Fast Multichannel Inverse Design through Augmented Partial Factorization.

Computer-automated design and discovery have led to high-performance nanophotonic devices with diverse functionalities. However, massively multichannel systems such as metasurfaces controlling many incident angles and photonic-circuit components coupling many waveguide modes still present a challenge. Conventional methods require Min forward simulations and Min adjoint simulations-2Min simulations in total-to compute the objective function and its gradient for a design involving the response to Min input channels. Here, we develop a formalism that uses the recently proposed augmented partial factorization method to obtain both the objective function and its gradient for a massively multichannel system in a single or a few simulations, achieving over 2 orders of magnitude speedup and reduced memory usage. We use this method to inverse design a metasurface beam splitter that separates the incident light to the target diffraction orders for all incident angles of interest, a key component of the dot projector for 3D sensing. This formalism enables efficient inverse design for a wide range of multichannel optical systems.

Fast multi-source nanophotonic simulations using augmented partial factorization.

Numerical solutions of Maxwell's equations are indispensable for nanophotonics and electromagnetics but are constrained when it comes to large systems, especially multi-channel ones such as disordered media, aperiodic metasurfaces and densely packed photonic circuits where the many inputs require many large-scale simulations. Conventionally, before extracting the quantities of interest, Maxwell's equations are first solved on every element of a discretization basis set that contains much more information than is typically needed. Furthermore, such simulations are often performed one input at a time, which can be slow and repetitive. Here we propose to bypass the full-basis solutions and directly compute the quantities of interest while also eliminating the repetition over inputs. We do so by augmenting the Maxwell operator with all the input source profiles and all the output projection profiles, followed by a single partial factorization that yields the entire generalized scattering matrix via the Schur complement, with no approximation beyond discretization. This method applies to any linear partial differential equation. Benchmarks show that this approach is 1,000-30,000,000 times faster than existing methods for two-dimensional systems with about 10,000,000 variables. As examples, we demonstrate simulations of entangled photon backscattering from disorder and high-numerical-aperture metalenses that are thousands of wavelengths wide.

End-Bonded Metal Contacts on WSe2 Field-Effect Transistors.

Contact engineering has been the central issue in the context of high-performance field-effect transistors (FETs) made of atomic thin transition metal dichalcogenides (TMDs). Conventional metal contacts on TMDs have been made on top via a lithography process, forming a top-bonded contact scheme with an appreciable contact barrier. To provide a more efficient pathway for charge injection, an end-bonded contact scheme has been proposed, in which covalent bonds are formed between the contact metal and channel edges. Yet, little efforts have been made to realize this contact configuration. Here, we bridge this gap and demonstrate seeded growth of end-bonded contact with different TMDs by means of chemical vapor deposition (CVD). Monolayer WSe2 FETs with a CVD-grown channel and end contacts exhibit improved performance metrics, including an on-current density of 30 µA/µm, a hole mobility of 90 cm2/V·s, and a subthreshold swing of 94 mV/dec, an order of magnitude superior than those of top-contact FET counterparts that share the same channel material. A fundamental NOT logic gate constructed using top-gated and end-bonded WSe2 and MoS2 FETs is also demonstrated. Calculations using density functional theory indicate that the superior device performance stems mainly from the stronger metal-TMD hybridization and substantial gap states in the end-contact configuration.

Ultrafast Monolayer In/Gr-WS2-Gr Hybrid Photodetectors with High Gain.

One of the primary limitations of previously reported two-dimensional (2D) photodetectors is a low frequency response (≪ 1 Hz) for sensitive devices with gain. Yet, little efforts have been devoted to improve the temporal response of photodetectors while maintaining high gain and responsivity. Here, we demonstrate a gain of 6.3 × 103 electrons per photon and a responsivity of 2.6 × 103 A/W while simultaneously exhibiting an ultrafast response time of 40-65 µs in a hybrid photodetector that consists of graphene-WS2-graphene junctions covered with indium (In) adatoms atop. The resultant responsivity is 6 orders of magnitude higher than that of conventional photodetectors comprising solely of a Au-WS2-Au junction. The photogain is provided mainly by the adsorbed In adatoms, from which photogenerated electrons can be transferred to the WS2 channel, while holes remain trapped in In adatoms, leading to a photogating effect as electrons are recirculating during the residence of holes in In adatoms. At a gate voltage near the Dirac point of graphene, a detectivity of D* = 2.2 × 1012 Jones and an ON/OFF ratio of 104 are achieved. The enhanced performance of the device can be attributed partly to the transparent graphene/WS2 contact and partly to the strong capacitive coupling of the In adatoms with the WS2 channel, which enables ultrafast carrier dynamics.

Stable 1T Tungsten Disulfide Monolayer and Its Junctions: Growth and Atomic Structures.

Transition-metal dichalcogenides in the 1T phase have been a subject of increasing interest, which is partly due to their fascinating physical properties and partly to their potential applications in the next generation of electronic devices, including supercapacitors, electrocatalytic hydrogen evolution, and phase-transition memories. The primary method for obtaining 1T WS2 or MoS2 has been using ion intercalation in combination with solution-based exfoliation. The resulting flakes are small in size and tend to aggregate upon deposition, forming an intercalant-TMD complex with small 1T and 1T' patches embedded in the 2H matrix. Existing growth methods have, however, produced WS2 or MoS2 solely in the 2H phase. Here, we have refined the growth approach to obtain monolayer 1T WS2 up to 80 µm in size based on chemical vapor deposition. With the aid of synergistic catalysts (iron oxide and sodium chloride), 1T WS2 can nucleate in the infant stage of the growth, forming special butterfly-like single crystals with the 1T phase in one wing and the 2H phase in the other. Distinctive types of phase boundaries are discovered at the 1T-2H interface. The 1T structure thus grown is thermodynamically stable over time and even persists at a high temperature above 800 °C, allowing for a stepwise edge epitaxy of lateral 1T heterostructures. Atomic images show that the 1T WS2-MoS2 heterojunction features a coherent and defectless interface with a sharp atomic transition. The stable 1T phase represents a missing piece of the puzzle in the research of atomic thin van der Waals crystals, and our growth approach provides an accessible way of filling this gap.