Multi-terminal memtransistors from polycrystalline monolayer molybdenum disulfide.

Memristors are two-terminal passive circuit elements that have been developed for use in non-volatile resistive random-access memory and may also be useful in neuromorphic computing. Memristors have higher endurance and faster read/write times than flash memory and can provide multi-bit data storage. However, although two-terminal memristors have demonstrated capacity for basic neural functions, synapses in the human brain outnumber neurons by more than a thousandfold, which implies that multi-terminal memristors are needed to perform complex functions such as heterosynaptic plasticity. Previous attempts to move beyond two-terminal memristors, such as the three-terminal Widrow-Hoff memristor and field-effect transistors with nanoionic gates or floating gates, did not achieve memristive switching in the transistor. Here we report the experimental realization of a multi-terminal hybrid memristor and transistor (that is, a memtransistor) using polycrystalline monolayer molybdenum disulfide (MoS2) in a scalable fabrication process. The two-dimensional MoS2 memtransistors show gate tunability in individual resistance states by four orders of magnitude, as well as large switching ratios, high cycling endurance and long-term retention of states. In addition to conventional neural learning behaviour of long-term potentiation/depression, six-terminal MoS2 memtransistors have gate-tunable heterosynaptic functionality, which is not achievable using two-terminal memristors. For example, the conductance between a pair of floating electrodes (pre- and post-synaptic neurons) is varied by a factor of about ten by applying voltage pulses to modulatory terminals. In situ scanning probe microscopy, cryogenic charge transport measurements and device modelling reveal that the bias-induced motion of MoS2 defects drives resistive switching by dynamically varying Schottky barrier heights. Overall, the seamless integration of a memristor and transistor into one multi-terminal device could enable complex neuromorphic learning and the study of the physics of defect kinetics in two-dimensional materials.

Polymorphism in Post-Dichalcogenide Two-Dimensional Materials.

Two-dimensional (2D) materials exhibit a wide range of atomic structures, compositions, and associated versatility of properties. Furthermore, for a given composition, a variety of different crystal structures (i.e., polymorphs) can be observed. Polymorphism in 2D materials presents a fertile landscape for designing novel architectures and imparting new functionalities. The objective of this Review is to identify the polymorphs of emerging 2D materials, describe their polymorph-dependent properties, and outline methods used for polymorph control. Since traditional 2D materials (e.g., graphene, hexagonal boron nitride, and transition metal dichalcogenides) have already been studied extensively, the focus here is on polymorphism in post-dichalcogenide 2D materials including group III, IV, and V elemental 2D materials, layered group III, IV, and V metal chalcogenides, and 2D transition metal halides. In addition to providing a comprehensive survey of recent experimental and theoretical literature, this Review identifies the most promising opportunities for future research including how 2D polymorph engineering can provide a pathway to materials by design.

Reconfigurable MoS2 Memtransistors for Continuous Learning in Spiking Neural Networks.

Artificial intelligence and machine learning are growing computing paradigms, but current algorithms incur undesirable energy costs on conventional hardware platforms, thus motivating the exploration of more efficient neuromorphic architectures. Toward this end, we introduce here a memtransistor with gate-tunable dynamic learning behavior. By fabricating memtransistors from monolayer MoS2 grown on sapphire, the relative importance of the vertical field effect from the gate is enhanced, thereby heightening reconfigurability of the device response. Inspired by biological systems, gate pulses are used to modulate potentiation and depression, resulting in diverse learning curves and simplified spike-timing-dependent plasticity that facilitate unsupervised learning in simulated spiking neural networks. This capability also enables continuous learning, which is a previously underexplored cognitive concept in neuromorphic computing. Overall, this work demonstrates that the reconfigurability of memtransistors provides unique hardware accelerator opportunities for energy efficient artificial intelligence and machine learning.

Self-Aligned van der Waals Heterojunction Diodes and Transistors.

A general self-aligned fabrication scheme is reported here for a diverse class of electronic devices based on van der Waals materials and heterojunctions. In particular, self-alignment enables the fabrication of source-gated transistors in monolayer MoS2 with near-ideal current saturation characteristics and channel lengths down to 135 nm. Furthermore, self-alignment of van der Waals p-n heterojunction diodes achieves complete electrostatic control of both the p-type and n-type constituent semiconductors in a dual-gated geometry, resulting in gate-tunable mean and variance of antiambipolar Gaussian characteristics. Through finite-element device simulations, the operating principles of source-gated transistors and dual-gated antiambipolar devices are elucidated, thus providing design rules for additional devices that employ self-aligned geometries. For example, the versatility of this scheme is demonstrated via contact-doped MoS2 homojunction diodes and mixed-dimensional heterojunctions based on organic semiconductors. The scalability of this approach is also shown by fabricating self-aligned short-channel transistors with subdiffraction channel lengths in the range of 150-800 nm using photolithography on large-area MoS2 films grown by chemical vapor deposition. Overall, this self-aligned fabrication method represents an important step toward the scalable integration of van der Waals heterojunction devices into more sophisticated circuits and systems.

Ultrafast Exciton Dissociation and Long-Lived Charge Separation in a Photovoltaic Pentacene-MoS2 van der Waals Heterojunction.

van der Waals heterojunctions between two-dimensional (2D) layered materials and nanomaterials of different dimensions present unique opportunities for gate-tunable optoelectronic devices. Mixed-dimensional p-n heterojunction diodes, such as p-type pentacene (0D) and n-type monolayer MoS2 (2D), are especially interesting for photovoltaic applications where the absorption cross-section and charge transfer processes can be tailored by rational selection from the vast library of organic molecules and 2D materials. Here, we study the kinetics of excited carriers in pentacene-MoS2 p-n type-II heterojunctions by transient absorption spectroscopy. These measurements show that the dissociation of MoS2 excitons occurs by hole transfer to pentacene on the time scale of 6.7 ps. In addition, the charge-separated state lives for 5.1 ns, up to an order of magnitude longer than the recombination lifetimes from previously reported 2D material heterojunctions. By studying the fractional amplitudes of the MoS2 decay processes, the hole transfer yield from MoS2 to pentacene is found to be ∼50%, with the remaining holes undergoing trapping due to surface defects. Overall, the ultrafast charge transfer and long-lived charge-separated state in pentacene-MoS2 p-n heterojunctions suggest significant promise for mixed-dimensional van der Waals heterostructures in photovoltaics, photodetectors, and related optoelectronic technologies.

Spiking neurons from tunable Gaussian heterojunction transistors.

Spiking neural networks exploit spatiotemporal processing, spiking sparsity, and high interneuron bandwidth to maximize the energy efficiency of neuromorphic computing. While conventional silicon-based technology can be used in this context, the resulting neuron-synapse circuits require multiple transistors and complicated layouts that limit integration density. Here, we demonstrate unprecedented electrostatic control of dual-gated Gaussian heterojunction transistors for simplified spiking neuron implementation. These devices employ wafer-scale mixed-dimensional van der Waals heterojunctions consisting of chemical vapor deposited monolayer molybdenum disulfide and solution-processed semiconducting single-walled carbon nanotubes to emulate the spike-generating ion channels in biological neurons. Circuits based on these dual-gated Gaussian devices enable a variety of biological spiking responses including phasic spiking, delayed spiking, and tonic bursting. In addition to neuromorphic computing, the tunable Gaussian response has significant implications for a range of other applications including telecommunications, computer vision, and natural language processing.

Electronic Coupling in Metallophthalocyanine-Transition Metal Dichalcogenide Mixed-Dimensional Heterojunctions.

Mixed-dimensional heterojunctions, such as zero-dimensional (0D) organic molecules deposited on two-dimensional (2D) transition metal dichalcogenides (TMDCs), often exhibit interfacial effects that enhance the properties of the individual constituent layers. Here we report a systematic study of interfacial charge transfer in metallophthalocyanine (MPc) - MoS2 heterojunctions using optical absorption and Raman spectroscopy to elucidate M core (M = first row transition metal), MoS2 layer number, and excitation wavelength effects. Observed phenomena include the emergence of heterojunction-specific optical absorption transitions and strong Raman enhancement that depends on the M identity. In addition, the Raman enhancement is tunable by excitation laser wavelength and MoS2 layer number, ultimately reaching a maximum enhancement factor of 30x relative to SiO2 substrates. These experimental results, combined with density functional theory (DFT) calculations, indicate strong coupling between nonfrontier MPc orbitals and the MoS2 band structure as well as charge transfer across the heterojunction interface that varies as a function of the MPc electronic structure.

Charge Separation at Mixed-Dimensional Single and Multilayer MoS2/Silicon Nanowire Heterojunctions.

Layered two-dimensional (2-D) semiconductors can be combined with other low-dimensional semiconductors to form nonplanar mixed-dimensional van der Waals (vdW) heterojunctions whose charge transport behavior is influenced by the heterojunction geometry, providing a new degree of freedom to engineer device functions. Toward that end, we investigated the photoresponse of Si nanowire/MoS2 heterojunction diodes with scanning photocurrent microscopy and time-resolved photocurrent measurements. Comparison of n-Si/MoS2 isotype heterojunctions with p-Si/MoS2 heterojunction diodes under varying biases shows that the depletion region in the p-n heterojunction promotes exciton dissociation and carrier collection. We measure an instrument-limited response time of 1 µs, which is 10 times faster than the previously reported response times for planar Si/MoS2 devices, highlighting the advantages of the 1-D/2-D heterojunction. Finite element simulations of device models provide a detailed understanding of how the electrostatics affect charge transport in nanowire/vdW heterojunctions and inform the design of future vdW heterojunction photodetectors and transistors.

Mechanisms of Ultrafast Charge Separation in a PTB7/Monolayer MoS2 van der Waals Heterojunction.

Mixed-dimensional van der Waals heterojunctions comprising polymer and two-dimensional (2D) semiconductors have many characteristics of an ideal charge separation interface for optoelectronic and photonic applications. However, the photoelectron dynamics at polymer-2D semiconductor heterojunction interfaces are currently not sufficiently understood to guide the optimization of devices for these applications. This Letter reports a systematic exploration of the time-dependent photophysical processes that occur upon photoexcitation of a type-II heterojunction between the polymer PTB7 and monolayer MoS2. In particular, photoinduced electron transfer from PTB7 to electronically hot states of MoS2 occurs in less than 250 fs. This process is followed by a 1-5 ps exciton diffusion-limited electron transfer from PTB7 to MoS2 and a sub-3 ps photoinduced hole transfer from MoS2 to PTB7. The equilibrium between excitons and polaron pairs in PTB7 determines the charge separation yield, whereas the 3-4 ns lifetime of photogenerated carriers is probably limited by MoS2 defects.

Rotationally Commensurate Growth of MoS2 on Epitaxial Graphene.

Atomically thin MoS2/graphene heterostructures are promising candidates for nanoelectronic and optoelectronic technologies. Among different graphene substrates, epitaxial graphene (EG) on SiC provides several potential advantages for such heterostructures, including high electronic quality, tunable substrate coupling, wafer-scale processability, and crystalline ordering that can template commensurate growth. Exploiting these attributes, we demonstrate here the thickness-controlled van der Waals epitaxial growth of MoS2 on EG via chemical vapor deposition, giving rise to transfer-free synthesis of a two-dimensional heterostructure with registry between its constituent materials. The rotational commensurability observed between the MoS2 and EG is driven by the energetically favorable alignment of their respective lattices and results in nearly strain-free MoS2, as evidenced by synchrotron X-ray scattering and atomic-resolution scanning tunneling microscopy (STM). The electronic nature of the MoS2/EG heterostructure is elucidated with STM and scanning tunneling spectroscopy, which reveals bias-dependent apparent thickness, band bending, and a reduced band gap of ∼0.4 eV at the monolayer MoS2 edges.

Mutual Photoluminescence Quenching and Photovoltaic Effect in Large-Area Single-Layer MoS2-Polymer Heterojunctions.

Two-dimensional transition metal dichalcogenides (TMDCs) have recently attracted attention due to their superlative optical and electronic properties. In particular, their extraordinary optical absorption and semiconducting band gap have enabled demonstrations of photovoltaic response from heterostructures composed of TMDCs and other organic or inorganic materials. However, these early studies were limited to devices at the micrometer scale and/or failed to exploit the unique optical absorption properties of single-layer TMDCs. Here we present an experimental realization of a large-area type-II photovoltaic heterojunction using single-layer molybdenum disulfide (MoS2) as the primary absorber, by coupling it to the organic π-donor polymer PTB7. This TMDC-polymer heterojunction exhibits photoluminescence intensity that is tunable as a function of the thickness of the polymer layer, ultimately enabling complete quenching of the TMDC photoluminescence. The strong optical absorption in the TMDC-polymer heterojunction produces an internal quantum efficiency exceeding 40% for an overall cell thickness of less than 20 nm, resulting in exceptional current density per absorbing thickness in comparison to other organic and inorganic solar cells. Furthermore, this work provides insight into the recombination processes in type-II TMDC-polymer heterojunctions and thus provides quantitative guidance to ongoing efforts to realize efficient TMDC-based solar cells.