Session Y56: Devices from 2D Materials: Function, Fabrication and Characterization - IV

Live

In this talk, we review the ultrafast direct measurement on the transient ferroelectric polarization switching in hafnium zirconium oxide with a crossbar metal-insulator-metal (MIM) structure including materials development, device fabrication and optimization, ultrafast electrical pulse measurement setup. A record low sub-nanosecond characteristic switching time of 925 ps was achieved on a single crossbar structure and a record fast polarization switching of 360 ps was achieved with 0.1 μm² crossbar array device structure, supported by the nucleation limited switching model. The impact of electric field, film thickness and device area on the polarization switching speed is systematically studied. The work is in close collaborations with X. Lyu, M. Si from Purdue University and P. R. Shrestha, K. P. Cheung from NIST.   

Live

The past decade has seen the meteoric rise of machine learning and its integration into a wide variety of applications. However, Si-based computational systems are not ideally suited for machine learning applications, due to their power requirements and memory-related bottlenecks, which hinder the use of machine learning on portable, lower-power devices such as cell phones. Resistive switching memory (RSM) is one promising technology for overcoming these issues, as it is expected to enable low power, non-volatile memory cells which also show potential for in-memory computing. Transition metal dichalcogenides (TMDs) with sub-nanometer thickness have previously been used as the active layer in RSM devices. However, it is yet unknown to what degree such TMD-based RSM devices can be scaled down laterally. In this talk, we will discuss our efforts to probe the local switching characteristics of a model RSM device comprising a graphene/MoS₂/Au heterostructure. Our investigations utilize a scanning tunneling microscope probe tip to perform switching operations over sub-100nm regions of the device, as well as assess the morphological and electronic properties of the device’s surface. Using these data, we will discuss the atomic-scale resistive switching behavior of our model RSM device.   

Live

We report on the electronic modulation of the phase and amplitude of scattered light from monolayer molybdenum diselenide (MoSe₂) coupled to quasi bound states in the continuum (BIC) modes in nanoscale AlGaAs resonators. Fermi level control via electronic gating of carrier density gives rise to dynamic phase and amplitude tunability of the reflected light. Our structure exploits the large excitonic reflectivity dominated by radiative emission in MoSe₂ at low temperatures. By measuring the gate-dependent tuning of the reflectivity of MoSe₂ at the A exciton emission wavelength, we can develop a model for the change in the complex dielectric function. We then enhance the emission rate and linewidth of monolayer MoSe₂ by integration with a metasurface comprised of subwavelength AlGaAs cylinders that support high quality factor quasi-BIC modes. This resonant structure is based on a cylinder that utilizes the destructive interference between optical modes to obtain a significant enhancement in the quality factor of the cavity. Gate tunable MoSe₂ monolayers coupled to high Q resonators thus allow us to electronically modulate the phase and amplitude of the reflected light and is a potential candidate for visible frequency active antenna elements comprising a gate-tunable metasurface.   

Live

Layered two-dimensional material (2DM) heterostructures offer anisotropic physical properties that can be tailored via layer assembly without the epitaxial constraints of the bulk. In this work, we analyze the strain modulation of the interlayer current for a variety of nanometer-thick crystalline heterostructures formed with 2DMs, facilitated by their weak mechanical bonding. Descriptions of the tunneling and thermionic emission transport regimes are analyzed using density functional theory and quantum transport calculation in terms of their composition (graphene, boron nitride and transition metal dichalcogenides) and junction length. We develop a model to rationalize these phenomena in terms of the physical properties of the heterostructure constituents (e.g. tunneling rates, density of states, band gap, and Schottky barrier heights). We find piezoresistive gauges in these easily formed systems is comparable to those found in ultraclean bulk systems. Hence, the results of this work show that 2DM heterostructures can expand the repertoire of functional materials and may find applications in sensor technology.   

Live

A flurry of recent developments has established 2D material and van der Waals heterostructure as an ideal solid-state platform for studying novel emergent phenomena as it offers versatile experimental controls for characterizing and manipulating the underlying order. However, experimental efforts to explore the multi-dimensional phase space, e.g. Landau fan measurement in graphene/hexagonal boron nitride structure, often require prolonged measurements. Here we explore a method to accelerate data acquisition by reconstructing an undersampled data set. Landau fan maps measured in graphene/hexagonal boron nitride structure will serve as an example to demonstrate the method’s viability and efficiency. To reconstruct undersampled data, we explore a traditional method as compressive sensing and two deep learning techniques: an enhanced deep residual network for a single image super-resolution model (EDSR), and a Noise2Noise neural network. For the sampling ratio 0.11, the EDSR and Noise2Noise methods indicate comparable performance, while both of the deep learning techniques demonstrate better performance compared to compressed sensing.   

Live

In this talk I report our progress towards the development of van der Waals heterostructures with coupled magnetoelectric properties. Our approach focuses on the MM'P₂S₆ (MM' = CuCr, CuIn, MnCo, MnNi, etc.) family of bi-metallic thiophosphates. These materials are generally Mott or charge-transfer insulators desirable for their diverse functional properties¹; have good air stability; and can be easily grown by our newly developed reactive flux method. I show that these materials exfoliate down to the few layer regime; are amenable to dry transfer stacking; plasma and ion etching; and transport device fabrication in combination with transition metal dichalcogenides (TMDs). A particular focus will be transport approaches towards gate control and detection of magnetization in such systems. Additionally, I present attempts to use TMD heterostructures to probe magnetic ordering in MnNiP₂S₆, a new material whose parent compounds MnPS₃ and NiPS₃ obey dissimilar magnetic orderings². Potential functionalizations such as dark-matter detection through the magnetoelectric effect will also be discussed.

[1] M. A. Susner, Adv. Mater., 29, 2017
[2] P. A. Joy, Phys. Rev. B, 46, 1992   

Live

We describe a method to control the insulator-metal transition at the LaAlO₃/SrTiO₃ interface using ultra-low-voltage electron beam lithography (ULV-EBL). Compared with previous reports that utilize conductive atomic-force-microscope lithography (c-AFM), this approach can provide comparable resolution (~10 nm) at write speeds (10 mm/s) that are up to 10,000x faster than c-AFM. The writing technique is non-destructive and the conductive state is reversible via prolonged exposure to air. Transport properties of representative devices are measured at milli-Kelvin temperatures, where superconducting behavior is observed. We also demonstrate the ability to create conducting devices on graphene/LaAlO₃/SrTiO₃ and hBN/graphene/LaAlO₃/SrTiO₃ heterostructures. The underlying mechanism is believed to be closely related to the same mechanism regulating c-AFM-based methods.   

Live

Graphene, with its vanishing heat capacity and weak electron-phonon coupling at cryogenic temperatures, is a promising material for ultrasensitive calorimetry and single-photon detection. Here, we present thermometry measurements performed on a tunable graphene Josephson Junction (gJJ) integrated into a resonant microwave circuit. Like other forms of resonant readout, resonant temperature readout has the advantage of being fast, continuously monitorable, and straightforward to frequency multiplex. We present results of heating measurements under electron- and hole-doping, a discussion of physical mechanisms, and performance metrics for use as a bolometer and single-photon detector. Our results pave the way for fast bolometers and calorimeters based on van der Waals materials.   

Live

Graphene’s thermal properties are extraordinary, with a predicted thermal conductivity (κ) of 5’000 W/(mK). However, the reported experimental values show variations of one order of magnitude, ranging from 500 to 5’000 W/(mK), with large experimental uncertainties. To overcome these issues, we propose a novel approach to spatially resolve the thermal conductivity of suspended graphene membranes, combining Raman microscopy with finite element method (FEM) calculations. In our method, temperature maps of the graphene membranes are obtained from the shift of the 2D Raman peak with temperature. These maps are then used as input for FEM calculations to extract the corresponding thermal conductivity maps. Moreover, we demonstrate that the thermal conductivity of graphene can be locally tuned in a controlled manner by introducing defects using the focused ion beam technique. This approach paves the way to investigate the thermal properties of other layered materials.   

Live

Thermal properties are important aspects in understanding quantum materials, complementary to electrical properties. For instance, in strongly correlated insulators quantum excitations may carry no electrical charge, which makes their electrical detection difficult. However, for the large family of atomically thin crystals, it has been a long-standing challenge in probing their thermal properties, such as thermal conductivity and heat capacity. In this talk, I will present our effort in detecting thermal properties of a class of two-dimensional materials based on novel nano-device designs, including our preliminary results on experiments in monolayer WTe₂.   

On Demand

Miniaturisation of electronic devices has resulted in increased interconnect resistance–capacitance delay and high power dissipation. Integration of low-k dielectrics—insulating materials that exhibit weak polarisation under applied electric fields—which also serve as diffusion barriers, facilitates miniaturisation beyond the current state-of-the-art. Recommendations of the International Roadmap for Devices and Systems require low-k materials to possess dielectric (k values ≤ 2 by 2028, be mechanically robust, and serve as diffusion barriers against interconnect-atom (typically Cu) migration into semiconductors. However, typical non-polar low-k materials, such as oxide derivatives (SiCOH), organic compounds, and aerogels, exhibit k values exceeding 2 and poor thermo-mechanical properties. In this talk, I will demonstrate realisation of ultra-low k values of 1.89 and 1.29 at 100 kHz and 1 MHz, respectively, in amorphous boron nitride (a-BN) via complementary metal-oxide semiconductor (CMOS)-compatible deposition at 400 °C.[1] The resulting structure is mechanically robust, with excellent diffusion-barrier characteristics. Detailed structural characterisation indicates that a-BN is sp²-hybridised, with no measurable crystallinity. The breakdown strength of a 3-nm thick a-BN sample was 7.3 MV/cm – high enough for contemporary applications. Cross-sectional transmission electron micrographs revealed no diffusion of metal atoms across a-BN under harsh conditions when compared against TiN barriers considered as reference. Hence, our results suggest that the amorphous counterpart of two-dimensional hexagonal boron nitride possesses ideal characteristics for use in next-generation low-k dielectrics for high-performance electronic applications.

[1] S. Hong et al., Nature 582, 511-514 (2020).