Exponential Escape Rate of Filamentary Incubation in Mott Spiking Neurons

Abstract

Mott materials such as vanadium oxides, when subject to a strong applied voltage, present an inhomogeneous insulator-to-metal transition with formation of metallic filaments within the insulating bulk. This property is enabling the development of compact and power-efficient neuromorphic devices known as Mott neurons. However, the nature of the transition has not been fully understood yet, as it may be attributed to different effects, including Joule self-heating and hot-carrier injection. Moreover, the experimental determination of the threshold voltage needed to induce the transition has proven to be challenging, as the transition becomes increasingly unpredictable when the threshold is approached. The physical understanding of these issues would not only deepen our understanding of Mott insulators, but would also be an important step toward the realization of neuromorphic devices based on such materials. In this work we use numerical simulations based on the Mott resistor network model to study the nature of the filament incubation and formation process. We show that both electronic and thermal effects, in the form of current density focusing and Joule self-heating, respectively, contribute to the filamentary incubation and growth. Remarkably, we find that the percolation of the metallic filaments near the threshold is intrinsically stochastic, qualitatively similar to the familiar Arrhenius activated behavior and to the stochastic firing of biological neurons. More precisely, we characterize the filament percolation as a Poisson point process, which has the same probability distribution as mathematical models of neuronal firing with an exponential escape rate. Finally, we support the numerical simulation results by performing experiments in VO 2 that are in agreement with the exponential escape rate behavior. Thus, we establish a functionality of Mott insulators that opens a path toward implementing neuromorphic hardware with quantum materials.

Quantum materials for energy-efficient neuromorphic computing

“…It should be kept in mind that the interaction should be a priori scalable to multiple neurons, since building networks is the ultimate goal. Since Mott neurons rely on an electro-thermal incubation process 157 we envision different types of neuron-neuron interactions based on both electric and thermal coupling.…”

Interplay of Thermal and Electronic Effects in the Mott Transition of Nanosized VO2 Phase Change Memory Devices

“…After releasing the voltage the device relaxes back to the original high resistance state (HRS). These volatile resistive switches exhibit excellent characteristics with stable and reproducible operation up to 10 9 cycles, and simple circuits including one or two VO 2 switches, capacitors, resistors, and batteries can realize artificial neurons that can mimic most of the known biological neuronal dynamic patterns. ,,,, Alternatively, the underlying switching mechanism exhibits remarkable complexity both in the temporal and spatial domains. The former is reflected in the subthreshold switching characteristics for rapidly repeated programming pulses, whereas spatially an inhomogeneous switching region was discovered, where the formation of conductive regions is more pronounced at the device edges. , The description of these phenomena requires complex modeling tools, like a percolation network model. − …”

“…Besides the structural and electronic properties, the optical features also change remarkably between the two phases. Due to the near room temperature phase transition, VO 2 is an attractive material for various applications including electromechanical devices, sensors and transistors. − Recently, the advantageous properties of VO 2 were also exploited in the field of neuromorphic computing, taking advantage of the volatile resistive switching characteristics of VO 2 devices. − …”

Dynamics of an Electrically Driven Phase Transition in Ca2RuO4 Thin Films: Nonequilibrium High‐Speed Resistive Switching in the Absence of an Abrupt Thermal Transition

“…

important research topic because most of the critical properties of Mott neuromorphic devices, such as the firing rate of artificial neurons [3][4][5][6][7] and the potentiation/ depression rate and temporal plasticity of artificial synapses, [5,8] directly depend on the formation/retention time of the metallic phases, as well as because of the fundamental interest in the physics of the transition pathway. [9][10][11][12][13] In previous studies, dynamics analysis and performance characterization of the metal-insulator transition have been conducted by fabricating two-terminal electrodes on a material and inducing the transition via electric field application or current injection.

…”

“…[9][10][11][12][13] In previous studies, dynamics analysis and performance characterization of the metal-insulator transition have been conducted by fabricating two-terminal electrodes on a material and inducing the transition via electric field application or current injection. Using the resistance changes under fields as a probe of the formation of the metallic phase, the dynamics and speed of the transition have been investigated for a variety of strongly correlated materials, including vanadium oxides (VO 2 , V 2 O 3 , and V 3 O 5 ), [5][6][7][9][10][11][13][14][15] metal chalcogenides (AM 4 X 8 (A = Ga and Ge, M = V, Nb, and Ta, X = S and Se), [3,4,8,[16][17][18][19] Ni(S,Se) 2 , [19] and 1T-TaS 2 [20,21] ), and rare-earth perovskite manganites [22] and nickelates, [12] and high-speed resistive switching with a switching time down to <100 ns has been demonstrated in some materials. [6,11,12,17,20] Recent investigations have indicated that the electrically driven resistive switching in the strongly correlated materials is generally produced by the increase in the internal temperature due to Joule heating, and temperature-driven metal-insulator transitions that are the same as those induced by temperature In Mott-type resistive switching phenomena, which are based on the metalinsulator transition in strongly correlated materials, the presence of an abrupt temperature-driven transition in the material is considered essential for achieving high-speed and large-resistance-ratio switching.…”