RESUMEN
Modulating semiconducting channel potential has been used for electrical switching in transistors without biological plasticity operations that are critical for energy-efficient neuromorphic computing. To achieve efficient data processing, alternative transport mechanisms, such as tunneling and thermionic emission, have been introduced with 2D materials. Here, a polymorphic memtransistor based on atomically thin Mo0.91 W0.09 Te2 is presented, where the lattice and electronic structures of the lateral device channel can be tuned as either metallic (1T') or semiconducting (2H) phases by electrical gating. The structural and electronic phase change of the channel material, optimized in Mo0.91 W0.09 Te2 , is explored using transport and optical measurements at the device scale. Based on the phase transition, the polymorphic memtransistor demonstrates a high on/off ratio (up to 105 ), low subthreshold swing (down to 80 mV dec-1 ), and various memristive behaviors, which are distinguished from traditional phase-change memory, transistors, and passive memristors for diverse neuromorphic and in-memory computing.
RESUMEN
Polymorphism allows the symmetry of the lattice and spatial charge distributions of atomically thin materials to be designed. While various polymorphs for superconducting, magnetic, and topological states have been extensively studied, polymorphic control is a challenge for robust ferroelectricity in atomically thin geometries. Here, the atomic and electric manipulation of ferroelectric polymorphs in Mo1- x Wx Te2 is reported. Atomic manipulation for polymorphic control via chemical pressure (substituting tungsten for molybdenum atoms) and charge density modulation can realize tunable polar lattice structures and robust ferroelectricity up to T = 400 K with a constant coercive field in an atomically thin material. Owing to the effective inversion symmetry breaking, the ferroelectric switching withstands a charge carrier density of up to 1.1 × 1013 cm-2 , developing an original diagram for ferroelectric switching in atomically thin materials.
RESUMEN
Tungsten disulfide (WS2) has tunable bandgaps, which are required for diverse optoelectronic device applications. Here, we report the bandgap modulation in WS2 monolayers with two-dimensional core-shell structures formed by unique growth mode in chemical vapor deposition (CVD). The core-shell structures in our CVD-grown WS2 monolayers exhibit contrasts in optical images, Raman, and photoluminescence spectroscopy. The strain and doping effects in the WS2, introduced by two different growth processes, generate PL peaks at 1.83 eV (at the core domain) and 1.98 eV (at the shell domain), which is distinct from conventional WS2 with a primary PL peak at 2.02 eV. Our density functional theory (DFT) calculations explain the modulation of the optical bandgap in our core-shell-structured WS2 monolayers by the strain, accompanying a direct-to-indirect bandgap transition. Thus, the core-shell-structured WS2 monolayers provide a practical method to fabricate lateral heterostructures with different optical bandgaps, which are required for optoelectronic applications.
RESUMEN
Synaptic computation, which is vital for information processing and decision making in neural networks, has remained technically challenging to be demonstrated without using numerous transistors and capacitors, though significant efforts have been made to emulate the biological synaptic transmission such as short-term and long-term plasticity and memory. Here, we report synaptic computation based on Joule heating and versatile doping induced metal-insulator transition in a scalable monolayer-molybdenum disulfide (MoS2) device with a biologically comparable energy consumption (â¼10 fJ). A circuit with our tunable excitatory and inhibitory synaptic devices demonstrates a key function for realizing the most precise temporal computation in the human brain, sound localization: detecting an interaural time difference by suppressing sound intensity- or frequency-dependent synaptic connectivity. This Letter opens a way to implement synaptic computing in neuromorphic applications, overcoming the limitation of scalability and power consumption in conventional CMOS-based neuromorphic devices.
