Vertical MoS₂ Double-Layer Memristor with Electrochemical Metallization as an Atomic-Scale Synapse with Switching Thresholds Approaching 100 mV

Abstract

Atomically thin two-dimensional (2D) materialssuch as transition metal dichalcogenide (TMD) monolayers and hexagonal boron nitride (hBN)and their van der Waals layered preparations have been actively researched to build electronic devices such as field-effect transistors, junction diodes, tunneling devices, and, more recently, memristors. Twodimensional material memristors built in lateral form, with horizontal placement of electrodes and the 2D material layers, have provided an intriguing window into the motions of ions along the atomically thin layers. On the other hand, 2D material memristors built in vertical form with top and bottom electrodes sandwiching 2D material layers may provide opportunities to explore the extreme of the memristive performance with the atomic-scale interelectrode distance. In particular, they may help push the switching voltages to a lower limit, which is an important pursuit in memristor research in general, given their roles in neuromorphic computing. In fact, recently Akinwande et al. performed a pioneering work to demonstrate a vertical memristor that sandwiches a single MoS 2 monolayer between two inert Au electrodes, but it could neither attain switching voltages below 1 V nor control the switching polarity, obtaining both unipolar and bipolar switching devices. Here, we report a vertical memristor that sandwiches two MoS 2 monolayers between an active Cu top electrode and an inert Au bottom electrode. Cu ions diffuse through the MoS 2 double layers to form atomic-scale filaments. The atomic-scale thickness, combined with the electrochemical metallization, lowers switching voltages down to 0.1−0.2 V, on par with the state of the art. Furthermore, our memristor achieves consistent bipolar and analogue switching, and thus exhibits the synapse-like learning behavior such as the spike-timing dependent plasticity (STDP), the very first STDP demonstration among all 2D-material-based vertical memristors. The demonstrated STDP with low switching voltages is promising not only for low-power neuromorphic computing, but also from the point of view that the voltage range approaches the biological action potentials, opening up a possibility for direct interfacing with mammalian neuronal networks.

Ultralow Power Wearable Heterosynapse with Photoelectric Synergistic Modulation

“…Long‐term retention (10 3 s) was measured after ten consecutive negative pulses (pulse width:100 ns, amplitude: −4 V), as shown in Figure 3j. The energy consumption of our flexible MoS 2 ‐based device was compared with other synapses, including biological and electronic synapses, [ 5,6,10,44,46–56 ] and its advantages are demonstrated in Figure 3k. The energy consumption of our synaptic device was orders of magnitude lower than that of biological systems (10 fJ per synaptic event).…”

Chemical vapor deposition synthesis of two-dimensional freestanding transition metal oxychloride for electronic applications

“…As shown in Figure 4(f), the resistance of VOCl based memristive device gradually shows an increase or a reduction with consecutive negative (−1.15 V, 500 µs) or positive (1.6 V, 50 µs) voltage pulses, respectively. This suggests that our VOCl based memristors have potential applications in neuromorphic computing [48,[52][53][54][55][56][57].…”

Two-Dimensional Near-Atom-Thickness Materials for Emerging Neuromorphic Devices and Applications

“…In fact, various working principles and structure-property relations exist despite the structural similarity of 2D TMDs. For instance, 2D TMD-based vertical memristive devices are able to tune multiple resistances by forming and rupturing conductive filaments through the migration of ions stemming from their electrodes ( Dev et al., 2020 ; Xu et al., 2019 ). The spatial pathways of these ions can be further tuned by the intrinsically existing in-plane grain boundaries of 2D TMD layers ( Sangwan et al., 2015 ), as well as by the externally introduced intercalations in between each 2D layer ( Zhang et al., 2019 ; Zhu et al., 2019 ).…”

“…Their desirable properties, including atomic thickness, dangling-bond-free surfaces, mechanical strength, high integration density, tunable electrical transport, and optical properties, as well as low energy consumption, make them ideal candidates for applications in a wide range of electronic devices ( Gupta et al., 2015 ; Mas-Ballesté et al., 2011 ; Xia et al., 2017 ). More recently, the applications of 2D materials have been extensively studied for energy-efficient and high-performing artificial synapses ( Arnold et al., 2017 ; Chen et al., 2019c ; Dev et al., 2020 ; Hu et al., 2019 ; Jiang et al., 2017 ; Kalita et al., 2019 ; Kim et al., 2019c ; Krishnaprasad et al., 2019 ; Kumar et al., 2019 ; Li et al., 2018 ; Liu et al., 2019 ; Mao et al., 2019 ; Paul et al., 2019 ; Pradhan et al., 2020 ; Xie et al., 2018a , 2018b ; Xu et al., 2019 ; Yan et al., 2019a ; Yi et al., 2018 ; Zhu et al., 2019 ). Furthermore, owing to their dangling-bonds-free surface and atomically thin nature, a variety of 2D materials-based heterostructures have been developed in spite of their lattice mismatch ( Novoselov et al., 2016 ).…”

“…Furthermore, owing to their dangling-bonds-free surface and atomically thin nature, a variety of 2D materials-based heterostructures have been developed in spite of their lattice mismatch ( Novoselov et al., 2016 ). 2D layered materials can be utilized as active materials in diode-type neuromorphic device architecture with two-electrode configuration, allowing for the storage and updating of synaptic weights via various mechanisms, e.g., the formation of filaments ( Lee et al., 2018 ; Shi et al., 2018 ; Xu et al., 2019 ), the transformation of phases ( Zhu et al., 2019 ), or redistribution of atomic vacancies ( Zhu et al., 2019 ). The unrivaled thinness of 2D materials also enables rapid electrical switching in memristor devices circumventing short channel effects, leading to improved energy efficiency ( Stanford et al., 2018 ).…”