia computing.

Understanding the Influence of Metal Oxide Layer Thickness and Defects on Resistive Switching Behavior Through Numerical Modeling

The resistive switching dynamics in metal–insulator–metal (MIM) structures may be explained through the formation and rupture of metallic filaments in the dielectric layer. Considerable research has been performed as an attempt to explain the microscopic origins of how regions that are originally insulating can change into conductors. Looking at how these conducting local regions are interconnected, the simulation of these localized switches can help understand how material properties such as thickness and defects influence the switching behavior and parameters. Here, the random circuit breaker (RCB) model is implemented and the dependence of the forming, Set, and Reset voltages on the oxide thickness and defect percentage is compared with experimental data. Only the forming voltage increases with the thickness, showing that, after the first step, the conductive filament formation and rupture occur only at a thinner thickness. It is observed that a higher percentage of initial conductive defects in the oxide layer promotes the forming step.

Multifunctional Ag–In–Zn–S/Cs3Cu2Cl5‐Based Memristors with Coexistence of Non‐Volatile Memory and Volatile Threshold Switching Behaviors for Neuroinspired Computing

AbstractLead‐free all‐inorganic Cs3Cu2Cl5 perovskites, a member of the metal‐metal halide material family, have attracted pronounced attention owing to their low toxicity, facile fabrication strategies, considerable ambient stability, and intriguing photoelectric properties. However, the application of environmentally friendly copper‐based Cs3Cu2Cl5 in memristors has been rarely reported to the authors’ knowledge. Herein, multifunctional memristors with the coexistence of non‐volatile memory (MS) and volatile threshold switching (TS) behaviors are introduced based on an innovative Ag–In–Zn–S/Cs3Cu2Cl5 heterostructure. The inserted Ag–In–Zn–S quantum dots layer may provide an effective method for guiding the formation of the dominant metallic Ag filaments, resulting in considerably stable and controllable multiple switching behaviors. Additionally, the heterostructure memristor is capable of imitating some essential biological synaptic functions, including long‐term potentiation (LTP), long‐term depression (LTD), and the short‐term memory (STM) to long‐term memory (LTM) transition. Furthermore, the famous conditioning Pavlov's dog experiment corresponding to associative learning is electronically simulated by the studied device. Moreover, utilizing the devices’ LTP and LTD properties, relatively high recognition accuracies for small and large digits datasets are achieved through a three‐layer artificial neural network, revealing the feasibility of implementing neuromorphic computation using heterostructure memristors.

Regulating WORM/Flash electrical memory behavior of metallopolymers through varying metal centers

To study the influence of metal atoms on the performance of organic polymer memory devices, four new metallopolymers P0, P1, P2 and P3 were designed and synthesized for organic resistive random-access memory (RRAM) applications. Among them, the device based on P1 shows nonvolatile binary write-once-read-many-times (WORM) memory behavior, while the devices based on P2 and P3 show vastly different nonvolatile binary Flash-type memory behavior. The charge transfer (CT) effect between the platinum (Pt) ligand and porphyrin moieties of metallopolymers is determined to be the origin of the electrical switching behavior of P1, which is proved by the electrochemical and DFT calculations. While the reason of the Flash type memory characteristic of P2 and P3 is ascribed to the formation of metal filaments because of the introduction of metal ions (Co and Ni) into the porphyrin ring. It is suggested that the metal ions can form an internal electrode and act as a bridge during the CT process, which contributes to both the CT and back CT, consequently regulating the WORM/Flash conversion. The results indicate the feasibility of modulating the memory behavior of metallopolymers by varying the central metal ions, manifesting the significance of metal complexes in the application of organic memory devices and will attract wide research interest of scientists. The as-fabricated device based on P3 was also successfully applied to memory logic gates and display function, which offers great application prospect as smart sensor in artificial intelligence (AI) network.

Remotely tuned multistate resistive switching in MoS2/NiMnIn thin film heterostructure for highly flexible ReRAM application

In this article, the Cu/MoS2/NiMnIn memory structure was fabricated over stainless steel substrate for flexible electronics applications. The bipolar resistive switching characteristics were observed in two different fabricated devices (i) without sulfur vacancies (D1) and (ii) with sulfur vacancies (D2) in MoS2 thin film. Three different resistance states such as the high resistance state (HRS), intermediate resistance state (IRS) and low resistance state (LRS) have been detected for the D2 memory device. It could be ascribed to the two filamentary models based on the movement of sulfur vacancies and the formation of Cu metallic filament with applied bias voltage. Moreover, the Ohmic and modified space charge limited conduction mechanisms clearly explain the current conduction in different resistance states of device D2. The fabricated MoS2 thin film-based memory structure exhibits stable resistive switching behavior with a high OFF/ON ratio of ∼3.6 × 103, good consistency of ∼3600 endurance cycles and excellent data retention capability up to 3000 s. Moreover, the remote tuning of device D2 was thoroughly investigated and an appreciable change in SET voltage was detected with external temperature and magnetic field. Additionally, the external magnetic field altered the switching states and enhanced the multi-bit data storage capability of the memory device. This can be ascribed to the effect of the Lorentz force on ionic movement in the presence of the external magnetic field. The mechanical flexibility of the memory structure was tested for 1000 bending cycles at various bending angles in both the tensile and compressive bending modes. Hence, the present study opens up new ways for the futuristic flexible device for high data storage and neuromorphic computing applications.

Investigation of Barrier Layer Effect on Switching Uniformity and Synaptic Plasticity of AlN Based Conductive Bridge Random Access Memory

In this work, we investigated the effect of the tungsten nitride (WNx) diffusion barrier layer on the resistive switching operation of the aluminum nitride (AlN) based conductive bridge random access memory. The WNx barrier layer limits the diffusion of Cu ions in the AlN switching layer, hence controlling the formation of metallic conductive filament in the host layer. The device operated at a very low operating voltage with a Vset of 0.6 V and a Vreset of 0.4 V. The spatial and temporal switching variability were reduced significantly by inserting a barrier layer. The worst-case coefficient of variations (σ/µ) for HRS and LRS are 33% and 18%, respectively, when barrier layer devices are deployed, compared to 167% and 33% when the barrier layer is not present. With a barrier layer, the device exhibits data retention behavior for more than 104 s at 120 °C, whereas without a barrier layer, the device fails after 103 s. The device demonstrated synaptic behavior with long-term potentiation/depression (LTP/LTD) for 30 epochs by stimulating with a train of identical optimized pulses of 1 µs duration.

Multifunctionality in ferromagnetic shape memory alloy-based resistive switching memory for flexible ReRAM application

Multifunctional flexible electronics is the ongoing demand for fabricating wearable data storage and communication devices. The magnetoelectric (ME) heterostructure consisting of piezoelectric (AlN) and ferromagnetic magnetic shape memory alloy [FSMA (Ni–Mn–In)] was fabricated over stainless steel (SS) substrate for resistive random access memory application. The Cu/AlN/FSMA/SS metal–insulator–metal based memory cell displays bipolar resistive switching (RS) behavior. The formation of Cu metallic filament at a particular SET voltage leads the memory cell in a low resistance state (LRS) from its pristine high resistance state (HRS). The LRS and HRS are explained well by Ohmic and space charge limited conduction mechanisms, respectively. The fabricated memory cell displays excellent endurance and data retention capability with a high OFF/ON ratio of ∼1.2 × 103. Furthermore, the multifunctionality of the ME heterostructure-based RAM was investigated by tuning the SET voltage with ambiance temperature and external magnetic field remotely. A significant change in the SET voltage could be ascribed to the temperature and magnetic field-induced strain transferred to the AlN piezoelectric layer from the magnetostrictive FSMA (Ni–Mn–In) bottom electrode. The residual Lorentz force explains the remotely tuned LRS and HRS in the transverse magnetic field for multi-bit data storage applications. Moreover, the RS characteristics remain stable even after 800 bending cycles as well as with bending angle (0°–180°). Hence, the present ME heterostructure integrated with flexible SS substrate can be a better choice for highly flexible, low-cost, and multifunctional futuristic RAM applications.

N, S-codoped cross-linked polymers for low switch voltage and high thermal stable nonvolatile memory

Heteroatoms doping is an effective strategy to boost polymer materials with desirable functions for advanced electronics and optoelectronics. Herein, two N or S heteroatoms were employed to construct large-area two-dimensional N, S-codoped conjugated microporous polymer (N, S-CMP) film for nonvolatile memory. The CMP films exhibited excellent thermal stability due to the cross-linked structure. The vertical diode with the sandwich structure of ITO/N, S-CMP/Al showed rewritable flash memory function, with a low switch voltage of 0.9 V, a high ON/OFF ratio of 1.3 × 10 2 and long retention time of up to 8 × 10 3 s. The flash behavior is probably attributed to the strongly coordinating S atoms with metal ions, which can lead the reproducible metal filament formation process. Impressively, even after annealing at 200 °C, the device also showed flash performance, indicating its excellent thermal stability. Our work provides a promising strategy to prepare CMPs for highly stable polymer memory. Text: A two-dimensional N, S-codoped conjugated microporous polymer was developed for memory device. The device exhibited flash memory performance with low switch voltage and high thermal stability. • A large-area, patterned 2D N, S-codoped conjugated microporous polymer (N, S-CMP) film was prepared. • The N, S-CMP based diode exhibited rewritable flash memory function with an ultralow switch voltage of 0.9 V. • The device also showed flash performance even after annealing at 200 °C, indicating its excellent thermal stability. • The heteroatoms in the CMP films play a key role for the reproducible metal filament formation and flash memory behavior.

Imaging of Electrothermal Filament Formation in a Mott Insulator

Resistive switching---the current- and voltage-induced change of electrical resistance---is at the core of memristive devices, which play an essential role in the emerging field of neuromorphic computing. This study is about resistive switching in a Mott insulator, which undergoes a thermally driven metal-to-insulator transition. Two distinct switching mechanisms are reported for such a system: electric-field-driven resistive switching and electrothermal resistive switching. The latter results from an instability caused by Joule heating. Here, we present the visualization of the reversible resistive switching in a planar $\mathrm{V}$${}_{2}$$\mathrm{O}$${}_{3}$ thin-film device using high-resolution wide-field microscopy in combination with electric transport measurements. We investigate the interaction of the electrothermal instability with the strain-induced spontaneous phase separation in the $\mathrm{V}$${}_{2}$$\mathrm{O}$${}_{3}$ thin film at the Mott transition. The photomicrographs show the formation of a narrow metallic filament with a minimum width $\ensuremath{\lesssim}500$ nm. Although the filament formation and the overall shape of the current-voltage characteristics (IVCs) are typical of an electrothermal breakdown, we also observe atypical effects such as oblique filaments, filament splitting, and hysteretic IVCs with sawtoothlike jumps at high currents in the low-resistance regime. We are able to reproduce the experimental results in a numerical model based on a two-dimensional resistor network. This model demonstrates that resistive switching in this case is indeed electrothermal and that the intrinsic heterogeneity is responsible for the atypical effects. This heterogeneity is strongly influenced by strain, thereby establishing a link between switching dynamics and structural properties.