Programmable Retention Characteristics in MoS2-Based Atomristors for Neuromorphic and Reservoir Computing Systems.

In this study, we investigate the coexistence of short- and long-term memory effects owing to the programmable retention characteristics of a two-dimensional Au/MoS2/Au atomristor device and determine the impact of these effects on synaptic properties. This device is constructed using bilayer MoS2 in a crossbar structure. The presence of both short- and long-term memory characteristics is proposed by using a filament model within the bilayer transition-metal dichalcogenide. Short- and long-term properties are validated based on programmable multilevel retention tests. Moreover, we confirm various synaptic characteristics of the device, demonstrating its potential use as a synaptic device in a neuromorphic system. Excitatory postsynaptic current, paired-pulse facilitation, spike-rate-dependent plasticity, and spike-number-dependent plasticity synaptic applications are implemented by operating the device at a low-conductance level. Furthermore, long-term potentiation and depression exhibit symmetrical properties at high-conductance levels. Synaptic learning and forgetting characteristics are emulated using programmable retention properties and composite synaptic plasticity. The learning process of artificial neural networks is used to achieve high pattern recognition accuracy, thereby demonstrating the suitability of the use of the device in a neuromorphic system. Finally, the device is used as a physical reservoir with time-dependent inputs to realize reservoir computing by using short-term memory properties. Our study reveals that the proposed device can be applied in artificial intelligence-based computing applications by utilizing its programmable retention properties.

Volatile and Nonvolatile Resistive Switching Coexistence in Conductive Point Hexagonal Boron Nitride Monolayer.

Recently, we demonstrated the nonvolatile resistive switching effects of metal-insulator-metal (MIM) atomristor structures based on two-dimensional (2D) monolayers. However, there are many remaining combinations between 2D monolayers and metal electrodes; hence, there is a need to further explore 2D resistance switching devices from material selections to future perspectives. This study investigated the volatile and nonvolatile switching coexistence of monolayer hexagonal boron nitride (h-BN) atomristors using top and bottom silver (Ag) metal electrodes. Utilizing an h-BN monolayer and Ag electrodes, we found that the transition between volatile and nonvolatile switching is attributed to the thickness/stiffness of chain-like conductive bridges between h-BN and Ag surfaces based on the current compliance and atomristor area. Computations indicate a "weak" bridge is responsible for volatile switching, while a "strong" bridge is formed for nonvolatile switching. The current compliance determines the number of Ag atoms that undergo dissociation at the electrode, while the atomristor area determines the degree of electric field localization that forms more stable conductive bridges. The findings of this study suggest that the h-BN atomristor using Ag electrodes shows promise as a potential solution to integrate both volatile neurons and nonvolatile synapses in a single neuromorphic crossbar array structure through electrical and dimensional designs.

Comparative Studies of Atomically Thin Proton Conductive Films to Reduce Crossover in Hydrogen Fuel Cells.

Hydrogen fuel cells based on proton exchange membrane fuel cell (PEMFC) technology are promising as a source of clean energy to power a decarbonized future. However, PEMFCs are limited by a number of major inefficiencies; one of the most significant is hydrogen crossover. In this work, we comprehensively study the effects of two-dimensional (2D) materials applied to the anode side of the membrane as H2 barrier coatings on Nafion to reduce crossover effects on hydrogen fuel cells, while studying adverse effects on conductivity and catalyst performance in the beginning of life testing. The barrier layers studied include graphene, hexagonal boron nitride (hBN), amorphous boron nitride (aBN), and varying thicknesses of molybdenum disulfide (MoS2), all chosen due to their expected stability in a fuel cell environment. Crossover mitigation in the materials studied ranges from 4.4% (1 nm MoS2) to 46.1% (graphene) as compared to Nafion 211. Effects on proton conductivity are also studied, suggesting high areal proton transport in materials previously thought to be effectively nonconductive, such as 2 nm MoS2 and amorphous boron nitride under the conditions studied. The results indicate that a number of 2D materials are able to improve crossover effects, with those coated with 8 nm MoS2 and 1 L graphene able to achieve greater crossover reduction while minimizing conductivity penalty.

Comparative Study between Sulfurized MoS2 from Molybdenum and Molybdenum Trioxide Precursors for Thin-Film Device Applications.

Two-dimensional (2D) materials have been studied as an emerging class of nanomaterials owing to their attractive properties in nearly every field of science and technology. Molybdenum disulfide (MoS2) is one of the more promising candidates of these atomically thin 2D materials for its technological potential. The facile synthesis of MoS2 remains a matter of broad interest. In this study, MoS2 was synthesized by chemical vapor deposition sulfurization at various temperatures (550 °C, 650 °C, and 750 °C) of either precursor molybdenum metal (Mo) or molybdenum trioxide (MoO3) deposited on silicon/silicon dioxide (Si/SiO2) via e-beam evaporation. Monolayer, bilayer, and few layers sulfurized samples have been grown and verified by Raman, photoluminescence spectroscopy, XRD, XPS, and AFM. MoO3 sulfurization provided monolayer growth in comparison to Mo sulfurization under the same conditions and precursor thicknesses. Optical microscopy showed the homogeneous nature of grown samples. A main finding of this work is that MoO3 sulfurization produced higher quality MoS2 as compared to those grown by an Mo precursor. Device characteristics based on monolayer MoO3 sulfurized MoS2-x include nonvolatile resistive switching with Ion/Ioff ≈ 104 at a relatively low operating bias of ±1 V. In addition, field-effect transistor characteristics revealed p-type material growth with a carrier mobility ∼ 41 cm2 V-1 s-1, which is in contrast to typically observed n-type characteristics.

The influence of aqueous solvent on the electronic structure and non-adiabatic dynamics of indole explored by liquid-jet photoelectron spectroscopy.

Understanding how the electronic structure of an aqueous solute is intricately bound up with the arrangement of a host liquid provides insight into how non-adiabatic photochemistry takes place in the condensed phase. For example, the presence of water provides additional solute-solvent interactions compared to non-polar solvents: changing the stability of ionized products and modifying the energies of low-lying excited valence states, as well as moving the point of intersection between potential surfaces. Thus, the locations and topography of conical intersections between these surfaces also change. The overall impact of the aqueous environment can be to modify the intricate photochemical and non-radiative pathways taking place after photoexcitation. Time-resolved photoelectron spectroscopy (TRPES) in a liquid micro-jet is implemented here to investigate the influence of water on the electronic structure and dynamics of indole, the chromophore of the amino acid tryptophan. TRPES is used to establish ultrafast relaxation pathways that vary as a function of excitation wavelength. In our experiment, aqueous indole was excited with femtosecond pulses centered at 292 nm and 266 nm. The vertical excitation energy of aqueous indole is extracted and found to be lowered by 0.5 eV in water relative to the gas phase. In the TRPES study, the spectral signature of 1La and evidence of solvated electron formation on an ultrafast timescale are observed. Our data also points to a possible contribution of the dissociative πσ* state, which can be accessed by a conical intersection (CI) with the 1La state.