Integrated 1D epitaxial mirror twin boundaries for ultrascaled 2D MoS2 field-effect transistors.

In atomically thin van der Waals materials, grain boundaries-the line defects between adjacent crystal grains with tilted in-plane rotations-are omnipresent. When the tilting angles are arbitrary, the grain boundaries form inhomogeneous sublattices, giving rise to local electronic states that are not controlled. Here we report on epitaxial realizations of deterministic MoS2 mirror twin boundaries (MTBs) at which two adjoining crystals are reflection mirroring by an exactly 60° rotation by position-controlled epitaxy. We showed that these epitaxial MTBs are one-dimensionally metallic to a circuit length scale. By utilizing the ultimate one-dimensional (1D) feature (width ~0.4 nm and length up to a few tens of micrometres), we incorporated the epitaxial MTBs as a 1D gate to build integrated two-dimensional field-effect transistors (FETs). The critical role of the 1D MTB gate was verified to scale the depletion channel length down to 3.9 nm, resulting in a substantially lowered channel off-current at lower gate voltages. With that, in both individual and array FETs, we demonstrated state-of-the-art performances for low-power logics. The 1D epitaxial MTB gates in this work suggest a novel synthetic pathway for the integration of two-dimensional FETs-that are immune to high gate capacitance-towards ultimate scaling.

Atomically Thin Synapse Networks on Van Der Waals Photo-Memtransistors.

A new type of atomically thin synaptic network on van der Waals (vdW) heterostructures is reported, where each ultrasmall cell (≈2 nm thick) built with trilayer WS2 semiconductor acts as a gate-tunable photoactive synapse, i.e., a photo-memtransistor. A train of UV pulses onto the WS2 memristor generates dopants in atomic-level precision by direct light-lattice interactions, which, along with the gate tunability, leads to the accurate modulation of the channel conductance for potentiation and depression of the synaptic cells. Such synaptic dynamics can be explained by a parallel atomistic resistor network model. In addition, it is shown that such a device scheme can generally be realized in other 2D vdW semiconductors, such as MoS2 , MoSe2 , MoTe2 , and WSe2 . Demonstration of these atomically thin photo-memtransistor arrays, where the synaptic weights can be tuned for the atomistic defect density, provides implications for a new type of artificial neural networks for parallel matrix computations with an ultrahigh integration density.