Heba Abunahla

Memristor technology offers a promising route toward energy-efficient computing but faces challenges including resistance drift, variability, and the need for electroforming. Filamentary resistive random-access memory, one of the most studied memristive platforms, typically requires a high-voltage electroforming step to initiate conductive filaments, leading to increased power overhead and reduced endurance. Here we report HfO₂-based forming-free memristive devices (PdNeuRAM) that operate at low voltages, support multi-bit functionality, and exhibit reduced variability. Through combined electrical and materials characterization, we identify a Pd-O-Hf interfacial configuration that lowers oxygen-vacancy formation and migration barriers, creating a dense network of shallow defect states. Together with a Ti top electrode acting as an oxygen reservoir and an ultrathin (5 nm) HfO₂ layer, this interfacial engineering enables charge redistribution at room temperature and eliminates the need for electroforming. The fabricated devices provide tunable resistance states and reduce programming and read energy by 43% and 38%, respectively, in spiking neural network inference tasks. These results provide mechanistic insight into forming-free resistive switching and demonstrate the potential of Pd/HfO₂ devices for energy-efficient neuromorphic computing.

This research topic focuses on various aspects related to the deployment of Memristor (MR) devices for neuromorphic and AI applications. Specifically, it covers the design, fabrication, and characterization of novel MR stacks that can effectively mimic the synaptic plasticity observed in neuronal behaviour within the brain.The first paper by Landsmeer et al. builds on the finding that memristive properties closely mimic the potassium and sodium ion channels in Hodgkin-Huxley-based brain models. The authors show how a volatile memristor structure made of Pt/NbOx/Ti can directly implement the Hodgkin-Huxley potassium channel. The paper starts from the memristor theoretical dynamics and demonstrates how this device reproduces the required sigmoidal gating and voltage-dependent time constants in the Hodgkin-Huxley potassium channel. This was achieved by fine-tuning the voltage and time scales of the fabricated device. The presented voltage-controlled physical memristor reports a unique implementation of the Hodgkin-Huxley potassium ion channel, providing the first direct one-to-one mapping between a device and a biological ion channel.The second paper by Lim et al. explores a memristor-MOS transistor hardware system design that employs a memristor-transistor configuration as a 1T1R synapse. The paper reports the characteristics of a memristor when combined at the drain or source terminal of the transistor. This configuration enables seamless array expansion. The design was investigated, and corresponding design rules were proposed. Using a SPICE-based simulator, optimal transistor parameters were determined and subsequently validated through simulation. The neuromorphic learning capabilities of the proposed structure were demonstrated by implementing an SNN inference accelerator. The accuracy of the developed network was evaluated using a reduced MNIST dataset. The unique features of this structure include the use of DPI and LIF neurons, which eliminate the need for ADC and DAC modules commonly used in DNN research, resulting in improved noise performance.The third paper by Li et al. demonstrates how analog memristors can be used in hardware event representation for event data utilization as a reservoir encoder. The authors utilized the inherent stochastic and nonlinear characteristics of memristors to enable effective and low-cost feature extraction of temporal information from event streams. They also proposed a simplified memristor model and memristor-based reservoir circuit for processing dynamic visual information and feature extraction from event data. The superior accuracy of their approach was demonstrated experimentally using four event datasets, showcasing the unique application of memristorbased processing for event systems.The fourth paper by Kipelkin et al. shows how memristor-based devices can be used to implement ion channels. They selected a simple FitzHugh-Nagumo equation system describing neuronal excitability and introduced nonlinearities by the voltage-gated ion channels. Based on their approach, they generated basic excitability of a single spike, self-oscillation stable limit cycle mode with periodic spike trains, and bistability between a fixed point and a limit cycle mode. They found that due to charge accumulation effects in the memristive device, the electronic synapse qualitatively imitated real biological synapses with potentiation effects, increasing the amplitude of the response to a spike sequence.This Research Topic provides great insight into how memristive devices can be combined with other electronic devices to better imitate or mimic biological behaviour. It highlights the ability to transition from device-based to systembased approaches, where a network of these devices can be used to mimic neural networks, leading to increased use of memristor devices in neuromorphic and AI applications.

Introduction: In 2012, potassium and sodium ion channels in Hodgkin-Huxley-based brain models were shown to exhibit memristive behavior. This positioned memristors as strong candidates for implementing biologically accurate artificial neurons. Memristor-based brain simulations offer advantages in energy efficiency, scalability, and compactness, benefiting fields such as soft robotics, embedded systems, and neuroprosthetics.Methods: Previous approaches used current-controlled Mott memristors, which poorly matched the voltage-controlled nature of ion channels. This study employs volatile, oxide-based memristors that leverage electric-field-driven oxygen-vacancy migration to emulate voltage-dependent channel behavior. We selected candidate WOx and NbOx memristors and modeled their dynamics to verify performance as Hodgkin-Huxley potassium channels.Results: The device exhibits sigmoidal gating and voltage-dependent time constants consistent with the theoretical model. By scaling the passive circuitry around the memristors, we show that they capture the essential mechanisms of potassium ion-channels, although spike height is reduced due to strong non-linear voltage-dependence. Still, by cascading multiple compartments, typical spike propagation is retained.Discussion: This is the first demonstration of a voltage-controlled memristor replicating the Hodgkin-Huxley potassium channel, validating its potential for more efficient brain simulation hardware.

Computational‐neuroscience research is increasingly in need of larger, biophysically realistic brain models. These analog‐in‐nature models build upon the Hodgkin‐Huxley (HH) formalism and are run on digital, high‐performance computing systems making simulation very computationally expensive. In circuit form, these models are theoretically suitable for efficient analog implementation. However, the ion‐channel components –predominantly, sodium and potassium– are nonlinear, time‐varying resistors, lacking an efficient implementation. Chua et al. proved that these ion‐channel models are in fact memristors –devices with a conductance as a function of applied‐voltage history– claiming that “memristors are the right stuff for building brains”. However, the kind of actual memristor implementation that is the right one for building brains is not defined. In this article, the device class and characteristics of such memristors are defined and existing memristive implementations of HH‐like designs are then reviewed. Surprisingly, although often misclassified as such, no physical implementation currently exists that replicates the original HH equations faithfully or efficiently. Having put forward the desired memristor properties, a design guide for screening suitable memristor designs is then proposed. Screening the existing literature reveals that suitable devices likely already exist for potassium ion‐channel emulation, while none exists for sodium; this calls for further investigation of higher‐order, voltage‐controlled and volatile memristors.

Advances in materials science and memory devices work in tandem for the evolution of Artificial Intelligence systems. Energy-efficient computation is the ultimate goal of emerging memristor technology, in which the storage and computation can be done in the same memory crossbar. In this work, an analog memristor device is fabricated utilizing the unique characteristics of single-wall carbon nanotubes (SWCNTs) to act as the switching medium of the device. Via the planar structure, the memristor device exhibits analog switching ability with high state stability. The device’s conductance and capacitance can be tuned simultaneously, increasing the device's potential and broadening its applications' horizons. The multi-state storage capability and long-term memory are the key factors that make the device a promising candidate for bio-inspired computing applications. As a demonstrator, the fabricated memristor is deployed in spiking neural networks (SNN) to exploit its analog switching feature for energy-efficient classification operation. Results reveal that the computation-in-memory implementation performs Vector Matrix Multiplication with 95% inference accuracy and few femtojoules per spike energy efficiency. The memristor device presented in this work opens new insights towards utilizing the outstanding features of SWCNTs for efficient analog computation in deep learning systems.

The advances in material science along with the development of fabrication techniques have enabled the realization of thin-film-based electronics on active substrates. This has substantially enhanced and supported the deployment of electronic devices in several emerging applications with flexible functionality. In this work, we report a novel fabrication of graphene oxide (GO)-based memristor devices on an active/shrinkable substrate. The standard lithography process is used to fabricate planar Au-rGO-Au devices on a polymer substrate that has the ability to shrink at a certain temperature (i.e., 170 °C). Upon heating, the devices are shrunk to 50% of their original size. A detailed electrical characterization has been carried out to study the switching behavior of the fabricated devices before and after shrinking. The results prove that upon shrinking, the device preserves its switching ability with enhanced electrical parameters (i.e., switching voltage). Also, material characterization performed for the deposited GO on the active substrate shows improved properties of the GO film due to the enhanced arrangement of GO flakes after shrinking. The novel approach proposed in this work provides new insights into and offers the ability to scale thin-film electronics postfabrication and thus overcome some of the device scaling challenges due to manufacturing limitations. It also unfolds a new path for the realization of GO-based electronic devices with improved electrical properties, which is a crucial aspect of the development of highly flexible and lightweight green electronics.

This work reports on the fabrication of a novel planar reduced graphene oxide (rGO) memristor (MR) device. For the first time in the literature, the MR tunable resistive switching behavior is controlled by the GO reduction time at a constant temperature. The device is fabricated using standard microfabrication techniques on a flexible cyclic olefin copolymer substrate (COC). Thermal reduction of the GO layer at low temperatures (100 °C) avoids the drawbacks of chemical reduction methods such as toxicity and electrode metal damage during fabrication, while allowing for fine-tuning of the MR’s switching behavior. The device has analog switching characteristics, with a range of different resistance states. By taking advantage of the slow nature of GO thermal annealing, the switching properties of the rGO memristors can be precisely controlled by adjusting the reduction period. At short annealing times (i.e., T < 20 h), the devices switch from high to low resistance states, while at longer annealing times the switching behavior is reversed, with the device switching from low to high resistance states (LRS to HRS). Resistive switching occurs as a result of the diffusion and removal of the oxygen functional groups in the GO film caused by Joule heating induced by the electric current. Complete electrical characterization tests are presented along with wettability and X-ray diffraction (XRD) tests. This work opens a new vision for realizing rGO-based MR devices with tunable switching properties, broadening the application horizon of the device.

Random spray retinex (RSR) is an effective image enhancement algorithm owing to its effectiveness in improving the image quality. However, the computing complexity of the algorithm, the required hardware resources, and memory access hamper its deployment in many application scenarios, for instance, in IoT systems with limited hardware resources. With the rise of artificial intelligence (AI), the use of image enhancement has become essential for improving the performance of many emerging applications. In this paper, we propose the use of RSR as a preprocessing filter before the task of semantic segmentation of low-quality urban road scenes. Using the publicly available Cityscapes dataset, we compared the performance of a pre-trained deep semantic segmentation network on dark and noisy images with that of RSR preprocessed images. Our findings confirm the effectiveness of RSR in improving segmentation accuracy. In addition, to address the computational complexity and suitability of edge devices, we propose a novel and efficient implementation of RSR using resistive random access memory (RRAM) technology. This architecture provides highly parallel analog in-memory computing (IMC) capabilities. A detailed, efficient, and low-latency implementation of RSR using RRAM-CMOS technology is described. The design was verified using SPICE simulations with measured data from the fabricated RRAM and 65 nm CMOS technologies. The approach presented here represents an important step towards a low-complexity, real-time hardware-friendly architecture and the design of retinex algorithms for edge devices.

Recently, Memristor (MR) device has received much attention in many applications, including memory, computing, neuromorphic, sensing, and security. Electrical characterization and testing of MR are among the main challenges facing MR technology. Available MR characterization tools are costly and bulky, limiting the research contributions in this field to big institutions. Furthermore, there is a need for a portable testing platform that provides the means to extract MR parameters, mainly when used in sensing or security applications. This work proposes a novel power-efficient, cost-effective, lightweight, and automated Arduino-based MR characterization platform named MemChar. A detailed circuit and system implementation of MemChar are presented. MemChar functionality results are compared to industry-standard device parameter analyzer tools, i.e., Keithley 4200A, with excellent correlation. MemChar is an easy-to-implement and user-friendly platform that provides a significant milestone toward utilizing MR devices in mobile, IoT, and sensing applications.