Search

Exploitation of the oxidation behaviour in an environmentally sensitive semiconductor is significant to modulate its electronic properties and develop unique applications. Here, we demonstrate a native oxidation-inspired InSe field-effect transistor as an artificial synapse in device level that benefits from the boosted charge trapping under ambient conditions. A thin InOx layer is confirmed under the InSe channel, which can serve as an effective charge trapping layer for information storage. The dynamic characteristic measurement is further performed to reveal the corresponding uniform charge trapping and releasing process, which coincides with its surface-effect-governed carrier fluctuations. As a result, the oxide-decorated InSe device exhibits nonvolatile memory characteristics with flexible programming/erasing operations. Furthermore, an InSe-based artificial synapse is implemented to emulate the essential synaptic functions. The pattern recognition capability of the designed artificial neural network is believed to provide an excellent paradigm for ultra-sensitive van der Waals materials to develop electric-modulated neuromorphic computation architectures., Developing efficient memory and artificial synaptic systems based on environmentally sensitive van der Waals materials remains a challenge. Here, the authors present a native oxidation-inspired InSe field-effect transistor that benefits from a boosted charge trapping behavior under ambient conditions.

The oxygen (O2)-dependent resistance change of multilayered molybdenum ditelluride (MoTe2) channels was characterized. A variation of the channel resistance could reproducibly determine relative O2...

Recently, researchers have focused on optoelectronics based on two-dimensional van der Waals materials to realize multifunctional memory and neuron applications. Layered indium selenide (InSe) semiconductors satisfy various requirements as photosensitive channel materials, and enable the realization of intriguing optoelectronic applications. Herein, we demonstrate InSe photonic devices with different trends of output currents rooted in the carrier capture/release events under various gate voltages. Furthermore, we reported an increasing/flattening/decreasing synaptic weight change index (∆Wn) via a modulated gate electric field, which we use to imitate medicine-acting metaplasticity with effective/stable/ineffective features analogous to the synaptic weight change in the nervous system of the human brain. Finally, we take advantage of the low-frequency noise (LFN) measurements and the energy-band explanation to verify the rationality of carrier capture-assisted optoelectronics applied to neural simulation at the device level. Utilizing optoelectronics to simulate essential biomedical neurobehaviors, we experimentally demonstrate the feasibility and meaningfulness of combining electronic engineering with biomedical neurology.

In the present study, we aim to help improve the design of van der Waals stacking, i.e., vertical 2D electronics, by probing charge transport differences in both parallel and vertical conducting channels of layered molybdenum disulfide (MoS2), with thin graphite acting as source and drain electrodes. To avoid systematic errors and variable contact contributions to the MoS2 channel, parallel and vertical electronics are all fabricated and measured on the same conducting material. Large differences in the on/off current ratio, mobility, and charge fluctuations, between parallel and vertical electronics are evident in electrical performance as well as in charge transport mechanisms. Further insights are drawn from a well-constrained analysis of both temperature-dependent current-voltage characteristics and low-frequency (LF) current fluctuations. This work offers significant insight into the fundamental understanding of charge transport and the development of future layered-materials-based integration technology.

Modern technology allows us to mimic biological functions with artificial devices. The human brain, including numerous neural cells that connect via synapses, enables us to handle complex tasks with ultra-low power consumption, which is one of the most important biological components that eager to emulate. Here, we propose and build a simple indium selenide (InSe)-based photonic synaptic device with unique gate tunable behaviours such as time-varying output current, auto-depression rate, and paired-pulse facilitation (PPF). A new inverse PPF behaviour is observed, characterized by the positive correlations to the time interval, which is opposite of those negative ones in previous studies. To unveil the origin of this new finding, both the substrate-dependent persistent photocurrent examination and low-frequency noise (LFN) measurements are performed to investigate the electric-controlled charge trapping/detrapping processes between the InSe and SiO2 interface. Furthermore, we systematically demonstrate such the specific evolution of the flexible plasticity within a low-operation voltage and a wide range of visible spectra. Interestingly, the inverse PPF can be employed to emulate more detailed characteristics of a biological brain such as the age-related changes of synaptic plasticity in real human brains. Thus, we believed that our findings provide a proof-of-concept for systematically mimicking the brain plasticity and advancing the development of energy-efficient artificial brains.

Multifunctional devices are expected to allow development of low-cost, highly integrated, and energy-efficient electronics for the internet of things (IoT) era. Here, we demonstrate an all-two-dimensional ReSe2/h-BN/graphene heterostructure (ReSe2 HS) consisting of a vertically coupled ReSe2 field-effect transistor and charge storage component, which possesses multifunctional characteristics for use in electronics and optoelectronics. As an electrically controlled non-volatile memory (NVM), the ReSe2 HS delivers high-performance multilevel data storage with a readout on/off current ratio exceeding 105 and excellent durability (>104 s) at an ultralow Vds of 50 mV, which benefits to the development of power saving devices. The ReSe2 HS design and high-photoresponsivity ReSe2 channel also achieve an energy efficient optical NVM with full-visible-spectrum distinction. Besides, the ReSe2 HS displays unique ambipolarity to operate as either an inverter or frequency doubler. We further propose a power-free ReSe2 HS optical memory matrix to simplify imaging systems. The ReSe2 HS shows promise for use in light-programmable information storage systems and neuromorphic computation with low power consumption.

Electrical contacts often dominate charge transport properties at the nanoscale because of considerable differences in nanoelectronic device interfaces arising from unique geometric and electrostatic features. Transistors with a tunable Schottky barrier between the metal and semiconductor interface might simplify circuit design. Here, germanium nanowire (Ge NW) transistors with Cu3 Ge as source/drain contacts formed by both buffered oxide etching treatments and rapid thermal annealing are reported. The transistors based on this Cu3 Ge/Ge/Cu3 Ge heterostructure show ambipolar transistor behavior with a large on/off current ratio of more than 105 and 103 for the hole and electron regimes at room temperature, respectively. Investigations of temperature-dependent transport properties and low-frequency current fluctuations reveal that the tunable effective Schottky barriers of the Ge NW transistors accounted for the ambipolar behaviors. It is further shown that this ambipolarity can be used to realize binary-signal and data-storage functions, which greatly simplify circuit design compared with conventional technologies.