Recent Advances in Synaptic Nonvolatile Memory Devices and Compensating Architectural and Algorithmic Methods Toward Fully Integrated Neuromorphic Chips

Filamentary and interface switching of CMOS-compatible Ta2O5 memristor for non-volatile memory and synaptic devices

Neuromorphic systems (hardware neural networks) derive inspiration from biological neural systems and are expected to be a computing breakthrough beyond conventional von Neumann architecture. Interestingly, in neuromorphic systems, the processing and storing of information can be performed simultaneously by modulating the connection strength of a synaptic device (i.e., synaptic weight). Previously investigated synaptic devices can emulate the functionality of biological synapses successfully by utilizing various nano-electronic phenomena; however, the impact of intrinsic synaptic device variability on the system performance has not yet been studied. Here, we perform a device-to-system level simulation of different synaptic device variation parameters in a designed neuromorphic system that has the potential for unsupervised learning and pattern recognition. The effects of variations in parameters such as the weight modulation nonlinearity (NL), the minimum-maximum weight (Gmin and Gmax), and the weight update margin (ΔG) on the pattern recognition accuracy are analyzed quantitatively. These simulation results can provide guidelines for the continued design and optimization of a synaptic device for realizing a functional large-scale neuromorphic computing system.

Neuromorphic computing systems are based on new architecture inspired by biological learning behavior and biological structure to overcome the shortcomings of the von Neumann system, such as high energy consumption and generation of excessive redundant data. However, common synaptic‐device designs are still need to carry an external power source, which is detrimental to both the miniaturization of smart devices and the further reduction of energy consumption. Therefore, self‐powered optoelectronic synapses hold great promise for achieving energy‐efficient artificial intelligence applications. Given the advantages of heterogeneous integration and ferroelectric performance, the 2D ferroelectric In2Se3 has become a typical candidate material for biological synaptic devices and neuromorphic computing. Although, various structures of ferroelectric synaptic devices are developed, most of which are controlled by optical or electrical pulses, hardly any reports discuss the operation in self‐powered mode. In this work, an In2Se3/MoS2 ferroelectric lateral heterojunction‐type optoelectronic synaptic device is fabricated to mimic biological synaptic functions in voltage mode or self‐driven mode. In addition, two simplified image pre‐processing methods based on device arrays are shown to improve the recognition accuracy of a back‐end neural network. The proposed approach suggests a new route for designing and applying high‐performance synaptic devices.

9 - Novel photoelectroactive memories and neuromorphic devices based on nanomaterials

To overcome the coming big-data era, significant efforts are needed to implement neuromorphic computing that mimics the functional and structural features of the human brain using electronic devices. Repeated updates mimicking biological synaptic plasticity degrade the endurance of synaptic devices during in situ training. This study demonstrates a two-dimensional (2D) ferroelectric α-In2Se3-based synaptic device with enhanced durability via self-curing performed from current annealing (CA) in synaptic fatigue occurring during repetitive learning. The conduction, degradation, and self-curing mechanisms of the 2D ferroelectric-based synaptic device are quantitatively elucidated by low-frequency noise (LFN) spectroscopy. The classification accuracy of the Canadian Institute for Advanced Research (CIFAR)-10 dataset with self-cured conductance is superior to that of the device with synaptic fatigue and recovers to the initial accuracy level. The simulation results of removing defect cells through self-curing in the 2D ferroelectric synaptic array can help reduce energy consumption in the long term. The experimental results emphasize adopting 2D ferroelectric materials for future neuromorphic computing.

As Moore's law approaches its limits, it is gradually harder to achieve the more and more complex tasks with computers with conventional von Neumann architecture. The hardware circuits for neuromorphic computing, with low power and high computing density, have attracted a lot of attention. Based on these requirements, how to choose the appropriate circuit components for neuromorphic computing becomes very important. Ambipolar synaptic devices with tunable and versatile behaviours, as a new type of circuit component, can mimic both excitatory and inhibitory behaviours, which has great potential in neuromorphic computing. In this chapter, we will introduce the definitions of ambipolar materials and synaptic devices. Then, the characteristics and applications of the synaptic devices will be discussed. The differences between conventional and ambipolar synaptic devices will also be compared. Finally, we will introduce the possibility of applying ambipolar synaptic devices in array to achieve more complex work.

While biological vision systems excel at in-memory processing with low power consumption, traditional silicon-based vision chips struggle with high energy demands. This gap motivates the exploration of alternative materials for artificial intelligence applications. This paper presents a VOx-based optoelectronic synaptic memristive device. The proposed artificial synaptic device ITO/VO x /Pt mimics biological functions such as potentiation (P), depression (D), long-term memory, short-term memory (STM), and paired-pulse facilitation (PPF). The PPF index, standing at 105%, suggests a favorable pattern in STM function. The device served as synapses within a spiking neural network showing an achievable pattern classification accuracy of 88.68%, highlighting the potential of the VO x synaptic device for pattern classification tasks. The suggested VO x -based synaptic devices could represent an efficient pattern recognition and visual perception application.

Synaptic devices based neuromorphic computing applications in artificial intelligence

Neuro-inspired deep learning algorithms have shown promising futures in artificial intelligence. Despite the remarkable progress in software-based neural networks, the traditional von-Neumann hardware architecture has suffered from limited energy efficiency while facing unprecedented large amounts of data. To meet the performance requirements of neuro-inspired computing, large-scale vector-matrix multiplication is preferred to be performed in situ, namely compute-in-memory. Non-volatile memory devices with different materials have been proposed for weight storage as synaptic devices. Among them, HfO2-based ferroelectric devices have attracted great attention because of their low energy consumption, good complementary-metal-oxide-semiconductor (CMOS) compatibility and multi-bit per cell potential. In this review, recent trends and prospects of the ferroelectric synaptic devices are surveyed. First, we present the three-terminal synaptic devices based on the ferroelectric field effect transistor (FeFET), and discuss the switching physics of the intermediate states, the back-end-of-line integration and the 3D NAND architecture design. Then, we introduce a hybrid precision synapse concept that leverages the volatile charges on the gate capacitor of the FeFET and the non-volatile polarization on the gate dielectric of the FeFET. Lastly, we review two-terminal synaptic devices using the ferroelectric tunnel junction (FTJ) and ferroelectric capacitor (FeCAP). The design margins of the crossbar array with FTJ and FeCAP analyzed.

Imaging sensors with inbuilt processing capability are expected to form the backbone of low‐latency and highly energy efficient artificial vision systems. A range of emerging atomically thin materials provide opportunities to exploit their electrical and optical properties for human vision and brain inspired functions. This work reports atomically thin nanosheets of β‐In2S3 which exhibit inherent persistent photoconductivity (PPC) under ultraviolet and visible wavelengths. This PPC effect enables β‐In2S3‐based optoelectronic devices to optically mimic the dynamics of biological synapses. Based on the material characterizations, the PPC effect is attributed to the intrinsic defects in the synthesized β‐In2S3 nanosheet. Furthermore, the feasibility of adopting these atomically thin synaptic devices for optoelectronic neuromorphic hardware is demonstrated by implementing a convolutional neural network for image classification. As such, the demonstrated atomically thin nanosheets and optoelectronic synaptic devices provide a platform for scaling up complex vision‐sensory neural networks, which can find many promising applications for multispectral imaging and neuromorphic computation.

The development of synaptic devices featuring metallic nanostructures with brain-analog hierarchical architecture, capable of mimicking cognitive functionalities, has emerged as a focal point in neuromorphic computing. However, existing challenges, such as inconsistent and unpredictable switching, high voltage requirements, unguided filament formation, and detailed fabrication processes, have impeded technological progress in the domain. The present study addresses some of these challenges by leveraging periodic nanostructures of Ag fabricated via plasma-assisted nanosphere lithography (NSL). The triangular nanostructures with a preferred orientation offer enhanced localized electric fields, facilitating low voltage electromigration at the sharp edges to guide predictive filament formation. A thorough investigation into gap control between the nanostructures through oxygen plasma treatment enables the attainment of an optimized low switching voltage of 0.86 V and retention at an ultra-low current compliance of 100 nA. The optimized device consumes low power, typically in the fJ range, akin to biological neurons. Furthermore, the device showcases intriguing synaptic characteristics, including controlled transition from short- to long-term potentiation, associative learning, etc., projecting its potential in perceptive learning, memory formation, and brain-inspired computing. COMSOL Multiphysics simulation, supported by ex situ electron microscopic imaging, confirms the controlled and predictable filament formation facilitated by electric field enhancement across the strategic nanostructures. Thus, the work highlights the potential of NSL-based cost-effective fabrication techniques for realizing efficient and biomimetic synaptic devices for neuromorphic computing applications.

Zero-power optoelectronic synaptic devices

Neuromorphic computing, an innovative technology inspired by the human brain, has attracted increasing attention as a promising technology for the development of artificial intelligence systems. This study proposes synaptic transistors with a Li1-xAlxTi2-x(PO4)3 (LATP) layer to analyze the conductance modulation linearity, which is essential for weight mapping and updating during on-chip learning processes. The high ionic conductivity of the LATP electrolyte provides a large hysteresis window and enables linear weight update in synaptic devices. The results demonstrate that optimizing the LATP layer thickness improves the conductance modulation and linearity of synaptic transistors during potentiation and degradation. A 20 nm-thick LATP layer results in the most nonlinear depression (αd = -6.59), whereas a 100 nm-thick LATP layer results in the smallest nonlinearity (αd = -2.22). Additionally, a device with the optimal 100 nm-thick LATP layer exhibits the highest average recognition accuracy of 94.8% and the smallest fluctuation, indicating that the linearity characteristics of a device play a crucial role in weight update during learning and can significantly affect the recognition accuracy.

In the era of “big data,” the cognitive system of the human brain is being mimicked through hardware implementation of highly accurate neuromorphic computing by progressive weight update in synaptic electronics. Low‐energy synaptic operation requires both low reading current and short operation time to be applicable to large‐scale neuromorphic computing systems. In this study, an energy‐efficient synaptic device is implemented comprising a Ni/Pb(Zr0.52Ti0.48)O3 (PZT)/0.5 wt.% Nb‐doped SrTiO3 (Nb:STO) heterojunction with a low reading current of 10 nA and short operation time of 20–100 ns. Ultralow femtojoule operation below 9 fJ at a synaptic event, which is comparable to the energy required for synaptic events in the human brain (10 fJ), is achieved by adjusting the Schottky barrier between the top electrode and ferroelectric film. Moreover, progressive domain switching in ferroelectric PZT successfully induces both low nonlinearity/asymmetry and good stability of the weight update. The synaptic device developed here can facilitate the development of large‐scale neuromorphic arrays for artificial neural networks with low energy consumption and high accuracy.

Nanoscale channel organic ferroelectric synaptic transistor array for high recognition accuracy neuromorphic computing