A skin-beyond tactile sensor as interfaces between the prosthetics and biological systems

Future-generation neuromorphic computing seeks to overcome the limitations of von Neumann architectures by co-locating logic and memory functions, thereby emulating the function of neurons and synapses in the human brain. Despite remarkable demonstrations of high-fidelity neuronal emulation, the predictive design of neuromorphic circuits starting from knowledge of material transformations remains challenging. VO2 is an attractive candidate since it manifests a near-room-temperature, discontinuous, and hysteretic metal-insulator transition. The transition provides a nonlinear dynamical response to input signals, as needed to construct neuronal circuit elements. We discuss strategies for tuning the transformation characteristics of VO2 based on modification of material properties, interfacial structure, and field couplings. Dynamical modulation of transformation characteristics through in-situ material processing is discussed as a means of imbuing synaptic function. We emphasize mechanistic understanding of site-selective modification; external, epitaxial, and chemical strain; defect dynamics; and interfacial field coupling in modifying local atomistic structure, the implications therein for electronic structure, and ultimately, the tuning of transformation characteristics. We highlight opportunities for inverse design and for using design principles related to thermodynamics and kinetics of electronic transitions learned from VO2 to inform the design of new Mott materials, as well as to go beyond energy-efficient computation to manifest intelligence. This article is protected by copyright. All rights reserved.

Harnessing the Metal-Insulator Transition of VO2 in Neuromorphic Computing.

A continuous exponential rise has been observed in the storage and processing of the data that may not curtail in the foreseeable future. The required data processing speed and power consumption are restricted by the buses between the logic and memory devices that are characteristic of the von Neumann computing architecture. Bio-mimicking neuromorphic computing has garnered considerable academic and industrial interest to resolve these challenges. Additionally, devices based on emerging nonvolatile memories capable of mimicking the behaviors of synapses and neurons, which are the main elements in biological computing systems (brains), are attracting significant interest from the device community. With the discovery of ferroelectricity in fluorite-structured oxides, such as HfO2 and ZrO2, which are compatible with the state-of-the-art complementary-metal-oxide-semiconductor processes, ferroelectric devices have rapidly evolved as the main direction of these research and development activities. Fundamental science related to fluorite-structured ferroelectrics has been intensively studied over the last decade. At present, the focus is gradually moving to practical applications, including neuromorphic computing and advanced classical processing or memory units in the conventional von Neumann architecture. However, despite its rapid development, the wealth of recent progress in neuromorphic computing devices based on fluorite-structured ferroelectrics has not been reviewed and systemized. This progress report comprehensively reviews and systemizes the recent progress in artificial synaptic and spiking neuron devices for neuromorphic computing based on fluorite-structured ferroelectrics. 

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/inf2.12380

Neuromorphic devices based on fluorite‐structured ferroelectrics

We demonstrate resistive switching and memristive behavior in devices consisting of ultrathin (4.5 nm) semiconducting, epitaxial ferroelectric Hf0.93Y0.07O2 (YHO) films on La0.7Sr0.3MnO3-buffered, Nb-doped SrTiO3 single crystal substrates with Au top electrodes. Unlike the tunneling-driven current–voltage characteristics of ferroelectric tunnel junctions which utilize ultrathin insulating (fully depleted) ferroelectric films, the semiconducting nature of our YHO films, i.e., the presence of free charge carriers introduced by Y doping, results in radically different current–voltage characteristics. Current–voltage measurements indicate a polarization-modulated transition from Schottky-barrier-controlled charge transport to Ohmic conduction in the YHO devices, which results in a large on/off ratio of up to 540. Moreover, voltage pulse train measurements reveal a broad range of accessible resistance states, which indicates the memristive behavior of the devices. Our results represent an important step toward the development of future nonvolatile memory and brain-inspired neuromorphic computing applications based on ultrathin semiconducting ferroelectric films.

Schottky-to-Ohmic switching in ferroelectric memristors based on semiconducting Hf0.93Y0.07O2 thin films

Organic field‐effect transistors with parallel transmission and learning functions are of interest in the development of brain‐inspired neuromorphic computing. However, the poor performance and high power consumption are the two main issues limiting their practical applications. Herein, an ultralow‐power vertical transistor is demonstrated based on transition‐metal carbides/nitrides (MXene) and organic single crystal. The transistor exhibits a high JON of 16.6 mA cm−2 and a high JON/JOFF ratio of 9.12 × 105 under an ultralow working voltage of −1 mV. Furthermore, it can successfully simulate the functions of biological synapse under electrical modulation along with consuming only 8.7 aJ of power per spike. It also permits multilevel information decoding modes with a significant gap between the readable time of professionals and nonprofessionals, producing a high signal‐to‐noise ratio up to 114.15 dB. This work encourages the use of vertical transistors and organic single crystal in decoding information and advances the development of low‐power neuromorphic systems.

Ultralow‐Power Vertical Transistors for Multilevel Decoding Modes

Abstract Neuromorphic perception systems inspired by biology have tremendous potential in efficiently processing multi-sensory signals from the physical world, but a highly efficient hardware element capable of sensing and encoding multiple physical signals is still lacking. Here, we report a spike-based neuromorphic perception system consisting of calibratable artificial sensory neurons based on epitaxial VO 2 , where the high crystalline quality of VO 2 leads to significantly improved cycle-to-cycle uniformity. A calibration resistor is introduced to optimize device-to-device consistency, and to adapt the VO 2 neuron to different sensors with varied resistance level, a scaling resistor is further incorporated, demonstrating cross-sensory neuromorphic perception component that can encode illuminance, temperature, pressure and curvature signals into spikes. These components are utilized to monitor the curvatures of fingers, thereby achieving hand gesture classification. This study addresses the fundamental cycle-to-cycle and device-to-device variation issues of sensory neurons, therefore promoting the construction of neuromorphic perception systems for e-skin and neurorobotics. 

/pdf/a-calibratable-sensory-neuron-based-on-epitaxial-vo2-for-2idv4ddz.pdf

A calibratable sensory neuron based on epitaxial VO2 for spike-based neuromorphic multisensory system

The human receives and transmits various information from the outside world through different sensory systems. The sensory neurons integrate various sensory inputs into a synthetical perception to monitor complex environments, and this fundamentally determines the way how we perceive the world. Developing multifunctional artificial sensory elements that can integrate multisensory perception plays a vital role in future intelligent perception systems, whereas prior spiking neurons reported can only handle single‐mode physical signals. Herein, a bioinspired haptic‐temperature fusion spiking neuron based upon a serial connection of piezoresistive sensor and VO2 volatile memristor is presented. The artificial sensory neuron is capable of detecting and encoding pressure and temperature inputs based on the voltage dividing effect and the intrinsic thermal sensitivity of metal–insulator transition in VO2. Recognition of Braille characters is achieved through multiple piezoresistive sensors, taking advantage of the spatial integration capabilities of such spiking neurons. Notably, the traditionally separate haptic and temperature signals can be fused physically in the sensory neuron when synchronizing the two sensory cues, which is able to recognize multimodal haptic/temperature patterns. The artificial multisensory neuron thus provides a promising approach toward e‐skin, neurorobotics, and human–machine interaction technologies. A preprint version of the article can be found at: https://www.authorea.com/doi/full/10.22541/au.164668806.60849882.

https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/aisy.202200039

Artificial Multisensory Neurons with Fused Haptic and Temperature Perception for Multimodal In‐Sensor Computing

Hafnia‐based compounds have considerable potential for use in nanoelectronics due to their compatibility with complementary metal–oxide–semiconductor devices and robust ferroelectricity at nanoscale sizes. However, the unexpected ferroelectricity in this class of compounds often remains elusive due to the polymorphic nature of hafnia, as well as the lack of suitable methods for the characterization of the mixed/complex phases in hafnia thin films. Herein, the preparation of centimeter‐scale, crack‐free, freestanding Hf0.5Zr0.5O2 (HZO) nanomembranes that are well suited for investigating the local crystallographic phases, orientations, and grain boundaries at both the microscopic and mesoscopic scales is reported. Atomic‐level imaging of the plan‐view crystallographic patterns shows that more than 80% of the grains are the ferroelectric orthorhombic phase, and that the mean equivalent diameter of these grains is about 12.1 nm, with values ranging from 4 to 50 nm. Moreover, the ferroelectric orthorhombic phase is stable in substrate‐free HZO membranes, indicating that strain from the substrate is not responsible for maintaining the polar phase. It is also demonstrated that HZO capacitors prepared on flexible substrates are highly uniform, stable, and robust. These freestanding membranes provide a viable platform for the exploration of HZO polymorphic films with complex structures and pave the way to flexible nanoelectronics.

Large‐Scale Hf0.5Zr0.5O2 Membranes with Robust Ferroelectricity

With their unique advantages in portability, shape adaptability, and human friendly surfaces, flexible electronics pave the way for the implementation of wearable electronic textiles and human–machine interfaces. Although organic materials are promising for flexible devices because of the low‐cost manufacturing and inherent flexibility, they meet challenges in harsh environments such as ultraviolet (UV) irradiation, which limits their applicability in UV sensors. Here, a flexible UV neuromorphic sensor is presented using inorganic vanadium dioxide (VO2) films grown on mica substrates. The flexible device shows UV photoinduced nonvolatile phase transition, and can be reversibly modulated using electrolyte gating. The optical responses remain almost unchanged after 10 000 bending cycles or at small bending radius, exhibiting high tolerance to the bending deformation. Besides, the variations in image recognition accuracy under different bending conditions keep within 1.6%, indicating that the device can be adapted to various deformation conditions. By constructing near‐/in‐sensor computing architectures using the flexible VO2 neuromorphic sensors with photoinduced nonvolatile phase transition, both static image processing and motion detection are realized without redundant and massive information transfer. This result lays the foundation for the development of flexible UV neuromorphic sensors.

Flexible VO2 Films for In‐Sensor Computing with Ultraviolet Light

A terahertz hybrid metamaterial incorporated with active media VO2 holds great promise for the realization of a new generation of reconfigurable and multifunctional devices. However, for the electrical control, many efforts on reducing high working threshold are usually based on the utilization of patterned VO2 patches or additional insulation layers, which will increase the complexity of the fabrication procedure. Here, we have proposed an effective strategy only by combining the surface microstructure and the unpatterned VO2 film to realize the tunability of working current and uncover its highly dependent correlation with the structural resonance responses. It is shown the fully modulated current in our hybrid metastructures can be reduced with the prominently separated hysteresis loops. Further developed binary encoders can perform not only the information transformation of the fixed code symbols but also the arbitrary encoding with the programmable current pulse. Additionally, the dynamic color display can be accomplished to illustrate the intriguing function of the information encryption and multi-image reappearance with the current as the decryption key. Our work provides an approach to reduce the operating current and paves a pathway for the development of photonic memory information processors.

Ge Li

Papers

A calibratable sensory neuron based on epitaxial VO2 for spike-based neuromorphic multisensory system

Artificial Multisensory Neurons with Fused Haptic and Temperature Perception for Multimodal In‐Sensor Computing

Large‐Scale Hf0.5Zr0.5O2 Membranes with Robust Ferroelectricity

Flexible VO2 Films for In‐Sensor Computing with Ultraviolet Light

Resonance dependence of electrically reconfigurable VO2-based THz metadevice for memory information processing