Symposium S.NM06—Theory and Characterization of 2D Materials—Bridging Atomic Structure and Device Performance

The atomic scale studies on the pristine lattice of TMDs have been carried out with aid from aberration-corrected (scanning) transmission electron microscope (AC-(S)TEM), enabling detailed studies on the evolution of atomic scale defect structures, but their perturbation on local electronic properties, which is essential for understanding their functionality in electronic and optoelectronic devices, are not readily available from direct imaging. Here, we reported the detailed analysis on adatoms, vacancies, line defects, nanopores, nanowires and edge structures in MoS₂ and WS₂ using 4D STEM, which record not only the scattering information, but also the direct beam, and this allows us to extract phase shift, atomic electric field and charge density map based on the shift of center of mass of direct beam.¹ We showed that the electric field map and phase reconstructed imaging are sensitive in detection of abnormal 1D states² and low atomic number atoms,³ and together provide a comprehensive anatomy on the electronic properties of these defect structures, which is of great importance for the theoretical study of 2D electronic devices.

References
[1] Müller-Caspary, K. et al. Ultramicroscopy 178, 62–80 (2017);
[2] Fang, S. et al. Nat. Commun. 10, 1–9 (2019);
[3] Wen, Y. et al. Nano Lett. 19, 6482–6491 (2019).

We report the distinctive defect structures and electron-driven dynamics of a novel 2D monolayer Pd₂Se₃ at the atomic level, which is controllably produced from restructuring few-layered PdSe₂ by thermal stimulation at an in-situ heating stage in scanning transmission electron microscopy (STEM). A rich variety of point vacancies, one-dimensional defects, grain boundaries (GBs) and defect ring complex, which are distinctly different from those in typical transition metal chalcogenides, are directly observed on monolayer Pd₂Se₃. The self-healing point vacancies, gradually consumed edges, and the fast-evolved peculiar defect complex and voids suggest the high mobility of the Se vacancies mobility under electron beam. Multiple unique defects without losing atoms are rendered by the behaviours of special covalent Se−Se dumbbells, which can elastically shift in a staggered way to buffer strains, forming the wave-like one-dimensional defects, and can also undergo Stone-Wales bond rotations which constitute the significant defect behaviour in Pd₂Se₃. The GBs can flexibly form in a meandering pathway and migrate by a sequence of Se−Se bond rotations without vacancies formation, and in the GB-rich corners and tilt GBs, the other highly symmetric vacancy-involved defects also occur to adapt to orientation turning. This report gives insights into the novel defects in new 2D materials featuring high adaptability, rich diversity and dynamics, which opens up more possibilities for exploiting versatile properties of new 2D material systems.

We present experimental evidence of extensive characterization for MoS₂ thin films. Our findings include: in-situ oxidation by induced Raman spectroscopy, scanning electron microscopy and grazing incidence x-ray diffraction and atom probe tomography. Results indicate Results confirm that nanowires plated shape with the 110-orientation are aligned perpendicular film substrate principal reflections at (002), (100), (101), (201), and Raman spectroscopy vibrational modes at E¹_2g at 378 cm⁻¹ and A_1g at 407 cm⁻¹ correspond to 2H-MoS₂. APT reveals MoS⁺², MoS⁺³ as predominant evaporated molecular ions on the sample, indicating no significant diffusion/segregation of Mo or S species within the ITO layer in correlation with STEM measurements. Mechanical properties indicate a harness value of 10 GPa and elastic modulus value of 136 GPa. Finally, some density functional theory calculations reveal honey-comb superlattices can posses ultra-rapid transitions from semiconducting to metallic as reported previously.

Black arsenic (bAs) is a two-dimensional layered material with similar atomic structures to black phosphorus (bP), and it is expected to exhibit attractive properties for future applications such as high carrier mobility, high anisotropy, tunable band structures by varying the number of layers, etc. [1]. Despite the theoretical predictions, it has only recently gained attention due to difficulties in synthesis of high-quality bAs [2]. Recent experimental reports have shown intriguing nature of bAs including high anisotropy [1,3]. Here, we explore the atomic and electronic structures of exfoliated high-quality bAs using high-angle annular dark-field (HAADF)-scanning transmission electron microscopy (STEM) imaging and spectroscopy and demonstrate changes in optical properties in few-layered bAs. Additionally, a stability study on bAs is conducted to gain insight on the key parameters affecting bAs at ambient conditions.
STEM experiments were carried out using an FEI Titan G2 60-300 (S)TEM at 200 keV equipped with energy dispersive X-ray (EDX)and monochromated electron energy-loss spectroscopy (EELS). HAADF-STEM image simulation was carried out using the TEMSIM code based on the Multislice approach [4,5].
Atomic resolution HAADF-STEM images of bulk bAs were obtained from five different orientations including three major crystallographic directions - armchair direction (x-axis), zigzag direction (y-axis), and plan-view direction (z-axis) - confirming anisotropic atomic structures of bAs. Plan-view images were also obtained from few-layered bAs and compared with simulated images that are computed as a function of the number of layers, by which the thickness of the examined layers was measured precisely. The electronic structure of bAs was investigated by using EELS. Low-loss EELS reveals bulk and surface plasma excitation energies and interband transitions, and characteristics of core-loss EELS are described. Low-loss EELS spectra are then acquired as a function of the number of bAs layers and evidence the changes in optical properties in few-layered bAs from bulk bAs. Lastly, degradation of bAs accompanying structure change is analyzed using HAADF-STEM, EDX, and diffraction pattern of specimens. The effect of air, i.e. O₂, and water on the degradation is further investigated and discussed.

Acknowledgment
P.G. and S.J.K. were supported by the NSF under Award No. ECCS-1708769.

[1] Y. Chen, C. Chen, R. Kealhofer, H. Liu, Z. Yuan, L. Jiang, et al., "Black Arsenic: A Layered Semiconductor with Extreme In-Plane Anisotropy," Advanced Materials, 30 1800754 (2018).
[2] O. Osters, T. Nilges, F. Bachhuber, F. Pielnhofer, R. Weihrich, M. Schöneich, et al., "Synthesis and Identification of Metastable Compounds: Black Arsenic—Science or Fiction?," Angewandte Chemie International Edition, 51 2994 (2012).
[3] L. Yu, Z. Zhu, A. Gao, J. Wang, F. Miao, Y. Shi, et al., "Electrically tunable optical properties of few-layer black arsenic phosphorus," Nanotechnology, 29 484001 (2018).
[4] J. M. Cowley, A. F. Moodie, "The scattering of electrons by atoms and crystals. I. A new theoretical approach," Acta Crystallographica, 10 609 (1957).
[5] E. J. Kirkland, "Advanced computing in electron microscopy," Springer, Boston, MA, 2010.

Two-dimensional (2D) materials and heterostructures are promising systems for soft robotics, deformable electronics, and nanoelectromechanical systems because they combine the high charge carrier mobilities of hard materials with the pliability of soft materials. Understanding their bending properties is crucial for the development of these next-generation devices. However, experimental measurements of bending stiffness in 2D materials have been widely divergent; for example, the reported bending stiffnesses of bilayer and trilayer graphene range from 3.4-160 eV and 7-690 eV, respectively [1-3]. Even less is known about the bending stiffness of 2D heterostructures. Here, we show that electron microscopy can provide a powerful platform for measuring the bending properties of 2D materials. We use aberration-corrected scanning transmission electron microscopy (STEM) to image graphene and 2D heterostructures draped over a series of atomically sharp hexagonal boron nitride steps. This approach enables atomic-resolution studies of their bending conformation, producing insight into both the bending stiffness and mechanisms of bending. In combination with density functional theory (DFT) and continuum mechanics modeling, we derive a unifying model for bending in 2D materials and their heterostructures.

First, we investigated the bending stiffness of single and few-layer graphene [4]. We derive mechanical models that relate the bending stiffness to geometric parameters—such as the radius of curvature, bending angle, and step height—measured directly from STEM images. For monolayer graphene, we obtain bending stiffness values of 1.2-1.7 eV, consistent with DFT predictions [5] and the experimental value of 1.2 eV derived from graphite phonon modes [6]. In multilayer graphene, we find that the bending stiffness varies strongly as a function of bending angle, tuning by almost 400% for trilayer graphene. For ten-layer graphene, we show that the bending stiffness can be as low as 18 eV, three orders of magnitude lower than the bending stiffness predicted by conventional thin-film mechanics. This unusual behavior results from the atomic-scale bending mechanism in 2D multilayers, which is dominated by interlayer shear and slip.

Our findings have profound implications on 2D heterostructures, where we demonstrate that the bending stiffness can be controlled by tailoring the interfacial interactions between vertical homo- and heterointerfaces. We investigated heterostructures and showed that, by simply changing their stacking order, we can dramatically tune their bending stiffnesses. Using a combination of DFT and classical simulations, we produce a model to predict the bending stiffness of arbitrary 2D heterostructures. Together, our results provide a new lower limit of bending stiffness for the fabrication of ultrasoft, high mobility electronics and demonstrate new methods to fabricate 2D heterostructures with tailored bending stiffnesses.

References:
[1] Akinwande, D., et al. Extreme Mechanics Letters 13, 42-77 (2017).
[2] Zhang, D.B., et al. Physical Review Letters 106, 3-6 (2011).
[3] Koskinen, P., et al. Physical Review B 82, 1-5 (2010).
[4] Han, E. and Yu, J. et al., Nature Materials (2019). doi: 10.1038/s41563-019-0529-7
[5] Ertekin, E., et al. Physical Review B 79 (2009).
[6] Nicklow, R., et al. Physical Review B 5, 4951-4962 (1972).

Weak interlayer coupling in 2-dimensional layered materials such as graphene gives rise to rich mechanical and electronic properties in particular in the case where the two atomic lattices at the interface are rotated with respect to one another. The reduced crystal symmetry leads to anti-correlations and cancellations of the atomic interactions across the interface, leading to low friction¹ and low interlayer electrical transport². Using our recent nanomanipulation technology³, based on atomic force microscopy, we show that combined electro-mechanical characterization can uniquely address open fundamental questions related to the dielectric interlayer interactions and electronic charge transport through stacking faulted structures⁴. In addition, we studied experimentally and theoretically the interlayer charge transport in twisted bilayer graphene systems separately for edges and bulk parts. We find that interlayer edge currents are several orders of magnitude larger than in the bulk and therefore govern the transport up until very large critical diameters depending on the potential across the adjacent layers.


[1] R. Yaniv et al., Advanced Functional Materials, 1901138 (2019).
[2] E. Koren et al., Nature Nanotech., 9 (2016) 752.
[3] E. Koren et al., Science, 6235 (2015) 679.
[4] R. Bessler et al., Nanoscale advances., 1 (2019) 1702.

Apart from being a powerful imaging technique, atomic force microscopy allows a talented user to forge nanostructures out of tiny objects (e.g. metal clusters) deposited and manipulated on atomically flat surfaces. This is important for fundamental investigations in nanotribology as well as for building up conductive networks with potential applications in molecular electronics. Due to the length scales involved, the control over the morphology of these structures remains, nevertheless, very problematic. Generally speaking, one has to distinguish between nanomanipulations performed by displacing the nano-objects one by one in a time-consuming way or collectively. In this talk we will focus on the second case and discuss “scan-induced-assembly” experiments on Au nanoclusters on mono- and multilayer MoS₂ aimed to (i) understand if the direction of manipulation can be controlled by a proper choice of the scan pattern, and (ii) using the information so obtained for fabricating structures with a desired morphology, such as long but tiny Au stripes [1,2]. While the first goal could be reached with the motion of the clusters precisely related to the angle of attack of the tip on both mono- and multilayers, this is not (yet) the case with the second goal. On MoS₂ monolayers the Au could be indeed rearranged in the form of crumpled stripes with a separation depending on the cluster size and concentration, but the clusters within each stripe remain slightly disconnected, which is possibly due to the roughness of the underlying SiO₂ substrate and/or to charge transfer effects. The presentation is supplemented by molecular dynamics simulations reproducing the displacement of single Au clusters on MoS₂. Differences with collective manipulation experiments on larger and irregularly shaped Sb islands forming incommensurate contacts with MoS₂ [3] will be also discussed.

[1] F. Trillitzsch, R. Guerra, A. Janas, N. Manini, F. Krok and E. Gnecco, Directional and angular locking in the driven motion of Au islands on MoS₂, Phys. Rev. B 98 (2018) 165417
[2] F. Trillitzsch, A. Janas, A. Özogul, C. Neumann, A. George, B.R. Jany, F. Krok, A. Turchanin, and E. Gnecco, Scanning-probe-induced assembling of gold striations on mono- and bi-layered MoS₂ on SiO₂, in preparation
[3] P. Nita, S. Casado, D. Dietzel, A. Schirmeisen, and E. Gnecco, Spinning and translational motion of Sb nanoislands manipulated on MoS2, Nanotechnology 24 (2013) 325302

Charge-transfer (CT) excitons at hetero-interfaces play a critical role in light to electricity conversion using nanostructured materials. However, how CT excitons migrate at these interfaces is poorly understood. Atomically thin and two-dimensional (2D) nanostructures provide a new platform to create architectures with sharp interfaces for directing interfacial charge transport. Here we investigate the formation and transport of interlayer CT excitons in van der Waals (vdW) heterostructures based on semiconducting transition metal dichalcogenides (TMDCs) employing transient absorption microscopy (TAM) with a temporal resolution of 200 fs and spatial precision of 50 nm.
We have recently imaged the transport of interlayer CT excitons in 2D organic-inorganic vdW heterostructures constructed from WS₂ layers and tetracene thin films. To investigate driving force for exciton dissociation, we perform measurements on heterostructures constructed with different WS₂ thickness ranging from l layer to 7 layers. Photoluminescence (PL) measurements confirm the formation of interlayer excitons with a binding energy of ~ 0.3 eV. Electron and hole transfer processes at the interface between monolayer WS₂ and tetracene thin film are very rapid, with time constant of ~ 2 ps and ~ 3 ps, respectively. TAM measurements of exciton transport at these 2D interfaces reveal coexistence of delocalized and localized CT excitons, with diffusion constant of ~ 1 cm²s^-1 and ~ 0.04 cm²s^-1, respectively. The high mobility of the delocalized CT excitons could be the key factor to overcome large CT exciton binding energy in achieving efficient charge separation.
We have also investigated interlayer exciton dynamics and transport modulated by the moiré potentials in WS₂-WSe₂ heterobilayers in time, space, and momentum domains using transient absorption microscopy combined with first-principles calculations. Experimental results verified the theoretical prediction of energetically favorable K-Q interlayer excitons and unraveled exciton-population dynamics that was controlled by the twist-angle-dependent energy difference between the K-Q and K-K excitons. Spatially- and temporally-resolved exciton-population imaging directly visualizes exciton localization by twist-angle-dependent moiré potentials of ~100 meV. Exciton transport deviates significantly from normal diffusion due to the interplay between the moiré potentials and strong many-body interactions, leading to exciton-density- and twist-angle-dependent diffusion length. These results have important implications for designing vdW heterostructures for exciton and spin transport as well as for quantum communication applications.

We present stable long-range diffusion of photocarriers in an innovative 0D-2D heterostructure through contactless all solid state photodoping. We implement 0D indium tin oxide nanocrystals (ITO NCs) as light-driven charge injection sources for localized contactless injection of carriers into the underlying monolayer of molybdenum disulfide (2D MoS₂). This technique gives us the opportunity to follow carrier diffusion in the 2D transition metal dichalcogenide without the need of invasive and broad physical contacts. With light beyond the bandgap we locally excite the nanocrystals within a diffraction limited excitation spot, promoting electrons from the valence band to the conduction band of ITO. The photogenerated holes spontaneously transfer to the underlying 2D material with charge injection densities comparable to p-type doping in electronically gated MoS₂ samples (in the range of 6x10¹²cm^-2). We collect hyperspectral maps of the 0D-2D hybrid, before and after light-driven charge injection to spatially resolve variations in the emission of the monolayer MoS₂. We observe a blue-shift of more than 35 meV of the photoluminescence peak energy and a significant variation in the relative contribution of excitons and trions to the emission, as a consequence of the annihilation of negatively charged excitons. Remarkably, the injected carriers diffuse tens of microns through the monolayer, covering long distances away from the local micron sized excitation spot. The excited holes accumulate preferably in regions that are initially higher in n-doping and preferentially along edges and grain boundaries, following the initial local electronic landscape of the two-dimensional layer. The comparison between hyperspectral maps collected immediately after the photodoping and after months shows no significant variation in the photoluminescence, indicating that the photodoping process is irreversible and stable up to, at least, 73 days. These works demonstrate a new possibility to remotely and locally dope two-dimensional transition metal dichalcogenides in a contact-less way. The accumulation of carriers in specific regions of the monolayer, as observed by us, might play an important role in unveiling the contributions of structural defects, vacancies and strain to the optical properties of 2D materials and in the same time open novel design principles for future applications in the field of energy storage and contactless light-driven nanoelectronics.

A variety of techniques have proven to be important in characterizing the optical and electronic properties of atomically thin transition metal dichalcogenides (TMDCs). Among these two-dimensional (2D) materials, molybdenum disulfide (MoS₂) is considered a good candidate for applications in optoelectronics and photovoltaics. However, predicted device performance is significantly affected by localized defects such as grain boundaries and non-uniformity in large-area films, including regions with varying number of atomic layers. Improvement of techniques for spatially localized investigation of these materials will help accelerate device development from these materials.
In this work, we demonstrate an optical microscope setup capable of transmittance, photoluminescence (PL), and photocurrent spectroscopy with ~1 μm spatial resolution. This spatial-spectral mapping of 2D materials is shown for the visible and near-infrared part of the spectrum. By correlating each of these measurements over the same material and device region, we can draw connections between material properties and device performance. We do this by studying optoelectronic and photovoltaic (PV) devices fabricated on chemical vapor-deposition (CVD) grown MoS₂ samples. We concurrently measure the fundamental properties of various excitons, e.g. transmission at the A, B and C peaks, and identify spatial regions of peak photoresponsivity leading to efficient photocurrent generation in MoS₂ photodetectors.
Our light source is a supercontinuum laser (Fianium/NKT Photonics) with laser line tunable filter, facilitating wavelength dependent excitation and optical studies of our MoS₂. PL mapping of MoS₂ flakes and films within the device active area is carried out by exciting single and few-layer thick MoS₂ samples with a monochromatic 532 nm excitation; emitted PL is measured using an Ocean Optics QePro spectrometer. Transmittance measurements are carried out with Thorlabs FDS1010-CAL calibrated photodiode mounted 1mm beneath the sample on a transparent stage. Device photocurrent is measured using a Keithley 2450 sourcemeter.
While each technique individually yields insightful information on the uniformity and properties of the 2D material, the integrated setup enables spatial correlation of properties as they relate to the performance of 2D PV, photodetectors, and other optoelectronics. From correlated PL and transmission measurements we extract absorption and photogeneration information that enables mapping localized external radiative efficiency. Transmittance measurements probe wavelength dependent absorption and reflectance and their dependence on local defects, while correlated spatial mapping of absorption and photocurrent vs. wavelength reveals the effects on carrier collection and device performance resulting from material uniformity, contact choice, grain boundaries, surface treatments, and defects. Localized layering of materials is visualized through mapping PL intensity and percentage transmittance. We are using this correlative mapping of photonic properties to better understand, design, model, and prototype advanced optoelectronics from 2D materials.

Two-dimensional (2D) van der Waals layered materials have attracted wide interest in condensed matter physics, materials, chemistry and engineering. For allotropes of phosphorus, besides black phosphorus with narrow band gap, a single layered phosphorus with bulked honeycomb structure and wide band gap was studied recently and was called blue phosphorus. Here, by considering many-body effects arising from strong electron-electron and electron-hole interactions in low-dimensional systems, we have studied the electronic and optical properties of 2D blue phosphorus. The intrinsic indirect quasiparticle band gap is 3.76 eV, with a sharp bright and strongly bound exciton located at 2.85 eV, with a binding energy of 0.9 eV. By applying the isotropic strain within 6%, the 1s exciton could be tuned effectively from 2.4 eV to 3.2 eV. In addition, the band topology and winding numbers have been illustrated, explaining the existence of p-like excitons with relatively high oscillator strength. Given recent advances in the successful production of two-dimensional blue phosphorus crystals, we expect them to be profoundly applied in sensor, solar, and lighting technologies.

The motivation behind this project is to study the structure of complex hetero ion based lamellar materials and their physical properties, including ferroelectricity and ferromagnetism, at the 2D limit. By mechanically exfoliating bulk single crystals of CuInP₂S₆, CuCrP₂S₆, and CuCrP₂S_6, a top down approach is used to obtain flakes that are as thin as possible. By isolating a monolayer of these materials, future studies will be conducted to probe for the existence of subnanometer ferroelectric, ferromagnetic, or magnetoelectric multiferroic responses from each respective material systems. Furthermore, studies on heterostructures formed from stacking different monolayers of material in order to study the coupling between different layers are also considered. In hope of its success, we can formulate a device that can take advantage of its magnetoelectric coupling in applications of spintronics and capacitors for more advanced computing.

Using Atomic Force Microscope (AFM) and Raman Spectroscopy, I developed a system to measure the thickness of flakes and to correlate them to Raman spectra and optical images. Despite its challenges, I have successfully measured a range of flake thickness that correlates to a trend in the intensity of the Raman spectra. This allows us a way to identify thin flakes of materials to pursue future studies on the multiferroic character of these materials systems at the 2D limit.

Among the many chemical and physical processes capable of modifying 2D materials, laser plasma processing provides an attractive suite of unique experimental parameters. Using appropriate laser irradiation conditions of a solid target, the ionization fraction, density, temperature, and kinetic energy of a laser-generated plasma can all be adjusted. These plasmas can be used to modify atomically-thin materials. This approach has been employed in the use of a laser-generated sulfur plume for alloying and creation of lateral heterostructures in transition metal dichalcogenides (TMDs) [1]. Further development in this area may be possible with a quantitative understanding of the formation/expansion process of laser-produced chalcogen plumes. In this work, we report on full-scale fluid dynamic simulations of laser-generated chalcogen ion plasmas from initial light-matter interactions to long-range expansion away from a solid target. Contrary to previous studies, our simulation is carried out to centimeter distances, which is the length scale relevant for materials modification. The simulation makes use of implicit partial differential equation solvers and is implemented in a state-of-the-art, multidimensional adaptive Cartesian mesh (ACM) framework. The ACM approach is essential since the spatial resolution required in the early stage of laser-matter interaction is a fraction of a micron while the entire computational domain is several centimeters. As the plasma begins to expand, the fine resolution at the target surface is no longer required, allowing the grid to be coarsened. A dense mesh is still required at the moving plasma front throughout the entire domain and so the ACM follows the front, and allows a coarse mesh elsewhere. The ACM technique makes the computation fast and enables large scale, multidimensional simulations. We simulate the plasma produced during pulsed nanosecond ablation (e.g., KrF excimer) of elemental tellurium and selenium and calculate the kinetic energy distribution of neutral and ionic species, analyzing in detail the evolution of their spatial dynamics. Compared to other metals like copper, bulk tellurium and selenium have drastically lower melting and boiling points as well as mass density. The result of these physical properties is that plasma absorption of the laser pulse begins earlier in the ablation process, and is sustained for longer times, leading to greater vapor densities and temperatures. For instance, at the surface of a copper target, vapor densities are in the 10²³ m^-3 range while a selenium or tellurium target irradiated with the same laser power exhibits vapor densities that are an order of magnitude higher, with a greater number of highly ionized species. Both 1D and 2D-axisymmetric simulations show that singly and doubly ionized species expand outwards with high kinetic energy ahead of the high-density neutral vapor. At high vacuum conditions, reducing the laser irradiance below the plasma absorption threshold minimizes the number of energetic species and the thermal component of the plume is dominant. Simulations such as these provide quantitative understanding of the physical properties of chalcogen ion plasmas and can guide experiments in laser plasma modification of TMDs.

[1] Mahjouri-Samani, M. et. al. Patterned arrays of lateral heterojunctions within monolayer two-dimensional semiconductors. Nat. Commun. 6, 7749 (2015)

The emergence of two dimensional (2D) materials opened up many potential avenues for novel device applications such as nanoelectronics, topological insulators, field effect transistors, microwave and terahertz photonics and many more. To date there are over 1,000 theoretically predicted 2D materials. Of those theoretical materials, only 55 have been experimentally synthesized. The incorporation of this technology into device applications has been hindered due to the difficulty synthesizing and stabilizing the 2D materials. Traditional methods such as chemical vapor deposition result in films with defects and grain boundaries. Controlling the growth of these films requires a systematic understanding of the crucial factors of the film-substrate adhesion strength and mismatch strain. In this work, we present a multi-scale computational approach to model the mesoscopic growth mechanisms of 2D materials on various substrates using density functional theory calculations with van der Waals corrections and phase field methods capturing both atomistic and mesoscopic materials properties.

Various 2D materials are placed epitaxially on substrate surfaces which possess hexagonal symmetry. The 2D materials are pre-screened to ensure that each 2D materials will not possess a strain that is more than 10\%. For all substrate-2D systems the binding and formation energy is computed using a high-throughput approach to density functional theory performed within the atomate framework. These parameters, coupled with the elastic constants for the 2D material are used to fit phase field models to predict the mesoscopic growth of various 2D materials.

1. We gratefully acknowledge ASU’s HPC staff for support and assistance with computing resources along with the Extreme Science and Engineering Discovery Environment (XSEDE), for support by National Science Foundation grant number ACI-1548562, through award number TG-DMR-150006.

From last decade, the interest of graphene is rising spanning almost all research areas due to its outstanding electrical, optical, mechanical, etc. properties inherited by its 2D structure [1]. Due to this remarkable discovery, the material graphene oxide (GO) became famous as a precursor for graphene synthesis, even though it was discovered about 150 years ago [2]. The GO is oxygenated form of graphene with many oxygen-containing groups (O-groups) and lattice defects. Hence, to fabricate graphene from GO, removal of these O-groups and repairing of the lattice defects is vital. This process is known as reduction [3]. However, to obtain pristine graphene from GO is the most challenging task due to the difficulty in complete removal of O-groups and repairing of lattice defects by conventional reduction method.
It has been reported that, reduction of GO by high temperature annealing (>1500 ^oC) is an effective reduction method, particularly in producing conducting RGO films [4]. However, high temperature annealing transforms into high-energy consumption. Further, high temperatures cannot be used on all types of substrates in making graphene films. Therefore, the road to make pristine graphene like material by reducing GO still poses obstacles. As an alternative strategy, thermal annealing of GO in the presence of a carbon source (C-source) such as, ethanol, methane, ethylene, and acetylene via chemical vapour deposition (CVD) has been reported (<1000 ^oC) [5,6]. In our previously published work, fabrication of conducting RGO films by the restoration of graphitic structure via ethanol-CVD was reported [7]. As a continuation of this work, here we report, with the objective to understand the restoration behaviour of lattice defects, thermal annealing of GO in the presence of various organic solvents such as ethanol, methanol, isopropanol, tert-butanol, ethylene glycol, and toluene.
To mention the methodology in brief, first a dispersion of GO was spin-coated on SiO₂/Si substrates and thermal annealing was done in the presence of the above solvents using a CVD device. Characterization of the synthesized samples was done by Raman spectroscopy, atomic force microscopy (AFM) along with Kelvin probe force microscopy (KPFM), and X-ray photoelectron spectroscopy (XPS). Preliminary results showed that there is a significant effect on restoration of lattice defects depending on the organic solvent used. Particularly, the Raman spectroscopic results show the evolution of the G’-peak, which is an indication of restoration of lattice defects, and depending on the C-source the spectral properties change (intensity of G’-peak increases with more carbon atoms). The AFM results showed that the thickness of a single GO sheet has reduced from 1 nm to 0.4 nm due to removal of O-groups upon thermal treatment. Further, conductivity measurements of the synthesized RGO films will be carried out to clarify the solvent dependence on the restoration of lattice defects.
References
[1] Novoselove, K.S., et al., Nature, 2012, 490, 192-200
[2] Brodie, B.C., Phil. Trans. R. Soc. Lond., 1859, 149, 249-259
[3] Pei, S., Cheng, H., Carbon, 2012, 50, 3210-3228
[4] Rozada, R., et al., Nano Res., 2013, 6, 216-233
[5] Su, C., et al., ACS Nano, 2010, 4, 5285-5292
[6] Lopez, V., et al., Adv. Mater., 2009, 21, 4683-4686
[7] De Silva, K.K.H., et al., Jap. J. Appl. Phys., 2019, 58, SIIB07

The unique structures of 2D materials make them to show many excellent electronic, chemical and thermal properties that are absent from their 3D counterparts. To date, tremendous efforts have been devoted to the synthesis of large-area high-quality 2D materials. Moreover, studies have shown that etching, which has been one of the key techniques in the semiconductor society, can be used to create various strictures in 2D materials, which can not be obtained by direct growth (PNAS 110 20386 (2013)). Therefore, A deep understanding on the growth and etching behaviors of 2D materials is crucial for their controllable fabrication. Here, we propose a general model for the growth and etching of two dimensional (2-D) crystals. In this model, low index edges are treated as the basis, while high index ones are described to be composed of terraces along the basis directions and kinks that connect the terraces. Considering the high formation energy penalty of forming a 1D nucleus on a terrace and the easy addition or removal of atoms to or from the kink sites, symmetry dependent edge growth and etching rate profiles of 2D crystals are constructed. By applying the kinetic Wulff construction (KWC) for growth and our modified KWC for etching, we simulated the shape evolutions of 2-, 3-, 4- and 6-fold symmetric 2D crystals during their growth and etching processes, which are well consistent with many experimental observations. This proposed model is expected to be a standard model for the growth and etching of 2D crystals.

Atomic precision phosphorus doping of silicon by scanning tunneling microscope (STM) based hydrogen resist lithography is a promising fabrication platform for creating advanced silicon based electrical and quantum structures. In this process, termed atomic precision advanced manufacturing (APAM), the doping is accomplished by selective depassivation of bound hydrogen with the STM tip, incorporation of phosphine molecules, and capping with epitaxial silicon. APAM devices could provide a platform to learn about the device physics relevant to future transistor technologies. Moreover, these structures can carry a surprising 2 mA/μm of current, potentially enough to integrate APAM structures and devices with modern CMOS transistors. However, these devices have limited applicability because they only operate at cryogenic temperatures, and their ability to withstand the operational environments of CMOS is an open question. In this work, we report on our attempts to demonstrate room temperature devices realized in silicon-on-insulator devices and delta layer analogs, and the discuss the robustness and failure mechanisms at elevated temperature and current densities. These results show utility and challenges of this material platform for future discovery platforms.

This work was supported by the Laboratory Directed Research and Development Program at Sandia National Laboratories and was performed, in part, at the Center for Integrated Nanotechnologies, a U.S. DOE, Office of Basic Energy Sciences user facility. Sandia National Labs is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government.

DNA-assisted bottom-up nanofabrication has demonstrated promise in creating both metal and non-metal nanostructures with diverse applications. To date, different-shaped and electrically connected metal nanostructures have been created and electrically characterized. However, the self-assembly and electrical measurement of semiconducting materials on DNA at the nanoscale have not been demonstrated. We have examined the fabrication yield and electrical properties of various Au nanowire structures created on DNA origami tiles by site-specific attachment of Au nanorods to molecularly programmed sites.¹ We are now working with metals and semiconductors on these DNA nanostructures. We have attached Au and Te nanorods to designed locations on individual DNA origami templates and connected the nanorods via electroless Au deposition technique to create metal-semiconductor junctions. In addition, two-point probe electrical measurements were performed to verify the electrical continuity of the junctions, and non-linear current vs voltage curves were obtained. This work is a step toward self-assembling metals and semiconductors at the nanoscale using DNA origami structures, with potential future nanoelectronics applications.
Reference
1) Aryal, B. R.; Westover, T. R.; Ranasinghe, D. R.; Calvopina, D. G.; Uprety, B.; Harb, J. N.; Davis, R. C.; Woolley, A. T. Four-Point Probe Electrical Measurements on Templated Gold Nanowires Formed on Single DNA Origami Tiles. Langmuir 2018, 34 (49), 15069-15077.

Controlled and meticulous tunability of electronic properties of layered two-dimensional (2D) transition metal dichalcogenides (TMDs) is essential challenge to fabricate TMDs functional devices for the specific applications. Structural defects, crystallinity of the film and dopants used in two-dimensional (2D) transition metal dichalcogenides (TMDs) can significantly modify the material properties. Significant effort has been done to tune the optical and electrical properties of MoS₂ with plasma exposure by maupulating the defects, doping and band structure. However, the observation of structural defects repair and phase change of MoS₂ by plasma exposure has been still debatable. In this study, we will present the tunability of the electrical properties MoS₂ thin film prepared by low pressure chemical vapor deposition (LPCVD) method with low power oxygen plasma treatment. Our electrical transport measurements elucidate the effect of the mild oxygen to the MoS₂ film and able to tune the transport properties of the MoS₂.



This work was supported by U.S. National Science Foundation under grant No. 1728309.

Characterization of the thickness and continuity of wide band gap 2D materials with monolayer sensitivity over large areas has proven to be very challenging. A prime example is 2D hexagonal boron nitride (hBN). Optical contrast methods suffer from the lack of visible absorption in the material; Raman spectral signatures are weak and often not conclusive; and electrical measurements are not possible due to a high electrical resistivity. In this contribution, we will demonstrate an experimental method based on the ellipsometry technique, which makes it possible to map the thickness and continuity of large-area hBN monolayers and bilayers transferred to Si/SiO₂ substrates. The method has sub-monolayer thickness sensitivity, is relatively fast, non-destructive, and can be easily automated. Importantly, artifacts in the measured thickness due to polymer residuals from the transfer process can be deconvolved under most conditions. With some assumptions on the optical functions of hBN, the thickness of an as-transferred hBN monolayer on SiO₂ is measured as 4.1 Å ± 0.1 Å, whereas the thickness of an air- annealed hBN monolayer on SiO₂ is measured as 2.5 Å ± 0.1 Å. The most likely cause of this discrepancy is the presence of a water layer trapped between the SiO₂ surface and the hBN layer in the latter case. The number of hBN layers measured in this study has been confirmed by Raman spectroscopy, x-ray photoemission spectroscopy, and by a series of ellipsometry control experiments. We will present a workflow of our experimental procedure, so that other researchers can extend this characterization method to other 2D materials and hopefully accelerate their development.

Modern medical applications require the design of multifunctional polymers. These substances shall fulfill several roles while implanted in the human body, for example as mechanically supporting devices that can release a drug, followed by a full, non-toxic degradation. A concept for implanting multiple functions is molecular integration through well-defined polymer architectures. Here, structure parameters, which can be varied include composition and ratio of different molecular building blocks as well as the branching type. Moreover, the aqueous biological environment can influence polymer behavior, like hydrogel swelling in water that may trigger both drug release and subsequent degradation.
Material design therefore implies knowledge on structure-function relationships that enable rapid prediction of material properties under specific conditions and external stimuli over sufficiently long time scales. Especially, polymer degradation has major impact on material physicochemical properties and requires thorough understanding of the molecular processes [1]. Importantly, degradation at the polymer surface links to medical applications. For example, medical drugs are delivered in the human body without functional loss if they are encapsulated in a polymer matrix which degrades at the surface without effecting its bulk composition. Recent experimental studies have shown that defined two-dimensional thin film polymer layers are highly suitable to quantitatively study degradation kinetics with Langmuir techniques [2]. In this work we now apply computational simulations to provide an all-atomistic description of the polymer degradation process at the two dimensional interface. A molecular modeling approach is used to show how a molecular monolayer polymer surface film at the water interface behaves at different stages of the degradation process. The models are validated by exploring chemically established commercially available implant materials such as PLGAs (poly(lactide-co-glycolide)s and PCL (poly(ε-caprolactone)). The validation takes random chains as well as end-chain cutting as degradation mechanisms into consideration.
Finally, the model will be used as a predictive tool for digitally assembled copolyesters comprising the same repeating units, but aligned in sequence structure, which in part are challenging to synthesize such as strictly alternating structures. The gained knowledge will form the basis for the design of the next generation of degradable implant materials.

Molybdenum disulphide (MoS₂) as a transition metal dichalcogenides (TMDCs) sketching a wide research interests with unique optical and electronic properties then a bulk. Bandgap engineering technique is a potential way for tuning to its optimization. MoS₂ owing to band gap tailoring from 1.3 to 1.9 eV for bulk (indirect) and single-layered (direct) structures makes it a promising material in the field of energy applications. The present article deals with variation in the properties along with dimensional (1-3D) change in molybdenum disulphide nanostructure. Syntheses, structural, morphological, optical properties of bulk to few layers of nanostructures are investigated. The exfoliation of MoS₂ Flake like nanostructure has been studied for its unique properties. Uniform 3D flower like nanostructures has been prepared by one-pot hydrothermal method. Nanostructure flakes by a solution based exfoliation method are compared with a bulk MoS₂. Structural, morphological and optical properties were investigated by XRD, FESEM, EDAX, UV–Visible for the bulk and nanostructures properties.

Due to their unparalleled mechanical strength and high electronic mobility two dimensional (2D) materials represent the ultimate limit in size of both mechanical atomic membranes and molecular electronics. Bringing these capabilities together make 2D materials and heterostructures promising for emerging technologies like origami robotics, stretchable electronics and mechanically reconfigurable nanoelectromechanical systems. Additionally, many of the most interesting properties of 2D materials and new functionality arise from the interfaces between layers and in engineering multilayer heterostructures. Building an understanding of the mechanics of the van der Waals interface is critical to these next generation devices. In this presentation, we will examine the role of the van der Waals interface in determining the mechanics of deformation in atomic membranes under bending, stretching, and shear and the impact on the behavior of 2D electromechanical systems.
First, we use aberration-corrected scanning transmission electron microscopy to image multilayer graphene and 2D heterostructures draped over a series of atomically sharp hexagonal boron nitride steps and extract. Combining the measurements with atomistic simulations and energy conservation models we extract the bending stiffness over a range of thicknesses and step heights. We find an bending angle and interface orientation dependent bending stiffness. At high bending angles, the bending stiffness to scale linearly with the number of layers due to free slip between layers. At low angles, materials made from commensurate interfaces become stiffer, converging to the cubic scaling with thickness predicted from continuum mechanics. In contrast, heterostructures with an incommensurate interface maintain the linear scaling at all bending angles due to the low interfacial friction. By tailoring the ordering of layers within the heterostructure, we demonstrate tuning of the bending stiffness. These results provide a united description for bending in 2D materials which resolve several years of contradictory measurements and enable the design of bending stiffness in 2D heterostructures with mixed commensurate and incommensurate interfaces.
Next, we engineer electromechanical drumhead resonators from 2D membranes such as twisted bilayer graphene, commensurate bilayer graphene and graphene/MoS₂ bimorphs. In the commensurate bilayer graphene, we observe discrete jumps in frequency which do not appear in monolayer resonators. We find the magnitude of the frequency jumps follows a simple model for changes in stress due to the dynamics of solitons at the interface. Moreover, we find enhanced structural dissipation in twisted 2D interfaces due to the inelastic sliding between layers. Finally, we examine the evolution of buckle and fold instabilities in 2D membranes and heterostructures under compression, and examine the behavior of stretchable electronics made from crumpled 2D heterostructure devices.
Taken together, these experiments show that interfacial slip strongly impacts the mechanics of 2D materials and heterostructures and leads to membranes which are orders of magnitude more deformable than conventional 3D materials.

A major drawback in shrinking down mechanical components to the nanoscale is the excessive dissipation coming from friction at contacting surfaces. A physical regime where this could be substantially avoided is structural superlubricity, wherein an atomically flat and incommensurate van der Waals interface will lead to ultra-low coefficients of friction of <0.001, less than 1% of conventional interfaces. Superlubricity has been experimentally observed via scanning probe measurements of 2D material surfaces, as well as in static sliding of twisted bilayer graphene [1,2] and heterostructure interfaces [3,4]. Moreover, many theoretical studies have predicted and proposed superlubricity in a variety of systems. Yet, nearly three decades after being predicted, superlubric behavior has not been demonstrated in a nanoelectromechanical system, nor has the energy dissipation from superlubric friction ever been directly measured.
In this study, we directly probe the dissipation due to incommensurate interfaces by engineering and comparing mechanical drumhead resonators made from twisted (incommensurate) bilayer, Bernal-stacked (commensurate) bilayer, and monolayer graphene. Because of the ultra-low mass of graphene, the total energy in the resonators is orders of magnitude less than in conventional MEMS systems, making their response extraordinarily sensitive to changes in the energy dissipation. We find that the incommensurate bilayer graphene resonators have damping coefficients more than twice than that of the commensurate counterparts. The extra dissipation in the twisted interface is well described by a structural damping model common in macroscale vibrating components arising from small slips at mechanical joints [5,6]. In the twisted bilayer graphene, the damping arises from inelastic sliding between layers during vibration. From this structural damping model, we extract an upper limit for the interlayer friction stress to be 0.86 mN/m, in good agreement with the reported value obtained in static measurements of interlayer sliding [2]. Additionally, by measuring the mechanical quality factor versus temperature, we observe that friction decreases with decreasing temperature, down to a value < 0.1 mN/m, in agreement with theoretical predictions of a phonon limiting mechanism for superlubric friction. These results unveil the important role of the superlubric interface in mechanical dissipation and provides a principle for designing nanoelectromechanical systems with 2D heterostructures.


[1] M.Dienwiebel, et al. Superlubricity of graphite. Phys. Rev. Lett., 92:126101, 2004.
[2] E. Koren, et al. Adhesion and friction in mesoscopic graphite contacts. Science, 348(6235):679–683, 2015.
[3] S. Kawai, et al. Superlubricity of graphene nanoribbons on gold surfaces. Science, 351(6276):957–961, 2016.
[4] C.M. Harris and A.G. Piersol, Harris’ shock and vibration handbook, McGraw-Hill handbooks, McGraw-Hill, 2002.
[5] Yiming Song, et al. Robust microscale superlubricity in graphite/hexagonal boron nitride layered heterojunctions. Nature Materials, 17(10):894–899, 2018.
[6] W. Chen and X. Deng. Structural damping caused by micro-slip along frictional interfaces. International Journal of Mechanical Sciences, 47(8):1191 – 1211, 2005

Switching mechanism in transistors based on transformation of transition metal dichalcogenides (TMD) between semiconducting and semi-metallic phases may offer superior performance by eliminating static and dynamic power consumption problems associated with scaling of conventional field effect transistors. It has been shown recently that the transformation between semi-metallic TMD 1T’-MoTe₂ and semiconducting MoTe₂ using nanoscale strain engineering results in such switching behavior at device scale. Since strain is the key factor in controlling such electrical properties in these multilayer materials, we use a combination of modelling and experimental approaches to examine and quantify the efficiency of strain transfer in multilayer TMDs and how strain is tied to electrical properties. In this study, molecular dynamics (MD) models are used to predict properties of both monolayer and multilayer structured TMDs under different mechanical loading such as uniaxial, biaxial strain, and nanoindentation. Using the reactive empirical bond order (REBO) potential available in literature along with local structural identification methods such as polyhedral template matching, phase transformation behavior is studied under mechanical loading. Additionally, nanoindentation is used for experimental measurement of properties of different structure types of TMDs (2H and 1T’) for verifying, as well as developing, interlayer van der Waals interaction parameters for atomistic models. Results from the atomistic model coupled with experimental observations using optical microscopy and Raman spectroscopy reveals the mechanism of strain transfer at device scale and its efficiency in various setups.

Modern 2D (two-dimensional) materials such as graphene and beyond are nanoscale materials of strong topical interest. They are recognized as materials with strong potential for diverse industrial applications ranging from quantum to structural material systems. In order to exploit their full potential, it is necessary to develop efficient techniques for their characterization and for atomic-scale understanding of lattice defects in these materials. One lattice defect of primary importance in 2D materials is a monovacancy. It can be treated as a building block of more complicated and extended defects in the materials.

Vacancies and their aggregates, such as antidots and Stonewall defects, play a pivotal role in determining mechanical, thermal, as well as electronic characteristics of 2D materials. One parameter that plays an important role in aggregation and coalescence of vacancies, is the strain field in the material due to the vacancy. Modeling of a monovacancy and its strain field is, therefore, of central interest for 2D materials.
Over the last several years, we have developed powerful techniques at NIST for calculation of Green’s functions (GF) for multiscale modeling of different materials and their application to mechanical and thermal problems. A major advantage of a GF based technique is its computational efficiency. Using this technique, we can model large crystallites containing several million atoms even on a desktop. Modeling large crystallites is especially important for 2D materials because these materials show a significant size effect. It arises because of the logarithmic nature of the response of these materials to a point stimulus. Lattice GF method makes it numerically convenient to account for the size effect. Further, since the GF gives the response of the whole solid at multiscales, it has been shown to seamlessly link the atomistic structure and bulk material performance.

In this talk, we will describe the lattice GF and its application to calculation of the lattice distortion and the strain field in 2D materials. We will present numerical results for silicene, phosphorene and, of course, graphene. Our model crystallite consists of about 2 million atoms. An important input to the GF calculation is the interatomic potential. We have used the functional form of the potentials as available in the literature.

One especially interesting feature of the GF is the possibility of its direct measurement. It has been proposed that the GF is not just a mathematical artefact but a physical entity that can be measured for 2D materials [Tewary et. al., "Green's function modeling of response of two-dimensional materials to point probes for scanning probe microscopy”; Physics Letters A380 (2016) 1750]. If this can be verified, the GF will become a very practical tool for characterization and atomistic scale modeling of 2D materials. This work is another step in that direction.

Transition metal dichalcogenides (TMDs) such as molybdenum disulfide (MoS₂) have gained much interest for scaled nanoelectronics, but their mobilities and drive currents must be improved in order to compete with existing technologies. Theoretical studies predict that tensile strain can increase the mobility of TMDs due to changes in band structure, leading to reduced intervalley scattering [1]. However, such mobility improvement has not yet been experimentally demonstrated.
In this work, we utilize uniaxial tensile strain to improve the electrical performance of monolayer (1L) MoS₂ transistors on flexible substrates. Using a two-point bending apparatus, we apply multiple strain levels between 0% and ~0.7%, which we estimate from the curvature radius of the bent substrates. We confirm these strain levels by Raman spectroscopy, using the known red-shift of the MoS₂ in-plane E’ Raman peak of -2 cm^-1 per percent of tensile strain [2].
We fabricate our devices on polyethylene naphthalate (PEN) flexible substrates. First, the gate metal is lithographically patterned, and then we form the gate dielectric with ~20 nm of Al₂O₃ by atomic layer deposition (ALD). Monolayer MoS₂ grown by chemical vapor deposition (CVD) on separate SiO₂/Si substrates [3] is then transferred to the flexible substrates using a polymer scaffold [4]. Finally, we pattern source and drain contacts onto the MoS₂ films, followed by an O₂ plasma etch to define the channels.
We characterize the effects of tensile strain on MoS₂ transistors with channel lengths from 1 to 15 μm, by extracting the field-effect mobility. Relatively long channels are used to limit extrinsic contributions of contact resistance. We obtain a continuous mobility improvement up to ~85% with increasing tensile strain up to ~0.7%, leading to a drain current I_D increase of ~2x at the same carrier density. Furthermore, we find that the effects are fully reversible as the device characteristics return to their initial state after strain release. The mobility increase with tensile strain also indicates that these devices could be used as strain sensors. An I_D change of ~2x at 0.7% strain corresponds to a gauge factor of nearly ~150, which is the highest value reported so far for piezoresistive 1L CVD MoS₂ strain sensors [5-6].
These results demonstrate the largest mobility and on-state current improvements to MoS₂ transistors using strain to date, revealing that strain engineering is a promising way to tune the electrical performance of TMD-based devices. Furthermore, using MoS₂ transistors as strain sensors can facilitate the realization of flexible and transparent sensor systems for strain mapping.
[1] M. Hosseini et al., J. Phys. D: Appl. Phys. 48, 375104 (2015)
[2] C. Rice et al., Phys. Rev. B 87, 081307 (2013)
[3] K.K.H. Smithe et al., 2D Materials 4, 011009 (2017)
[4] S. Vaziri et al., Sci. Adv. 5, eaax1325 (2019)
[5] Y. J. Park et al., ACS Nano 13, 3023 (2019)
[6] M. Park et al., Adv. Mater. 28, 2556 (2016)

Following the success of ambient-stable two-dimensional (2D) materials such as graphene, hexagonal boron nitride, and transition metal dichalcogenides, new classes of chemically reactive layered solids are being explored since their unique properties hold promise for improved device performance [1] and quantum phenomena [2]. For example, chemically reactive 2D semiconductors (e.g., black phosphorus (BP) and indium selenide (InSe)) have shown enhanced field-effect mobilities and optoelectronic properties under controlled conditions that minimize ambient degradation [3,4]. In addition, 2D boron (i.e., borophene) is an anisotropic metal with a diverse range of theoretically predicted phenomena including confined plasmons, charge density waves, and superconductivity [5], although its high chemical reactivity has limited experimental studies to inert ultrahigh vacuum conditions [6-9]. Therefore, to fully study and exploit the vast majority of 2D materials, methods for mitigating or exploiting their relatively high chemical reactivity are required [10]. In particular, covalent organic functionalization of BP minimizes ambient degradation, provides charge transfer doping, and enhances field-effect mobility [11], while noncovalent organic functionalization of borophene leads to the spontaneous formation of electronically abrupt lateral borophene-organic heterostructures [12]. On the other hand, sequential deposition of atomic carbon and boron on silver substrates results in rotationally commensurate vertical borophene-graphene heterostructures where the graphene adlayer provides robust encapsulation for the underlying borophene [13]. By combining organic and inorganic encapsulation strategies, even highly chemically reactive 2D materials (e.g., InSe and CrI₃) can be studied and utilized in ambient conditions [14].

[1] A. J. Mannix, et al., Nature Reviews Chemistry, 1, 0014 (2017).
[2] X. Liu et al., Nature Reviews Materials, 4, 669 (2019).
[3] J. Kang, et al., Advanced Materials, 30, 1802990 (2018).
[4] C. Husko, et al., Nano Letters, 18, 6515 (2018).
[5] A. J. Mannix, et al., Nature Nanotechnology, 13, 444 (2018).
[6] G. P. Campbell, et al., Nano Letters, 18, 2816 (2018).
[7] X. Liu, et al., Nature Materials, 17, 783 (2018).
[8] X. Liu, et al., Nature Communications, 10, 1642 (2019).
[9] Z. Zhang, et al., Science Advances, 5, eaax0246 (2019).
[10] X. Liu, et al., Advanced Materials, 30, 1801586 (2018).
[11] C. R. Ryder, et al., Nature Chemistry, 8, 597 (2016).
[12] X. Liu, et al., Science Advances, 3, e1602356 (2017).
[13] X. Liu et al., Science Advances, 5, eaax6444 (2019).
[14] S. A. Wells, et al., Nano Letters, 18, 7876 (2018).

Charge density wave (CDW) instabilities are a common phenomenon in many of the layered transition metal dichalcogenides which make them interesting both from a theoretical viewpoint and for practical applications.¹ Bulk TiTe₂ is a semimetal and does not undergo any CDW distortion but according to recent reports its monolayer undergoes a CDW instability² very similar to TiSe₂ bulk and monolayer.^3,4 We performed electronic structure and lattice dynamics calculations (using Phonopy code)⁵ for both of these Ti(IV) dichalcogenides using hybrid (HSE06) density functional theory. We also mapped the potential energy surfaces⁶ spanned by the imaginary mode eigenvectors to estimate the barrier associated with the transition and to track the route to the CDW phase. By treating these structures all at the same high level of theory, we can try to explain why specific phases undergo CDW transition and what the driving force is.
Our results successfully show that though TiTe₂ bulk has no lattice instability, its monolayer has an instability similar to but much weaker than that present in TiSe₂, highlighting the origin of a very weak coupling CDW. The semimetallic overlap in the electronic band structure of TiTe₂ monolayer is only 0.2 eV smaller than that in the bulk supporting the idea that narrow band overlaps thermodynamically drive the CDW distortion. This is the first attempt to perform a complete study of ab-intio electronic and lattice dynamics calculations of the two compounds within the same level of theory. The presence of phonon instability only in the structures that undergo CDW transition shows that the CDW distortion in these compounds is brought about by a lattice instability, and the semimetallic overlap is the key factor that determines how favoured the transition is. These results could have impact in prediction and understanding of CDW instabilities in other systems, particularly the methodology of following imaginary modes to track the transitions between phases.

¹J. A. Wilson et al., Adv. Phys., 1975, 24, 117.
²P. Chen et al., Nat. Commun., 2017, 8, 516.
³F. J. Di Salvo et al., Phys. Rev. B, 1976, 14, 4321
⁴P. Chen et al., Nat. Commun., 2015, 6, 8943.
⁵A. Togo and I. Tanaka, Scr. Mater., 2015, 108, 1-5.
⁶J. M. Skelton et al., Phys. Rev. Lett., 2016, 117, 075502.

Recent experiments found signatures of superconductivity in 'magic angle' twisted bilayer graphene and ABC-trilayer graphene/ hexagonal boron nitride moiré superlattice. Several theoretical works point that the electron-phonon coupling is strong enough to introduce this superconducting phenomenon. In our presentation, phonon spectrum and corresponding phonon modes will be demonstrated by molecular dynamics simulation. These illustrations will help the understanding the interaction between electron and phonon in these multi-layered materials.

We introduce reliable force fields and applications to electrolyte and organic interfaces for molybdenum sulfide and battery oxides, including lithium cobalt oxide as well as nickel and manganese substituted lithium cobalt oxides. Such models, except for MoS2, have not been available and are necessary to understand the dynamics and charge transport properties up to the large nanometer scale. Specifically, the parameters for MoS₂ reliably account for structural, interfacial, and mechanical properties and are compatible with many force fields for molecular simulations (Interface force field, CVFF, PCFF, CHARMM, AMBER, OPLS-AA, TEAM). We reproduce chemical bonding, X-ray structure, cleavage energy, infrared spectrum, bulk modulus, Young’s modulus, and interfacial properties with polar and nonpolar solvents within 0.1% to 5% deviation from measurements. Compared to prior models, structural and mechanical instabilities of the nanoscale layers were eliminated, and large deviations in computed interfacial properties (>50%) were reduced to quantitative agreement with experiment (<5%). Computed contact angles of water and diiodomethane agree within ±2° with experimental measurements on freshly cleaved MoS₂ surfaces. Similarly, new force field parameters for lithium cobalt oxide as well as Co-Ni-Mn mixed oxide phases are shared that reproduce lattice constants, surface energies, contact angles, and mechanical properties in excellent agreement with experiment. The models can be used as part of multiple force fields (IFF, CHARMM, AMBER, OPLS-AA, PCFF, TEAM) for accurate predictions of fully dynamic, large scale interfacial properties and local analysis by quantum mechanics.

The parameters rely on quantitative representation of chemical bonding using atomic charges, using methods that reach much higher accuracy than DFT calculations, bonded parameters from experimental data, and a consistent interpretation of Lennard-Jones parameters that lead to wide compatibility for multiphase materials. All parameters are supported by a physical chemical interpretation. As an example, we discuss the binding mechanism and adsorption energies of peptides on the MoS₂ basal plane using strongly binding and weakly binding sequences derived from phage display. Adsorption is driven by direct surface contact and replacement of surface-bound water molecules, with very widely tunable adsorption energies between -86 and -6 kcal/mol. Weak electrostatic interactions of backbone and side chains with the surface, hydrogen bonds between electrolyte-facing side groups and water, and contributions by hydrophobic groups play a key role to tune the interaction strength.

The models can be applied to any biomaterials and nanomaterials including these 2D compounds, including embedding in electrolytes, polymer matrices, understanding of charge transport, local defects and changes in stoichiometry. The accuracy is comparable to DFT methods at millionfold larger time and length scales, and extensions are feasible for reactive simulations (IFF-R). The models for battery oxides can help overcome problems in designing materials with a higher storage capacity.

Owing to the large surface area and layered nature, the electrical and catalytic properties of 2D materials can be tuned greatly by surface functionalization with tailored molecules with specific redox potentials and intercalation of various intercalants. The utility of surface functionalization and intercalation has been amply demonstrated in 2D transition metal dichalcogenides (TMDCs), including the modulation of carrier densities in monolayer MoS₂ field-effect transistors (FETs) and the enhancement of catalytic activities of hydrogen evolution reaction in Li⁺-intercalated WS₂. However, fundamental mechanistic understandings that govern molecule-TMDC interactions or intercalation-induced structural and electrical phase transformations remain basic.

In this talk, I will present our approach to elucidate the underlying mechanisms that govern surface functionalization and intercalation in 2D TMDCs, using MoS₂ as a material choice. For surface functionalization, we compare doping powers of a family of organic electron donors (OEDs) on MoS₂ to understand the OED-MoS₂ interactions. This requires accurate knowledge of surface coverage of OEDs, change in the carrier density after functionalization, and the nature of the OED bonding to MoS₂. For intercalation, we electrochemically intercalate Li⁺ ions into individual monolayer or heterostructure FETs and follow the change in crystal structure and electrical properties of the host 2D systems as a function of intercalation. We uncover a host of interesting phenomena, such as the role of a hBN/MoS₂ interface on intercalation kinetics and physics, phase transformation in WSe₂ by intercalation, and competition between intercalation kinetics and thermodynamics in MoS₂/graphene heterostructures. A central theme of our approach to understanding surface functionalization and intercalation is the integration of device physics with electrochemistry.

We employed first-principles calculations based on density functional theory to investigate the effect of structural disorder and nitrogen doping on the Li storage capacity of graphene nanomaterials as the anodes of Li-ion batteries. Our calculated results first revealed that the Li storage capacity of monolayer graphene does not necessarily increase with the number of carbon vacancy created but highly depends on the local geometry of the defect sites. The achievable Li capacity limit per vacancy contributed was found to decrease from four to one as the monovacancy grows into a hexavacancy complex, which implies that the coalescence of vacancy defects may tend to lower down the Li storage capacity of monolayer graphene. Our electronic structure analysis further revealed that the enhanced Li storage capacity by the carbon vacancy is mainly attributed to the increased amount of density of states lying just above the Fermi level, which can be much more increased by the local structural disorder around the vacancy sites. Moreover, our calculations also showed that the Li storage capacity of monolayer graphene can be effectively enhanced by the local ring disorder such as the Stone-Wales defect without the presence of any carbon vacancy, which appeared to increase proportionally to the number of the Stone-Wales defect on the basal plane. Our calculations further demonstrated that the amorphous graphene structure can possess a relatively high Li storage capacity primarily owing to the presence of many non-hexagonal ring defects therein. These topological disorders were found to create many electron-deficient regions on the basal plane, which can effectively induce p-type doping on graphene to accommodate more electrons from Li, thereby greatly enhancing the Li storage capacity of the graphene-based nanomaterials. On the other hand, our calculated results also indicate that nitrogen-doping can effectively reduce the vacancy formation energy in graphene, thereby increasing the amount of carbon vacancy to enhance the Li storage capacity. Moreover, the migration energy barriers of vacancy defects were found to increase by nitrogen-doping, which can thus suppress the Li capacity loss induced by the coalescence of vacancy defects in graphene. Nevertheless, our calculations also indicated that it is energetically favorable for nitrogen dopants to aggregate on the vacancy sites, which can thus result in significant irreversible capacity loss of Li because of its extremely high adsorption energy on the vacancy site. Our theoretical findings suggest that the concentration of nitrogen-doping should be carefully controlled in order to obtain optimal Li storage capacity of graphene-based materials.

The theories to describe the rate at which electrochemical reactions proceed, to date, do not consider explicitly the dimensionality or the discreteness and occupancy of the energy levels of the electrodes. We show experimentally that such quantum mechanical aspects are important for dimensionally confined nanostructured materials and yield unusual variation of the kinetic rate constants with applied voltage in single-layer graphene. The observed divergence from conventional electrokinetics was ascribed to the linear energy dispersion as well as a nonzero density of states at the Dirac point in the graphene. The obtained results justify the use of density of states-based rate constants and considerably add to Marcus–Hush–Chidsey kinetics. The talk includes theoretical, computational, and experimental results.

Single atom chalcogen vacancies are common in chemical vapor deposited monolayer transition metal dichalcogenides (TMDs). In some ways, the properties of two dimensional TMDs appear to be not influenced by the presence of up to 1-3% vacancies in the monolayers. That is, it is possible to achieve gate modulation and mobilities of tens of cm²-V^-1-s^-1 at room temperature in field effect transistors (FETs) with single layer TMD channels. Here, we describe the evolution of photoluminescence and FET characteristics as a function of single atom vacancy defects concentration. We find that the monolayer TMDs are defect tolerant – capable of preserving their properties up to defect concentration of ~ 10¹⁴ cm^-2. We also measure the catalytic properties of monolayer TMDs and find that the overall activity increases with number of atomic vacancies – with the best performance (lowest overpotential and highest turnover frequency for the reaction) occurring at a concentration where semiconductor to metal transition occurs. I will describe how our results provide new insights into the defect tolerance of 2D TMDs and how atomic structure engineering can be used to tune their properties.

The phosphorene is a 2D puckered sheet composed of sp³ phosphorus with each P atom is covalently bonded to three P atoms in the plane. The phosphorus atoms contain lone-pair electrons. It exhibits high charge carrier mobility as well as a thickness tuneable band-gap in the 0.3-2.0 eV range.[1] The conduction band minimum of phosphorene is appropriately positioned to effectively catalyze the H₂ evolution reaction by water spliting. Besides, phosphorus is an earth-abundant and environmentally benign non-metal element. Therefore, it is of great interest to exploit the potential of phosphorene as a metal-free photocatalyst for H₂ evolution. However, pristine phsphorene is ambient instable and it produces trace amounts of H₂ by water splitting.[2,3]

The advantage of the lone-pair electrons is that they can be utilized for chemical functionalization of phosphorene without significantly affecting its intrinsic properties. In this work, we present synthesis of ambient stable η–ν adducts of phosphorene, covalently cross-linked phosphorene-MoX₂ (X = S or Se)₂ nanocomposites, and their HER activity. The phosphorenes functionalized with InCl₃ and B(C₆F₅)₃ as well as ylide with a benzyl group are ambient stable and disperse well in aquous medium.[3] Their photocatalytic HER activity is enhanced significantly exhibiting H₂ yields of 6.6 mmol h⁻¹g⁻¹ in the case of phosphorene-B(C₆F₅)₃ (hydrogen yields for pristine phosphorene is 0.6 mmol h⁻¹g⁻¹). Phosphorene-MoX₂ nanocomposites are synthesized by forming amide as the cross-linkages.[4] Due to improved charge-transfer and suppressed charge recombination rate, phosphorene-MoS₂ exhibits excellent photochemical HER activity with H₂ yields of 26.8 mmol h^-1g^-1, while only a negligible amount is produced by their physical mixture. The phosphorene-MoS₂ composite shows high electrochemical HER activity with an onset overpotential of 110 mV, closer to that of Pt/C catalyst. The onset overpotential of a 1:1 physical mixture of phosphorene and MoS₂ is 450 mV. The enhanced HER activity of phosphorene-MoS₂ nanocomposite can be attributed to the ordered cross-linking of the 2D sheets, leading to increased interfacial area as well as the charge-transfer interaction between phosphorene and MoS₂ layers.

[1]. P. Vishnoi, K. Pramoda, C. N. R. Rao, ChemNanoMat, 2019, 5, 1062-1091.
[2]. J. Plutnar, Z. Sofer, M. Pumera, ACS Nano, 2018, 128, 8390-8396.
[3] P. Vishnoi, U. Gupta, R. Pandey, C. N. R. Rao, J. Mater. Chem. A, 2019, 7, 6631-6637.
[4] P. Vishnoi, K. Pramoda, U. Gupta, M. Chhetri, R. Geetha Balakrishna, C. N. R. Rao, ACS Appl. Mater. Interfaces, 2019, 11, 27780-27787.

We have utilized data mining approaches to elucidate over 1000 2D materials and several hundred 3D materials consisting of van der Waals bonded 1D subcomponents, or molecular wires. We find that hundreds of these 2D materials have the potential to exhibit observable piezoelectric effects, representing a new class of piezoelectrics. A further class of layered materials consists of naturally occurring vertical hetero structures, i.e. . bulk crystals that consist of stacks of chemically dissimilar van der Waals bonded layers like a 2-D super lattice. We further combine this data set with physics-based machine learning to discover the chemical composition of an additional 1000 materials that are likely to exhibit layered and two-dimensional phases but have yet to be synthesized. This includes two materials our calculations indicate can exist in distinct structures with different band gaps, expanding the short list of two-dimensional phase change materials. We find our model performs five times better than practitioners in the field at identifying layered materials and is comparable or better than professional solid-state chemists. Finally, we find that semi-supervised learning can offer benefits for materials design where labels for some of the materials are unknown.

Solution processing of graphene [1] allows simple and low-cost techniques such as inkjet printing [2, 3] to be used for fabrication of heterostructures of arbitrary complexity. However, the success of this technology is determined by the nature and quality of the inks used.

In this work we show a general formulation engineering approach to achieve highly concentrated, and inkjet printable water-based 2D crystal formulations, which also provide optimal film formation for heterostructure fabrication [4]. Examples of all-inkjet printed devices, such as large area arrays of photosensors on plastic [4], programmable logic memory devices [4], strain sensors on paper [5], capacitors [6] and transistors [7] will be discussed. In addition, our approach allows easy production of defects-free and biocompatible graphene flakes with positive or negative charge [4,8-10], which can find use in biomedical applications.

References
[1] Coleman et al., Science 2011, 331, 568.
[2] Torrisi et al, ACS Nano 2012, 6, 2992.
[3] Finn et al. J. Mat. Chem. C 2014, 2, 925.
[4] McManus et al, Nature Nanotechnology, 2017, doi:10.1038/nnano.2016.281.
[5] Casiraghi et al, Carbon, 2018, 129, 462.
[6] Worsley et al, ACS Nano, DOI: 10.1021/acsnano.8b06464 [7] Lu et al, ACS Nano, DOI: doi.org/10.1021/acsnano.9b04337 [8] Shin et al, Mol. Syst. Des. Eng., 2019, DOI:10.1039/C9ME00024K [9] Shin et al, submitted.
[10] Shin et al, in preparation.

Two-dimensional transition metal dichalcogenides (TMDCs) such as MoS₂ and WS₂, among others, have gained a lot of attention due to their phase and thickness dependent electrical and optical properties. There is also another class of two-dimensional materials involving similar elements as in TMDCs, which are known as transition metal trichalcogenides (TMTCs)¹_. Contrary to TMDCs, TMTCs are quasi-1D materials, which gives added benefits for applications as they have strong anisotropy in both electrical and optical properties. Among the various transition metal chalcogenides systems, some TMDCs exhibits metallic properties, while their TMTCs counterparts exhibits semiconducting properties (or vice versa). An example are Nb based sulfides: NbS₂ is metallic, whereas NbS₃ is semiconducting.
Lately, the prospects of combining metal-semiconductor transition metal chalcogenides to form heterostructures have gained momentum for future applications in opto-electronics. However, there are many challenges associated with the current heterostructure synthesis processes, such as process scalability, control over the thickness of the individual layers and high processing temperatures. Atomic layer deposition (ALD) can address most of the aforementioned challenges as it offers uniformity over large area substrates, precise thickness control, and low-temperature processing ².
In this work, we focus on fabricating metallic NbS₂, semiconducting NbS₃ heterostructures by ALD. NbS₂ and NbS₃ thin films were synthesized with control over the phase by plasma-enhanced atomic layer deposition (PE-ALD) using a metalorganic precursor and H₂S (+H₂) plasma at low temperatures (200 – 450 °C). The phase-control between NbS₂ and NbS₃ was achieved by two methods, first by varying the temperature (which we have also shown for TiS₂/TiS₃)³. Second, by careful optimization of H₂S:H₂ gas mixture ratio in the plasma co-reactant while maintaining a constant deposition temperature. For demonstrating this, the deposition temperature was maintained at 300 °C. The H₂S:H₂ ratio was varied from 1:0 to 0:1 while maintaining a constant total flow. A high H₂S gas fraction led to the synthesis of NbS₃, otherwise, NbS₂ was deposited. It was observed with optical emission spectroscopy that the increased H species in the plasma mixture could act as an S reducing agent and cause NbS₂ growth at low temperatures. We used this phase-controlled growth of NbS₂ on NbS₃ (or vice versa) by modulating the H₂S:H₂ ratio as a function of ALD cycles to form NbS₂/NbS₃ heterostructures. The formation of heterostructures was confirmed by high-resolution transmission electron microscopy (TEM) along with electron diffraction. Our experiments demonstrate that ALD enables the controlled synthesis of both TMDCs and TMTCs, while also enabling metal-semiconductor heterostructure formation at low temperatures over large scale substrates. This opens up new avenues to include both TMDCs and TMTCs in nano- or optoelectronic applications.
(1) Island, J. O.; Molina-Mendoza, A. J.; Barawi, M.; Biele, R.; Flores, E.; Clamagirand, J. M.; Ares, J. R.; Sánchez, C.; van der Zant, H. S. J.; D’Agosta, R.; et al. Electronics and Optoelectronics of Quasi-1D Layered Transition Metal Trichalcogenides. 2D Mater. 2017, 4 (2), 022003.
(2) Hao, W.; Marichy, C.; Journet, C. Atomic Layer Deposition of Stable 2D Materials. 2D Mater. 2018, 6 (1), 012001.
(3) Basuvalingam, S. B.; Zhang, Y.; Bloodgood, M. A.; Godiksen, R. H.; Curto, A. G.; Hofmann, J. P.; Verheijen, M. A.; Kessels, W. M. M.; Bol, A. A. Low Temperature Phase-Controlled Synthesis of Titanium Di- and Tri-Sulfide by Atomic Layer Deposition. Chem. Mater. 2019, acs.chemmater.9b02895. https://doi.org/10.1021/acs.chemmater.9b02895.

Atomically thin two-dimensional transition metal dichalcogenides (TMDs) have been extensively researched in recent years due to their unique thickness dependent optical and electronic properties, showing remarkable applications in FETs, digital logic circuits, memory devices, photovoltaics, sensors, flexible devices etc. The selenides are garnering a greater interest particularly for the applications in optoelectronics owing to a narrower bandgap and higher optical absorbance compared to sulfides [1]. Although mechanically exfoliated monolayer TMDs are widely used for different proof-of-concept experiments, scaling up the production of large area uniform TMDs to an industrial level still requires an optimized growth method. Several bottom up synthesis processes such as molecular beam epitaxy (MBE) [2] and chemical vapor deposition (CVD) [3] have already been reported in literature, but while the low grain sizes in MBE grown TMDs limits device performances [4], CVD comes with its own challenges e.g., poor control over thickness, low repeatability of growth results etc. In order to achieve large area repeatable growth of TMDs with precise thickness control, a detailed understanding of the role of different parameters influencing the growth dynamics is of utmost importance.
In this work we demonstrate atmospheric pressure CVD (APCVD) growth of MoSe₂ on Si/SiO₂ substrates, and investigate the effect of growth temperature and metal/chalcogen flux. We observe that the growth temperature strongly influences the morphology of the domains and the compact triangular or hexagonal domains ramify into fractals as the growth temperature is decreased. We also observe that higher substrate temperature helps reduce the formation of grain boundaries, which hinder carrier mobilities and introduce defects in the crystal domains by suppressing the nucleation density. The effect of chalcogen environment is studied where growth in a Se-rich environment helps restrict Mo-rich nuclei formation promoting lateral growth. Raman and Photoluminescence (PL) spectroscopy confirms the formation of crystalline monolayer MoSe₂.For the growth in Se-deficient environment, several multilayer islands are seen to form on the domains, showing vertical growth of multilayer MoSe₂ which can be attributed to insufficient Se-passivation. Atomic force microscopy (AFM) study confirms the lateral (monolayer)versus vertical growth under Se-rich and Se-deficient conditions, respectively. XPS analysis show a near perfect stoichiometry (Mo:Se=1:1.97) of MoSe₂ under Se-rich growth environment, whereas in the Se-deficient condition a ratio of Mo:Se=1:1.77 is observed indicating a deviation from the stoichiometric MoSe₂. This also supports our claim of forming metal rich nuclei (Mo_1+xSe_2-x) under Se-deficient condition leading to 3D island formation. This work offers a deeper understanding of the effects of several important growth parameters and proposes an optimized growth window for synthesizing large area 2D TMD compounds.

This work was supported in part by the Army Research Office (ARO) Grant # W911NF-17-1-0312 (MURI) and NSF NNCI (done at the Texas Nanofabrication Facility at the University of Texas at Austin supported by NSF grant NNCI-1542159).


[1] M. Bernardi et al., Nano Lett., 2013, 13, 3664-3670.
[2] R. Yue et al., 2D Mater., 2017, 4, 045019.
[3] X. Wang et al., ACS Nano, 2014, 8, 5125-5131.
[4] A. Roy et al.ACS Appl. Mater. Interfaces 2016, 8, 7396-7402.

Two-dimensional (2D) semiconducting materials, such as MoS₂, WS₂, phosphorene, etc., hold great potential for many important applications, such as in nanoelectronics, molecular and bio-sensors, thermoelectric conversion and solar energy harvesting. It is well recognized that structures and defects of these 2D materials play an import role in dictating the electronic, optical, magnetic and thermal properties. To fully explore the functionalities and potentials of 2D materials, their structure and defect engineering is often required, which can greatly widen their applications.

Neuromorphic computing, which mimics the biological neural architectures, is promising to address some of the challenges facing von Neumann computing system, such as energy efficiency, computational power, and robust learning. 2D materials are promising materials for use in such new computational paradigm.

In this talk, we report our work on the development of multi-terminal MoS₂-based memtransistor for neuromorphic computing. By combining first-principles calculations and theoretical modelling, we investigate the structures and energetics of intrinsic point defects in MoS₂, their evolution and reaction under thermal and electric fields, and their effects on the changes in electrical properties, such as electrical conductivity and Schottky barrier. We further examine the origin of synaptic behavior in the MoS₂-based multi-terminal memtransistor and the performance of such memtransistors, such as, long-term potentiation, long-term depreciation, etc. We also report our realization of an aerosol jet printed Ag/MoS₂/Ag memristor capable of storing and processing data on flexible substrates. This memresistor is realized in a cross bar structure by developing a scalable and low-temperature printing technique utilizing a functional MoS₂ ink platform. Interestingly, the MoS₂ memresistor exhibits both volatile and nonvolatile resistive switching behavior. By using first-principles calculations and kinetic Monte Carlo simulations, we further examine the role of defects and structures in the functionalities and performance of the memresistor.

Our studies here show that these nanodevices are capable of efficiently mimicking many interesting behaviors of biological synapses, demonstrating their potential to enable energy efficient artificial neuromorphic computing.

The metallic 1T phase of MoS₂ has excellent electrical conductivity but it is thermodynamically unstable. Hence there is great interest in realization and stabilization of this phase in different forms such as nanomaterials and thin films for various device applications. The thin film platform affords the element of substrate-induced strain as a tunable parameter that can control the phase equilibria. In the work reported here, we therefore examine the possibility of stabilizing the desirable 1T phase of MoS₂ in the case of thin films grown by pulsed laser deposition (PLD) on different crystalline substrates, namely c-Al₂O₃ (0001), LaAlO₃ (001), SrLaAlO₄ (001), SrTiO₃ (001) and MgO (001). The percent (%) lattice parameter mismatch between the most favorable growth planes of MoS₂ and these different substrates varies from 5% to 16%. c-Al₂O₃, LaAlO₃, SrLaAlO₄ are noted to induce tensile strain whereas SrTiO₃ and MgO cause compressive strain in MoS₂ thin film. Interestingly, Raman and X-ray Photoelectron spectroscopy reveal much enhanced and stable 1T phase contribution in MoS₂ thin films grown on SrTiO₃ and SrLaAlO₄ substrates. The X-ray diffraction, X-ray reflectivity, and atomic force microscopy data reveal high crystalline quality of MoS₂ thin films. The most enhanced 1T phase film of MoS₂ is noted in the case of the film grown on SrLaAlO₄, which also shows a reduction in room temperature resistivity and semi-metal behavior. The valence band spectroscopy (VBS) data for this case are also consistent with the expected metal-like nature of the 1T phase MoS₂ thin film. Furthermore, while increasing the thickness, structural transition from mixed 1T/2H phase to pure 2H hexagonal phase is observed; a consequence of strain relaxation, which clearly establishes the role of substrate-induced strain in 1T phase stabilization via octahedral rotation. This study will pave the way for making the vast family of transition-metal chalcogenides tetragonal phase thin films and brings out the potential of 1T phase thin film for unfolding phenomenon and technological applications.


References:
1. Swati Parmar, Abhijit Biswas, Sachin Kumar Singh, Bishakha Ray, Saurabh Parmar, Suresh Gosavi, Vasant Sathe, Ram Janay Choudhary, Suwarna Datar, and Satishchandra Ogale, Coexisting 1T/2H polymorphs, reentrant resistivity behavior, and charge distribution in MoS₂-hBN 2D/2D composite thin films, Phys. Rev. Materials, 3, 074007 (2019).

Two-dimensional of transition metal dichalcogenides (2D-TMDCs) stand out from the class of 2D materials due to their appealing properties, including atomic-scale thickness, direct band gap and strong spin-orbit coupling. The precise control of the atomic layer structure can pave new avenues for the integration of TMDCs in future optoelectronic devices, such as light-emitting diodes and van der Waals photovoltaics.
In this paper, we present our recent progress on the growth of 2D-TMDCs including MoS₂, WS₂, MoSe₂ and their alloys, by a bottom-up approach, namely Pulsed Laser Deposition (PLD). PLD is a well-established technique for the growth of functional oxide structures¹ and superlattices². It relies on vapor-phase transfer of a material from a target to substrate³. Furthermore, this catalyst-free approach has recently been used for the growth of layered TMDCs⁴. Nevertheless, the optoelectronic properties of 2D-TMDCs, particularly photoluminescence (PL) emission, are often poor. A good control of the chalcogenide vacancies at the monolayer limit is important to realize the full potential of 2D materials and their heterostructures in devices.
In this paper we will discuss the PLD-synthesis of large area, crystalline 2D-TMDCs on sapphire and SiO₂/Si. PLD synthesis at high temperature (~600⁰ C and above) using this one-step approach results in the formation of continuous monolayers with grain boundaries, as compared to the triangular-shaped structures, specific to the chemical vapor deposition process. We profile the vacancy concentration in MoS₂ monolayer on an atomic scale using annular-dark-field electron microscopy, with an absolute detection sensitivity of one to two sulfur vacancies. A rich variety of defects, including single and double sulfur vacancies, anti-site defects, as well as grain boundaries with periodic rings 8-4-4- rings defects are revealed. Room temperature PL spectra of MoS₂ monolayer exhibit two-distinct exciton peaks that can be used to qualitatively asses the as-grown monolayers. Finally, we will discuss the PLD growth of 2D materials using chalcogen-enriched targets as a path towards defect engineering of 2D materials.

1. Ziese, M. et al. Tailoring Magnetic Interlayer Coupling in La0.7Sr0.3MnO3/SrRuO3 Superlattices. Phys. Rev. Lett. 104, (2010).
2. Muller, D. A., Nakagawa, N., Ohtomo, A., Grazul, J. L. & Hwang, H. Y. Atomic-scale imaging of nanoengineered oxygen vacancy profiles in SrTiO3. Nature 430, 657–661 (2004).
3. Canulescu, S. et al. Nonstoichiometric transfer during laser ablation of metal alloys. Phys. Rev. Mater. 1, 73402 (2017).
4. Serna, M. I. et al. Large-Area Deposition of MoS2 by Pulsed Laser Deposition with In Situ Thickness Control. - ACS Nano 10, 6054

<!--[if !supportLineBreakNewLine]-->

Epitaxial Xenes are an emerging class of two-dimensional crystals made of elements spanning group III to group VI of the periodic table (e.g. borophene, silicene, phosphorene, antimomene, tellurene etc) [1]. These materials are synthesized by epitaxy on various substrates and are subject to degradation in environmental conditions. Our work is centered around Xenes grown on noble metal surfaces that are supported by mica, as they can readily undergo delamination and transfer on other substrates for device related studies. In particular, we focus on silicene-on-silver [2] and epitaxial phosphorene [3] as representative cases to establish a stabilization and transfer scheme. Encapsulation of the unstable Xenes is a critical step for achieving transfer of the Xene layer from its pristine growth substrate to a device or functional substrate. We propose the use of an Al₂O₃encapsulation as a stabilization strategy that can be universally applied to the whole class of Xenes. The integrity of the Xene layer after encapsulation is validated by means of Raman spectroscopy and X-ray photoelectron spectroscopy. We then propose a few different transfer schemes and present experimental results of film transfer. A technical step in the delamination and transfer process is the time-controlled etching of the metal substrate. We achieve this using an optimized KI/I₂silver etchant with a controllable etch rate. Finally, we demonstrate the stability of air-exposed multilayer silicene, prepared using our developed methods, via Raman spectroscopy, validating our successive delamination and transfer protocol.

References:
[1] A. Molle et al., Nature Materials (2017) 16, 163.
[2] A. Molle et al., Chemical Society Reviews (2018) 47, 6370.
[3] C. Grazianetti et al., Nanoscale (2019) 11, 18232.

The class of two-dimensional (2D) graphene-like lattices made of atoms out of carbon, collectively known as Xenes, today includes elements from the lightest boron to the heaviest tellurium [1]. The Xenes flow started with silicene that first paved the way to the chance of mimicking the graphene’s properties in an artificial way [2]. Although the widely studied silicene on Ag(111) looks promising for applications in electronics [3], conversely on such a substrate the optical properties can be hardly accessed [4]. The synthesis by molecular beam epitaxy (MBE) of silicene and silicon nanosheets on a transparent substrate like Al₂O₃(0001) allowed for the survey of the thickness-dependent behavior of the optical conductivity obtained from transmittance measurements via Kramers-Kronig constrained fit [5]. At the 2D limit, the optical conductivity is characterized by two main features at 1.4 and 4.5 eV that closely resemble those arising from π-π* and σ-σ* interband transition in freestanding silicene. Two distinct behaviors can be recognized: at the 2D limit, the optical conductivity is consistent with a Dirac-like energy bandstructure, whereas, conversely, for thicker silicon layers an anomalous optical behavior shows up suggesting a different energy bandstructure with respect to that of conventional silicon. On the other hand, limited to the group-14 (i.e. column-IVA) of the periodic table, it turns out that increasing the mass of the X element from carbon to tin, the spin-orbit coupling (SOC), a relativistic effect that scales as Z⁴ in elements of atomic number Z, converts a honeycomb lattice from an ideal 2D semimetallic state to a quantum spin Hall insulator (as predicted first for graphene) characterized by large bandgap opening and conductive dissipationless edge channels [6]. In this framework, the choice of a heavier element than silicon, like tin, would intriguingly afford to unravel the topological properties of the Xenes giving rise to the emergence of non-trivial topological properties even at room temperature. Interestingly, the Al₂O₃(0001) substrate turns out to be also well-suited even for stanene as predicted by theoretical modeling [7]. In close analogy with silicon [5], we investigated the optical properties of tin deposited by MBE on Al₂O₃(0001). The absorbance from THz to ultraviolet (6 meV - 5 eV) photon range measured on ultra-thin tin nanosheets show two spectral features centered at ~1.25 and 4 eV that can be related to π-π* and σ-σ* interband transition in freestanding stanene albeit broadened and shifted towards lower frequency. Remarkably, as also confirmed by means of optical conductivity, ultra-thin tin nanosheets show hints of a bandgap opening of ~40 (0.5 nm-thick) and ~90 (1.5 nm-thick) meV being consistent with SOC induced predicted values. Moreover in the 0.25-1.10 eV range the optical conductance are linear following a power-law frequency dependence that universally describes the interband optical response of D-dimensional Dirac electrons. By and large, the Xenes made of silicon and tin atoms grown on Al₂O₃(0001) might potentially pave the way to a new era in high-speed and low-power nanophotonics based on 2D materials.
References
[1] C. Grazianetti et al., Phys. Stat. Solidi RLL (2019) DOI: 10.1002/pssr.201900439
[2] C. Grazianetti et al., Research 2019, 8494606 (2019)
[3] L. Tao et al., Nature Nanotech. 10, 227 (2015) and C. Grazianetti et al., ACS Nano 11, 3376 (2017)
[4] E. Cinquanta et al., Phys. Rev. B 92, 165427 (2015)
[5] C. Grazianetti et al., Nano Lett. 18, 7124 (2018)
[6] A. Molle et al., Nature Mater. 16, 163 (2017)
[7] H. Wang et al., Phys. Rev. B 94, 035112 (2016)

In this talk, I will discuss our recent research progress in understanding the electronic, photonic and ferroelectric properties of emerging low-dimensional materials, and in developing them for sensing and memory applications. The first part of the talk will focus on discussing the basic properties of emerging 2D materials such as black phosphorus and our progress in developing the material for mid-infrared optoelectronics application. I will also discuss our work on utilizing 4-dimensional imaging techniques to study carrier dynamics in two-dimensional materials, including a study using the newly developed scanning ultrafast electron microscopy (SUEM) technique to image the photo-carrier transport in black phosphorus. In the second part of the talk, I will discuss our recent research on the ferroelectric monolayer materials for memory device applications. I will conclude with remarks on promising future research directions of low-dimensional material properties and devices, and how the emerging materials may benefit future generations of electronics and photonics technology in sensing and memory.

Two-dimensional layered materials have received much interest in recent years due to their ease of miniaturization by mechanical exfoliation along with strong light-matter interactions and tunable electronic properties. Quasi-one-dimensional (Quasi-1D) materials, however, have seen considerably less interest, but can express many of the same desirable properties as conventional layered materials, with an added dimension of anisotropy. A representative example of this Quasi-1D subclass is Titanium Trisulfide (TiS₃). We demonstrate similar ease of mechanical exfoliation of TiS₃ accompanied with theoretical calculations and show that these materials exfoliate into few-atomic-layer nanoribbons with very smooth edges. We emulated macroscopic exfoliation experiments on the nanoscale by applying a local shear force to TiS₃ crystals in different crystallographic directions using a tip of an atomic force microscopy (AFM) probe. In the AFM experiments, it was possible to slide the 2D TiS₃ layers relative to each other as well as to remove selected 1D chains from the layers. Further, their characterization by Raman spectroscopy shows a reliable, internally standardized shift of a few cm^-1 from monolayer to bulk demonstrating tunability typical of conventional layered materials. Devices made from this material show promising electronic properties with predicted mobilities in excess of 10,000 cm²V^-1s^-1 with low edge-scattering, as well as in optoelectronics with strong light-matter interactions, particularly along the crystallographic b-axis (anisotropic). Orientation-dependent properties of this nature show promise as filters and polarizers, while high surface areas associated with 2D materials hint towards gas-sensing and energy storage applications.

Since the first demonstration of atomic resolution in ultrahigh vacuum conditions more than twenty years ago, frequency modulation-based noncontact atomic force microscopy (FM-NC-AFM) has significantly matured and is now routinely applied to study problems that benefit from high-resolution surface imaging. In FM-NC-AFM, control of the tip’s vertical position is accomplished by detecting a shift in the cantilever’s resonance frequency upon approach to the sample. Consistently ensuring reliable distance control during extended data acquisition periods has nevertheless remained challenging, as most FM-mode-based control schemes employ three feedback loops that may interfere. As a consequence, sample throughput in FM-NC-AFM is often low compared to ambient condition AFM, where the easy-to-implement amplitude-modulation (AM) control scheme is predominantly used. Transfer of the AM methodology to high-resolution measurements in vacuum is, however, difficult as with AM-AFM, instabilities during approach are common; in addition, the lack of viscous air damping and the related significant increase of the cantilever’s quality factor generates prolonged settling times that cause the system’s bandwidth to become impractical for many applications. Here we introduce a greatly simplified approach to NC-AFM imaging and quantitative tip-sample interaction force measurement that prevents instabilities while simultaneously enabling data acquisition with customary scan speeds by externally tuning the oscillator’s response characteristics [1]. After discussing the background and basic measurement principles, examples for its application to controlled manipulation of molecules are provided.we will show that the manipulation path can be chosen at will and energy barriers between potential minima on that pathway can be quantified, as can the energy landscape around the molecule before and after manipulation. To explore the practicality of this novel pathway to catalysis research, we selected benzene molecules on a Cu (100) surface as a model system. We first choose a specific manipulation path and then move the tip at constant but continuously reduced heights zalong this path (xcoordinate) while recording the oscillation amplitude Aand phase phi with the microscope operated in our recently developed tuned-oscillator (TO) detection scheme [1]. To preserve the accuracy of recovered tip-sample interaction potentials and forces, we use oscillation amplitudes significantly larger than the decay length of the tip-sample interaction potential [2-5]. Analyzing the full (x, z, A, phi) data array then allows recovery of the potential energy U(x,z) acting between the tip and the sample, the force on the tip normal to the surface vertical tip-sample force F_n(x, z), and the force F_l(x, z) that acts on the tip along the manipulation path (i.e., lateral) with meV, pN, and pm resolution [4,5]. In 54 distinct manipulation events, the molecules were either pushed, pulled, jumped to the tip, or did not move depending on the chemical surrounding of the molecule and the chemical identity of the tip. For further insight, we compared the experimentally measured energy landscapes and manipulation outcomes with computational results, which highlights the decrease of the energy barrier with the variation of the chemical environment of the molecule.
References:
[1] O. E. Dagdeviren et al., Nanotechnology 27, 065703 (2016).
[2] O. E. Dagdeviren et al., Physical Review Applied 9,044040 (2018).
[3] O. E. Dagdeviren et al., Review of Scientific Instruments 90,033707 (2019).
[4] O. E. Dagdeviren et al., Sensors 19, 4510 (2019).
[5] O. E. Dagdeviren et al., Review of Scientific Instruments 90, 013703 (2019). <div id="UMS_TOOLTIP" style="position: absolute; cursor: pointer; z-index: 2147483647; background-color: transparent; top: -100000px; left: -100000px; background-position: initial initial; background-repeat: initial initial;"> </div>

Understanding nanoscale energy dissipation is nowadays among few priorities particularly in solid state systems. Breakdown of topological protection, loss of quantum information and disorder-assisted hot electrons scattering in graphene are just few examples of systems, where the presence of energy dissipation has a great impact on the studied object [1]. It is therefore critical to know, how and where energy leaks. Pendulum geometry Atomic Force Microscope (pAFM), oscillating like a pendulum over the surface, is perfectly suited to measure such tiny amount of dissipation [2,3], since a minimum detectable power loss is of the order of aW.
Here we report on a low temperature (T=5K) measurement of striking singlets or multiplets of dissipation peaks above graphene nanodrums surface. The stress present in the structure leads to formation of few nanometer sized graphene quantum dots (GDS) and the observed dissipation peaks are attributed to tip-induced charge state transitions in quantum-dot- like entities. The dissipation peaks strongly depend on the external magnetic field (B=0T-2T), the behavior we attributed to crossover from quantum dot carrier confinement to the confinement by magnetic field.



[1] – D. Halbertal, et.al., Nanoscale thermal imaging of dissipation in quantum systems, Nature539, (2016), 407–410.
[2] - B.C. Stipe, et.al., Noncontact Friction and Force Fluctuations between Closely Spaced Bodies, Phys. Rev. Lett.87, (2001), 096801.
[3] - M. Kisiel, et.al., Suppression of electronic friction on Nb films in the superconducting state, Nature Materials10, (2011), 119-122.

The electrical properties of deformed graphene nanoribbons (GNR) are of central interest in the development of high mass specific conductivity wiring [1], chemiresitive sensors [2], nanoscale strain gauges, and nanoelectromechanical systems. Despite considerable experimental and computational research, understanding of electromechanical coupling effects in GNR is limited. Recent computational research [3] has developed a general description of electron transmission along circular arcs in semiconducting (3M-1) aGNRs. The computational results suggest that: (1) length and total rotation (along a circular arc) are orthogonal generalized coordinates which determine the resistance of a semiconducting nanowire, (2) the ratio of the curved GNR conductance to that of a corresponding flat nanowire is a linear function of the total rotation over a wide operating range, and (3) at low rotations, the current in a curved GNR can exceed that of a corresponding flat nanowire, due to bandgap effects. The computational results have been shown to be consistent with published data obtained from lifting experiments on 5-aGNRs. An analytical conductance model developed from the computational results generalizes the well known exponential decay law for semiconducting nanowires, and may be applied in the design of experiments and new nanoscale devices.

[1] Zhang, Jie, and Eric P. Fahrenthold. Potassium-Doped Graphene Nanoribbons for High-Specific Conductivity Wiring. ACS Applied Nano Materials 2.5 (2019): 2873-2880.
[2] Zhang, Jie, and Eric P. Fahrenthold. Graphene-Based Sensing of Gas-Phase Explosives. ACS Applied Nano Materials 2.3 (2019): 1445-1456.
[3] Zhang, Jie, and Eric P. Fahrenthold. Conductance of Curved 3M–1 Arm- chair Graphene Nanoribbons. The Journal of Physical Chemistry C 123.35 (2019): 21805-21812.


This work was supported by the Office of Naval Research (Grant N00014-16-2357). Computer time support was provided by the Department of Defense High Performance Computing Modernization Program (Project ONRDC40983493) and the Texas Advanced Computing Center (Project G-815029) at the University of Texas at Austin.

CrTe₂ is a novel magnetic layered transition metal dichalcogenide that has been recently synthesized in a metastable 1T bulk form [1]. While promising strain-tunable magnetic ordering has been proposed in 1T monolayers [2], detailed theoretical insights into the phase and layer thickness dependent electronic and magnetic properties are imperative [3]. In this work, we have investigated the effect of phase type and quantum confinement on the structural, electronic, and magnetic properties of CrTe₂-based systems using first-principle simulations. In the 1T phase, although the inclusion of spin-orbit coupling (SOC) opens up a gap at the K-point in the Brillion zone, the relaxed structures appear to be insensitive to quantum confinement and maintain metallic from the bulk towards the monolayer limit. This weak metallic behavior is generally consistent with thickness dependent resistivity measurements in the few-monolayer regime on e.g., exfoliated 1T-CrTe₂ flakes as recently reported in the literature [4], as well as on our in-house prepared CrTe_2-δ-based thin films. The 1H and 2H phases, on the other hand, typically have finite and thickness-tunable energy gaps. Surprisingly, the 1H-CrTe₂ is nonmagnetic with a finite gap of 0.3-0.4 eV with thickness ranging from 1 to 5 layers, which becomes ferromagnetic and metallic at and above six layers. These tunable, anomalous electronic and magnetic properties controlled by the layer thickness and phase make the extended family of chromium tellurides a particularly rich and intriguing platform for further exploration of spintronic and topologically enabled devices.
1. Freitas, D.C., et al., Ferromagnetism in layered metastable 1T-CrTe₂. Journal of Physics: Condensed Matter, 2015. 27(17): p. 176002.
2. Lv, H.Y., et al., Strain-controlled switch between ferromagnetism and antiferromagnetism in 1T-CrX₂ (X = Se, Te) monolayers. Physical Review B, 2015. 92(21): p. 214419.
3. Bastos, C.M.O., et al., Ab initio investigation of structural stability and exfoliation energies in transition metal dichalcogenides based on Ti-, V-, and Mo-group elements. Physical Review Materials, 2019. 3(4): p. 044002.
4. Sun, X., et al. Room temperature 2D ferromagnetism in few-layered 1T-CrTe₂. 2019; Available from: https://arxiv.org/abs/1909.09797.

The atomic scale studies on the pristine lattice of TMDs have been carried out with aid from aberration-corrected (scanning) transmission electron microscope (AC-(S)TEM), enabling detailed studies on the evolution of atomic scale defect structures, but their perturbation on local electronic properties, which is essential for understanding their functionality in electronic and optoelectronic devices, are not readily available from direct imaging. Here, we reported the detailed analysis on adatoms, vacancies, line defects, nanopores, nanowires and edge structures in MoS₂ and WS₂ using 4D STEM, which record not only the scattering information, but also the direct beam, and this allows us to extract phase shift, atomic electric field and charge density map based on the shift of center of mass of direct beam.¹ We showed that the electric field map and phase reconstructed imaging are sensitive in detection of abnormal 1D states² and low atomic number atoms,³ and together provide a comprehensive anatomy on the electronic properties of these defect structures, which is of great importance for the theoretical study of 2D electronic devices.

References
[1] Müller-Caspary, K. et al. Ultramicroscopy 178, 62–80 (2017);
[2] Fang, S. et al. Nat. Commun. 10, 1–9 (2019);
[3] Wen, Y. et al. Nano Lett. 19, 6482–6491 (2019).

Black arsenic (bAs) is a two-dimensional layered material with similar atomic structures to black phosphorus (bP), and it is expected to exhibit attractive properties for future applications such as high carrier mobility, high anisotropy, tunable band structures by varying the number of layers, etc. [1]. Despite the theoretical predictions, it has only recently gained attention due to difficulties in synthesis of high-quality bAs [2]. Recent experimental reports have shown intriguing nature of bAs including high anisotropy [1,3]. Here, we explore the atomic and electronic structures of exfoliated high-quality bAs using high-angle annular dark-field (HAADF)-scanning transmission electron microscopy (STEM) imaging and spectroscopy and demonstrate changes in optical properties in few-layered bAs. Additionally, a stability study on bAs is conducted to gain insight on the key parameters affecting bAs at ambient conditions.
STEM experiments were carried out using an FEI Titan G2 60-300 (S)TEM at 200 keV equipped with energy dispersive X-ray (EDX)and monochromated electron energy-loss spectroscopy (EELS). HAADF-STEM image simulation was carried out using the TEMSIM code based on the Multislice approach [4,5].
Atomic resolution HAADF-STEM images of bulk bAs were obtained from five different orientations including three major crystallographic directions - armchair direction (x-axis), zigzag direction (y-axis), and plan-view direction (z-axis) - confirming anisotropic atomic structures of bAs. Plan-view images were also obtained from few-layered bAs and compared with simulated images that are computed as a function of the number of layers, by which the thickness of the examined layers was measured precisely. The electronic structure of bAs was investigated by using EELS. Low-loss EELS reveals bulk and surface plasma excitation energies and interband transitions, and characteristics of core-loss EELS are described. Low-loss EELS spectra are then acquired as a function of the number of bAs layers and evidence the changes in optical properties in few-layered bAs from bulk bAs. Lastly, degradation of bAs accompanying structure change is analyzed using HAADF-STEM, EDX, and diffraction pattern of specimens. The effect of air, i.e. O₂, and water on the degradation is further investigated and discussed.

Acknowledgment
P.G. and S.J.K. were supported by the NSF under Award No. ECCS-1708769.

[1] Y. Chen, C. Chen, R. Kealhofer, H. Liu, Z. Yuan, L. Jiang, et al., "Black Arsenic: A Layered Semiconductor with Extreme In-Plane Anisotropy," Advanced Materials, 30 1800754 (2018).
[2] O. Osters, T. Nilges, F. Bachhuber, F. Pielnhofer, R. Weihrich, M. Schöneich, et al., "Synthesis and Identification of Metastable Compounds: Black Arsenic—Science or Fiction?," Angewandte Chemie International Edition, 51 2994 (2012).
[3] L. Yu, Z. Zhu, A. Gao, J. Wang, F. Miao, Y. Shi, et al., "Electrically tunable optical properties of few-layer black arsenic phosphorus," Nanotechnology, 29 484001 (2018).
[4] J. M. Cowley, A. F. Moodie, "The scattering of electrons by atoms and crystals. I. A new theoretical approach," Acta Crystallographica, 10 609 (1957).
[5] E. J. Kirkland, "Advanced computing in electron microscopy," Springer, Boston, MA, 2010.

Two-dimensional (2D) materials and heterostructures are promising systems for soft robotics, deformable electronics, and nanoelectromechanical systems because they combine the high charge carrier mobilities of hard materials with the pliability of soft materials. Understanding their bending properties is crucial for the development of these next-generation devices. However, experimental measurements of bending stiffness in 2D materials have been widely divergent; for example, the reported bending stiffnesses of bilayer and trilayer graphene range from 3.4-160 eV and 7-690 eV, respectively [1-3]. Even less is known about the bending stiffness of 2D heterostructures. Here, we show that electron microscopy can provide a powerful platform for measuring the bending properties of 2D materials. We use aberration-corrected scanning transmission electron microscopy (STEM) to image graphene and 2D heterostructures draped over a series of atomically sharp hexagonal boron nitride steps. This approach enables atomic-resolution studies of their bending conformation, producing insight into both the bending stiffness and mechanisms of bending. In combination with density functional theory (DFT) and continuum mechanics modeling, we derive a unifying model for bending in 2D materials and their heterostructures.

First, we investigated the bending stiffness of single and few-layer graphene [4]. We derive mechanical models that relate the bending stiffness to geometric parameters—such as the radius of curvature, bending angle, and step height—measured directly from STEM images. For monolayer graphene, we obtain bending stiffness values of 1.2-1.7 eV, consistent with DFT predictions [5] and the experimental value of 1.2 eV derived from graphite phonon modes [6]. In multilayer graphene, we find that the bending stiffness varies strongly as a function of bending angle, tuning by almost 400% for trilayer graphene. For ten-layer graphene, we show that the bending stiffness can be as low as 18 eV, three orders of magnitude lower than the bending stiffness predicted by conventional thin-film mechanics. This unusual behavior results from the atomic-scale bending mechanism in 2D multilayers, which is dominated by interlayer shear and slip.

Our findings have profound implications on 2D heterostructures, where we demonstrate that the bending stiffness can be controlled by tailoring the interfacial interactions between vertical homo- and heterointerfaces. We investigated heterostructures and showed that, by simply changing their stacking order, we can dramatically tune their bending stiffnesses. Using a combination of DFT and classical simulations, we produce a model to predict the bending stiffness of arbitrary 2D heterostructures. Together, our results provide a new lower limit of bending stiffness for the fabrication of ultrasoft, high mobility electronics and demonstrate new methods to fabricate 2D heterostructures with tailored bending stiffnesses.

References:
[1] Akinwande, D., et al. Extreme Mechanics Letters 13, 42-77 (2017).
[2] Zhang, D.B., et al. Physical Review Letters 106, 3-6 (2011).
[3] Koskinen, P., et al. Physical Review B 82, 1-5 (2010).
[4] Han, E. and Yu, J. et al., Nature Materials (2019). doi: 10.1038/s41563-019-0529-7
[5] Ertekin, E., et al. Physical Review B 79 (2009).
[6] Nicklow, R., et al. Physical Review B 5, 4951-4962 (1972).

Owing to the large surface area and layered nature, the electrical and catalytic properties of 2D materials can be tuned greatly by surface functionalization with tailored molecules with specific redox potentials and intercalation of various intercalants. The utility of surface functionalization and intercalation has been amply demonstrated in 2D transition metal dichalcogenides (TMDCs), including the modulation of carrier densities in monolayer MoS₂ field-effect transistors (FETs) and the enhancement of catalytic activities of hydrogen evolution reaction in Li⁺-intercalated WS₂. However, fundamental mechanistic understandings that govern molecule-TMDC interactions or intercalation-induced structural and electrical phase transformations remain basic.

In this talk, I will present our approach to elucidate the underlying mechanisms that govern surface functionalization and intercalation in 2D TMDCs, using MoS₂ as a material choice. For surface functionalization, we compare doping powers of a family of organic electron donors (OEDs) on MoS₂ to understand the OED-MoS₂ interactions. This requires accurate knowledge of surface coverage of OEDs, change in the carrier density after functionalization, and the nature of the OED bonding to MoS₂. For intercalation, we electrochemically intercalate Li⁺ ions into individual monolayer or heterostructure FETs and follow the change in crystal structure and electrical properties of the host 2D systems as a function of intercalation. We uncover a host of interesting phenomena, such as the role of a hBN/MoS₂ interface on intercalation kinetics and physics, phase transformation in WSe₂ by intercalation, and competition between intercalation kinetics and thermodynamics in MoS₂/graphene heterostructures. A central theme of our approach to understanding surface functionalization and intercalation is the integration of device physics with electrochemistry.

The phosphorene is a 2D puckered sheet composed of sp³ phosphorus with each P atom is covalently bonded to three P atoms in the plane. The phosphorus atoms contain lone-pair electrons. It exhibits high charge carrier mobility as well as a thickness tuneable band-gap in the 0.3-2.0 eV range.[1] The conduction band minimum of phosphorene is appropriately positioned to effectively catalyze the H₂ evolution reaction by water spliting. Besides, phosphorus is an earth-abundant and environmentally benign non-metal element. Therefore, it is of great interest to exploit the potential of phosphorene as a metal-free photocatalyst for H₂ evolution. However, pristine phsphorene is ambient instable and it produces trace amounts of H₂ by water splitting.[2,3]

The advantage of the lone-pair electrons is that they can be utilized for chemical functionalization of phosphorene without significantly affecting its intrinsic properties. In this work, we present synthesis of ambient stable η–ν adducts of phosphorene, covalently cross-linked phosphorene-MoX₂ (X = S or Se)₂ nanocomposites, and their HER activity. The phosphorenes functionalized with InCl₃ and B(C₆F₅)₃ as well as ylide with a benzyl group are ambient stable and disperse well in aquous medium.[3] Their photocatalytic HER activity is enhanced significantly exhibiting H₂ yields of 6.6 mmol h⁻¹g⁻¹ in the case of phosphorene-B(C₆F₅)₃ (hydrogen yields for pristine phosphorene is 0.6 mmol h⁻¹g⁻¹). Phosphorene-MoX₂ nanocomposites are synthesized by forming amide as the cross-linkages.[4] Due to improved charge-transfer and suppressed charge recombination rate, phosphorene-MoS₂ exhibits excellent photochemical HER activity with H₂ yields of 26.8 mmol h^-1g^-1, while only a negligible amount is produced by their physical mixture. The phosphorene-MoS₂ composite shows high electrochemical HER activity with an onset overpotential of 110 mV, closer to that of Pt/C catalyst. The onset overpotential of a 1:1 physical mixture of phosphorene and MoS₂ is 450 mV. The enhanced HER activity of phosphorene-MoS₂ nanocomposite can be attributed to the ordered cross-linking of the 2D sheets, leading to increased interfacial area as well as the charge-transfer interaction between phosphorene and MoS₂ layers.

[1]. P. Vishnoi, K. Pramoda, C. N. R. Rao, ChemNanoMat, 2019, 5, 1062-1091.
[2]. J. Plutnar, Z. Sofer, M. Pumera, ACS Nano, 2018, 128, 8390-8396.
[3] P. Vishnoi, U. Gupta, R. Pandey, C. N. R. Rao, J. Mater. Chem. A, 2019, 7, 6631-6637.
[4] P. Vishnoi, K. Pramoda, U. Gupta, M. Chhetri, R. Geetha Balakrishna, C. N. R. Rao, ACS Appl. Mater. Interfaces, 2019, 11, 27780-27787.

We introduce reliable force fields and applications to electrolyte and organic interfaces for molybdenum sulfide and battery oxides, including lithium cobalt oxide as well as nickel and manganese substituted lithium cobalt oxides. Such models, except for MoS2, have not been available and are necessary to understand the dynamics and charge transport properties up to the large nanometer scale. Specifically, the parameters for MoS₂ reliably account for structural, interfacial, and mechanical properties and are compatible with many force fields for molecular simulations (Interface force field, CVFF, PCFF, CHARMM, AMBER, OPLS-AA, TEAM). We reproduce chemical bonding, X-ray structure, cleavage energy, infrared spectrum, bulk modulus, Young’s modulus, and interfacial properties with polar and nonpolar solvents within 0.1% to 5% deviation from measurements. Compared to prior models, structural and mechanical instabilities of the nanoscale layers were eliminated, and large deviations in computed interfacial properties (>50%) were reduced to quantitative agreement with experiment (<5%). Computed contact angles of water and diiodomethane agree within ±2° with experimental measurements on freshly cleaved MoS₂ surfaces. Similarly, new force field parameters for lithium cobalt oxide as well as Co-Ni-Mn mixed oxide phases are shared that reproduce lattice constants, surface energies, contact angles, and mechanical properties in excellent agreement with experiment. The models can be used as part of multiple force fields (IFF, CHARMM, AMBER, OPLS-AA, PCFF, TEAM) for accurate predictions of fully dynamic, large scale interfacial properties and local analysis by quantum mechanics.

The parameters rely on quantitative representation of chemical bonding using atomic charges, using methods that reach much higher accuracy than DFT calculations, bonded parameters from experimental data, and a consistent interpretation of Lennard-Jones parameters that lead to wide compatibility for multiphase materials. All parameters are supported by a physical chemical interpretation. As an example, we discuss the binding mechanism and adsorption energies of peptides on the MoS₂ basal plane using strongly binding and weakly binding sequences derived from phage display. Adsorption is driven by direct surface contact and replacement of surface-bound water molecules, with very widely tunable adsorption energies between -86 and -6 kcal/mol. Weak electrostatic interactions of backbone and side chains with the surface, hydrogen bonds between electrolyte-facing side groups and water, and contributions by hydrophobic groups play a key role to tune the interaction strength.

The models can be applied to any biomaterials and nanomaterials including these 2D compounds, including embedding in electrolytes, polymer matrices, understanding of charge transport, local defects and changes in stoichiometry. The accuracy is comparable to DFT methods at millionfold larger time and length scales, and extensions are feasible for reactive simulations (IFF-R). The models for battery oxides can help overcome problems in designing materials with a higher storage capacity.

Following the success of ambient-stable two-dimensional (2D) materials such as graphene, hexagonal boron nitride, and transition metal dichalcogenides, new classes of chemically reactive layered solids are being explored since their unique properties hold promise for improved device performance [1] and quantum phenomena [2]. For example, chemically reactive 2D semiconductors (e.g., black phosphorus (BP) and indium selenide (InSe)) have shown enhanced field-effect mobilities and optoelectronic properties under controlled conditions that minimize ambient degradation [3,4]. In addition, 2D boron (i.e., borophene) is an anisotropic metal with a diverse range of theoretically predicted phenomena including confined plasmons, charge density waves, and superconductivity [5], although its high chemical reactivity has limited experimental studies to inert ultrahigh vacuum conditions [6-9]. Therefore, to fully study and exploit the vast majority of 2D materials, methods for mitigating or exploiting their relatively high chemical reactivity are required [10]. In particular, covalent organic functionalization of BP minimizes ambient degradation, provides charge transfer doping, and enhances field-effect mobility [11], while noncovalent organic functionalization of borophene leads to the spontaneous formation of electronically abrupt lateral borophene-organic heterostructures [12]. On the other hand, sequential deposition of atomic carbon and boron on silver substrates results in rotationally commensurate vertical borophene-graphene heterostructures where the graphene adlayer provides robust encapsulation for the underlying borophene [13]. By combining organic and inorganic encapsulation strategies, even highly chemically reactive 2D materials (e.g., InSe and CrI₃) can be studied and utilized in ambient conditions [14].

[1] A. J. Mannix, et al., Nature Reviews Chemistry, 1, 0014 (2017).
[2] X. Liu et al., Nature Reviews Materials, 4, 669 (2019).
[3] J. Kang, et al., Advanced Materials, 30, 1802990 (2018).
[4] C. Husko, et al., Nano Letters, 18, 6515 (2018).
[5] A. J. Mannix, et al., Nature Nanotechnology, 13, 444 (2018).
[6] G. P. Campbell, et al., Nano Letters, 18, 2816 (2018).
[7] X. Liu, et al., Nature Materials, 17, 783 (2018).
[8] X. Liu, et al., Nature Communications, 10, 1642 (2019).
[9] Z. Zhang, et al., Science Advances, 5, eaax0246 (2019).
[10] X. Liu, et al., Advanced Materials, 30, 1801586 (2018).
[11] C. R. Ryder, et al., Nature Chemistry, 8, 597 (2016).
[12] X. Liu, et al., Science Advances, 3, e1602356 (2017).
[13] X. Liu et al., Science Advances, 5, eaax6444 (2019).
[14] S. A. Wells, et al., Nano Letters, 18, 7876 (2018).

The metallic 1T phase of MoS₂ has excellent electrical conductivity but it is thermodynamically unstable. Hence there is great interest in realization and stabilization of this phase in different forms such as nanomaterials and thin films for various device applications. The thin film platform affords the element of substrate-induced strain as a tunable parameter that can control the phase equilibria. In the work reported here, we therefore examine the possibility of stabilizing the desirable 1T phase of MoS₂ in the case of thin films grown by pulsed laser deposition (PLD) on different crystalline substrates, namely c-Al₂O₃ (0001), LaAlO₃ (001), SrLaAlO₄ (001), SrTiO₃ (001) and MgO (001). The percent (%) lattice parameter mismatch between the most favorable growth planes of MoS₂ and these different substrates varies from 5% to 16%. c-Al₂O₃, LaAlO₃, SrLaAlO₄ are noted to induce tensile strain whereas SrTiO₃ and MgO cause compressive strain in MoS₂ thin film. Interestingly, Raman and X-ray Photoelectron spectroscopy reveal much enhanced and stable 1T phase contribution in MoS₂ thin films grown on SrTiO₃ and SrLaAlO₄ substrates. The X-ray diffraction, X-ray reflectivity, and atomic force microscopy data reveal high crystalline quality of MoS₂ thin films. The most enhanced 1T phase film of MoS₂ is noted in the case of the film grown on SrLaAlO₄, which also shows a reduction in room temperature resistivity and semi-metal behavior. The valence band spectroscopy (VBS) data for this case are also consistent with the expected metal-like nature of the 1T phase MoS₂ thin film. Furthermore, while increasing the thickness, structural transition from mixed 1T/2H phase to pure 2H hexagonal phase is observed; a consequence of strain relaxation, which clearly establishes the role of substrate-induced strain in 1T phase stabilization via octahedral rotation. This study will pave the way for making the vast family of transition-metal chalcogenides tetragonal phase thin films and brings out the potential of 1T phase thin film for unfolding phenomenon and technological applications.


References:
1. Swati Parmar, Abhijit Biswas, Sachin Kumar Singh, Bishakha Ray, Saurabh Parmar, Suresh Gosavi, Vasant Sathe, Ram Janay Choudhary, Suwarna Datar, and Satishchandra Ogale, Coexisting 1T/2H polymorphs, reentrant resistivity behavior, and charge distribution in MoS₂-hBN 2D/2D composite thin films, Phys. Rev. Materials, 3, 074007 (2019).

Two-dimensional of transition metal dichalcogenides (2D-TMDCs) stand out from the class of 2D materials due to their appealing properties, including atomic-scale thickness, direct band gap and strong spin-orbit coupling. The precise control of the atomic layer structure can pave new avenues for the integration of TMDCs in future optoelectronic devices, such as light-emitting diodes and van der Waals photovoltaics.
In this paper, we present our recent progress on the growth of 2D-TMDCs including MoS₂, WS₂, MoSe₂ and their alloys, by a bottom-up approach, namely Pulsed Laser Deposition (PLD). PLD is a well-established technique for the growth of functional oxide structures¹ and superlattices². It relies on vapor-phase transfer of a material from a target to substrate³. Furthermore, this catalyst-free approach has recently been used for the growth of layered TMDCs⁴. Nevertheless, the optoelectronic properties of 2D-TMDCs, particularly photoluminescence (PL) emission, are often poor. A good control of the chalcogenide vacancies at the monolayer limit is important to realize the full potential of 2D materials and their heterostructures in devices.
In this paper we will discuss the PLD-synthesis of large area, crystalline 2D-TMDCs on sapphire and SiO₂/Si. PLD synthesis at high temperature (~600⁰ C and above) using this one-step approach results in the formation of continuous monolayers with grain boundaries, as compared to the triangular-shaped structures, specific to the chemical vapor deposition process. We profile the vacancy concentration in MoS₂ monolayer on an atomic scale using annular-dark-field electron microscopy, with an absolute detection sensitivity of one to two sulfur vacancies. A rich variety of defects, including single and double sulfur vacancies, anti-site defects, as well as grain boundaries with periodic rings 8-4-4- rings defects are revealed. Room temperature PL spectra of MoS₂ monolayer exhibit two-distinct exciton peaks that can be used to qualitatively asses the as-grown monolayers. Finally, we will discuss the PLD growth of 2D materials using chalcogen-enriched targets as a path towards defect engineering of 2D materials.

1. Ziese, M. et al. Tailoring Magnetic Interlayer Coupling in La0.7Sr0.3MnO3/SrRuO3 Superlattices. Phys. Rev. Lett. 104, (2010).
2. Muller, D. A., Nakagawa, N., Ohtomo, A., Grazul, J. L. & Hwang, H. Y. Atomic-scale imaging of nanoengineered oxygen vacancy profiles in SrTiO3. Nature 430, 657–661 (2004).
3. Canulescu, S. et al. Nonstoichiometric transfer during laser ablation of metal alloys. Phys. Rev. Mater. 1, 73402 (2017).
4. Serna, M. I. et al. Large-Area Deposition of MoS2 by Pulsed Laser Deposition with In Situ Thickness Control. - ACS Nano 10, 6054

<!--[if !supportLineBreakNewLine]-->

Charge-transfer (CT) excitons at hetero-interfaces play a critical role in light to electricity conversion using nanostructured materials. However, how CT excitons migrate at these interfaces is poorly understood. Atomically thin and two-dimensional (2D) nanostructures provide a new platform to create architectures with sharp interfaces for directing interfacial charge transport. Here we investigate the formation and transport of interlayer CT excitons in van der Waals (vdW) heterostructures based on semiconducting transition metal dichalcogenides (TMDCs) employing transient absorption microscopy (TAM) with a temporal resolution of 200 fs and spatial precision of 50 nm.
We have recently imaged the transport of interlayer CT excitons in 2D organic-inorganic vdW heterostructures constructed from WS₂ layers and tetracene thin films. To investigate driving force for exciton dissociation, we perform measurements on heterostructures constructed with different WS₂ thickness ranging from l layer to 7 layers. Photoluminescence (PL) measurements confirm the formation of interlayer excitons with a binding energy of ~ 0.3 eV. Electron and hole transfer processes at the interface between monolayer WS₂ and tetracene thin film are very rapid, with time constant of ~ 2 ps and ~ 3 ps, respectively. TAM measurements of exciton transport at these 2D interfaces reveal coexistence of delocalized and localized CT excitons, with diffusion constant of ~ 1 cm²s^-1 and ~ 0.04 cm²s^-1, respectively. The high mobility of the delocalized CT excitons could be the key factor to overcome large CT exciton binding energy in achieving efficient charge separation.
We have also investigated interlayer exciton dynamics and transport modulated by the moiré potentials in WS₂-WSe₂ heterobilayers in time, space, and momentum domains using transient absorption microscopy combined with first-principles calculations. Experimental results verified the theoretical prediction of energetically favorable K-Q interlayer excitons and unraveled exciton-population dynamics that was controlled by the twist-angle-dependent energy difference between the K-Q and K-K excitons. Spatially- and temporally-resolved exciton-population imaging directly visualizes exciton localization by twist-angle-dependent moiré potentials of ~100 meV. Exciton transport deviates significantly from normal diffusion due to the interplay between the moiré potentials and strong many-body interactions, leading to exciton-density- and twist-angle-dependent diffusion length. These results have important implications for designing vdW heterostructures for exciton and spin transport as well as for quantum communication applications.

We present stable long-range diffusion of photocarriers in an innovative 0D-2D heterostructure through contactless all solid state photodoping. We implement 0D indium tin oxide nanocrystals (ITO NCs) as light-driven charge injection sources for localized contactless injection of carriers into the underlying monolayer of molybdenum disulfide (2D MoS₂). This technique gives us the opportunity to follow carrier diffusion in the 2D transition metal dichalcogenide without the need of invasive and broad physical contacts. With light beyond the bandgap we locally excite the nanocrystals within a diffraction limited excitation spot, promoting electrons from the valence band to the conduction band of ITO. The photogenerated holes spontaneously transfer to the underlying 2D material with charge injection densities comparable to p-type doping in electronically gated MoS₂ samples (in the range of 6x10¹²cm^-2). We collect hyperspectral maps of the 0D-2D hybrid, before and after light-driven charge injection to spatially resolve variations in the emission of the monolayer MoS₂. We observe a blue-shift of more than 35 meV of the photoluminescence peak energy and a significant variation in the relative contribution of excitons and trions to the emission, as a consequence of the annihilation of negatively charged excitons. Remarkably, the injected carriers diffuse tens of microns through the monolayer, covering long distances away from the local micron sized excitation spot. The excited holes accumulate preferably in regions that are initially higher in n-doping and preferentially along edges and grain boundaries, following the initial local electronic landscape of the two-dimensional layer. The comparison between hyperspectral maps collected immediately after the photodoping and after months shows no significant variation in the photoluminescence, indicating that the photodoping process is irreversible and stable up to, at least, 73 days. These works demonstrate a new possibility to remotely and locally dope two-dimensional transition metal dichalcogenides in a contact-less way. The accumulation of carriers in specific regions of the monolayer, as observed by us, might play an important role in unveiling the contributions of structural defects, vacancies and strain to the optical properties of 2D materials and in the same time open novel design principles for future applications in the field of energy storage and contactless light-driven nanoelectronics.

Two-dimensional (2D) van der Waals layered materials have attracted wide interest in condensed matter physics, materials, chemistry and engineering. For allotropes of phosphorus, besides black phosphorus with narrow band gap, a single layered phosphorus with bulked honeycomb structure and wide band gap was studied recently and was called blue phosphorus. Here, by considering many-body effects arising from strong electron-electron and electron-hole interactions in low-dimensional systems, we have studied the electronic and optical properties of 2D blue phosphorus. The intrinsic indirect quasiparticle band gap is 3.76 eV, with a sharp bright and strongly bound exciton located at 2.85 eV, with a binding energy of 0.9 eV. By applying the isotropic strain within 6%, the 1s exciton could be tuned effectively from 2.4 eV to 3.2 eV. In addition, the band topology and winding numbers have been illustrated, explaining the existence of p-like excitons with relatively high oscillator strength. Given recent advances in the successful production of two-dimensional blue phosphorus crystals, we expect them to be profoundly applied in sensor, solar, and lighting technologies.

A variety of techniques have proven to be important in characterizing the optical and electronic properties of atomically thin transition metal dichalcogenides (TMDCs). Among these two-dimensional (2D) materials, molybdenum disulfide (MoS₂) is considered a good candidate for applications in optoelectronics and photovoltaics. However, predicted device performance is significantly affected by localized defects such as grain boundaries and non-uniformity in large-area films, including regions with varying number of atomic layers. Improvement of techniques for spatially localized investigation of these materials will help accelerate device development from these materials.
In this work, we demonstrate an optical microscope setup capable of transmittance, photoluminescence (PL), and photocurrent spectroscopy with ~1 μm spatial resolution. This spatial-spectral mapping of 2D materials is shown for the visible and near-infrared part of the spectrum. By correlating each of these measurements over the same material and device region, we can draw connections between material properties and device performance. We do this by studying optoelectronic and photovoltaic (PV) devices fabricated on chemical vapor-deposition (CVD) grown MoS₂ samples. We concurrently measure the fundamental properties of various excitons, e.g. transmission at the A, B and C peaks, and identify spatial regions of peak photoresponsivity leading to efficient photocurrent generation in MoS₂ photodetectors.
Our light source is a supercontinuum laser (Fianium/NKT Photonics) with laser line tunable filter, facilitating wavelength dependent excitation and optical studies of our MoS₂. PL mapping of MoS₂ flakes and films within the device active area is carried out by exciting single and few-layer thick MoS₂ samples with a monochromatic 532 nm excitation; emitted PL is measured using an Ocean Optics QePro spectrometer. Transmittance measurements are carried out with Thorlabs FDS1010-CAL calibrated photodiode mounted 1mm beneath the sample on a transparent stage. Device photocurrent is measured using a Keithley 2450 sourcemeter.
While each technique individually yields insightful information on the uniformity and properties of the 2D material, the integrated setup enables spatial correlation of properties as they relate to the performance of 2D PV, photodetectors, and other optoelectronics. From correlated PL and transmission measurements we extract absorption and photogeneration information that enables mapping localized external radiative efficiency. Transmittance measurements probe wavelength dependent absorption and reflectance and their dependence on local defects, while correlated spatial mapping of absorption and photocurrent vs. wavelength reveals the effects on carrier collection and device performance resulting from material uniformity, contact choice, grain boundaries, surface treatments, and defects. Localized layering of materials is visualized through mapping PL intensity and percentage transmittance. We are using this correlative mapping of photonic properties to better understand, design, model, and prototype advanced optoelectronics from 2D materials.

In this talk, I will discuss our recent research progress in understanding the electronic, photonic and ferroelectric properties of emerging low-dimensional materials, and in developing them for sensing and memory applications. The first part of the talk will focus on discussing the basic properties of emerging 2D materials such as black phosphorus and our progress in developing the material for mid-infrared optoelectronics application. I will also discuss our work on utilizing 4-dimensional imaging techniques to study carrier dynamics in two-dimensional materials, including a study using the newly developed scanning ultrafast electron microscopy (SUEM) technique to image the photo-carrier transport in black phosphorus. In the second part of the talk, I will discuss our recent research on the ferroelectric monolayer materials for memory device applications. I will conclude with remarks on promising future research directions of low-dimensional material properties and devices, and how the emerging materials may benefit future generations of electronics and photonics technology in sensing and memory.

Two-dimensional layered materials have received much interest in recent years due to their ease of miniaturization by mechanical exfoliation along with strong light-matter interactions and tunable electronic properties. Quasi-one-dimensional (Quasi-1D) materials, however, have seen considerably less interest, but can express many of the same desirable properties as conventional layered materials, with an added dimension of anisotropy. A representative example of this Quasi-1D subclass is Titanium Trisulfide (TiS₃). We demonstrate similar ease of mechanical exfoliation of TiS₃ accompanied with theoretical calculations and show that these materials exfoliate into few-atomic-layer nanoribbons with very smooth edges. We emulated macroscopic exfoliation experiments on the nanoscale by applying a local shear force to TiS₃ crystals in different crystallographic directions using a tip of an atomic force microscopy (AFM) probe. In the AFM experiments, it was possible to slide the 2D TiS₃ layers relative to each other as well as to remove selected 1D chains from the layers. Further, their characterization by Raman spectroscopy shows a reliable, internally standardized shift of a few cm^-1 from monolayer to bulk demonstrating tunability typical of conventional layered materials. Devices made from this material show promising electronic properties with predicted mobilities in excess of 10,000 cm²V^-1s^-1 with low edge-scattering, as well as in optoelectronics with strong light-matter interactions, particularly along the crystallographic b-axis (anisotropic). Orientation-dependent properties of this nature show promise as filters and polarizers, while high surface areas associated with 2D materials hint towards gas-sensing and energy storage applications.

Epitaxial Xenes are an emerging class of two-dimensional crystals made of elements spanning group III to group VI of the periodic table (e.g. borophene, silicene, phosphorene, antimomene, tellurene etc) [1]. These materials are synthesized by epitaxy on various substrates and are subject to degradation in environmental conditions. Our work is centered around Xenes grown on noble metal surfaces that are supported by mica, as they can readily undergo delamination and transfer on other substrates for device related studies. In particular, we focus on silicene-on-silver [2] and epitaxial phosphorene [3] as representative cases to establish a stabilization and transfer scheme. Encapsulation of the unstable Xenes is a critical step for achieving transfer of the Xene layer from its pristine growth substrate to a device or functional substrate. We propose the use of an Al₂O₃encapsulation as a stabilization strategy that can be universally applied to the whole class of Xenes. The integrity of the Xene layer after encapsulation is validated by means of Raman spectroscopy and X-ray photoelectron spectroscopy. We then propose a few different transfer schemes and present experimental results of film transfer. A technical step in the delamination and transfer process is the time-controlled etching of the metal substrate. We achieve this using an optimized KI/I₂silver etchant with a controllable etch rate. Finally, we demonstrate the stability of air-exposed multilayer silicene, prepared using our developed methods, via Raman spectroscopy, validating our successive delamination and transfer protocol.

References:
[1] A. Molle et al., Nature Materials (2017) 16, 163.
[2] A. Molle et al., Chemical Society Reviews (2018) 47, 6370.
[3] C. Grazianetti et al., Nanoscale (2019) 11, 18232.

Transition metal dichalcogenides (TMDs) such as molybdenum disulfide (MoS₂) have gained much interest for scaled nanoelectronics, but their mobilities and drive currents must be improved in order to compete with existing technologies. Theoretical studies predict that tensile strain can increase the mobility of TMDs due to changes in band structure, leading to reduced intervalley scattering [1]. However, such mobility improvement has not yet been experimentally demonstrated.
In this work, we utilize uniaxial tensile strain to improve the electrical performance of monolayer (1L) MoS₂ transistors on flexible substrates. Using a two-point bending apparatus, we apply multiple strain levels between 0% and ~0.7%, which we estimate from the curvature radius of the bent substrates. We confirm these strain levels by Raman spectroscopy, using the known red-shift of the MoS₂ in-plane E’ Raman peak of -2 cm^-1 per percent of tensile strain [2].
We fabricate our devices on polyethylene naphthalate (PEN) flexible substrates. First, the gate metal is lithographically patterned, and then we form the gate dielectric with ~20 nm of Al₂O₃ by atomic layer deposition (ALD). Monolayer MoS₂ grown by chemical vapor deposition (CVD) on separate SiO₂/Si substrates [3] is then transferred to the flexible substrates using a polymer scaffold [4]. Finally, we pattern source and drain contacts onto the MoS₂ films, followed by an O₂ plasma etch to define the channels.
We characterize the effects of tensile strain on MoS₂ transistors with channel lengths from 1 to 15 μm, by extracting the field-effect mobility. Relatively long channels are used to limit extrinsic contributions of contact resistance. We obtain a continuous mobility improvement up to ~85% with increasing tensile strain up to ~0.7%, leading to a drain current I_D increase of ~2x at the same carrier density. Furthermore, we find that the effects are fully reversible as the device characteristics return to their initial state after strain release. The mobility increase with tensile strain also indicates that these devices could be used as strain sensors. An I_D change of ~2x at 0.7% strain corresponds to a gauge factor of nearly ~150, which is the highest value reported so far for piezoresistive 1L CVD MoS₂ strain sensors [5-6].
These results demonstrate the largest mobility and on-state current improvements to MoS₂ transistors using strain to date, revealing that strain engineering is a promising way to tune the electrical performance of TMD-based devices. Furthermore, using MoS₂ transistors as strain sensors can facilitate the realization of flexible and transparent sensor systems for strain mapping.
[1] M. Hosseini et al., J. Phys. D: Appl. Phys. 48, 375104 (2015)
[2] C. Rice et al., Phys. Rev. B 87, 081307 (2013)
[3] K.K.H. Smithe et al., 2D Materials 4, 011009 (2017)
[4] S. Vaziri et al., Sci. Adv. 5, eaax1325 (2019)
[5] Y. J. Park et al., ACS Nano 13, 3023 (2019)
[6] M. Park et al., Adv. Mater. 28, 2556 (2016)

Due to their unparalleled mechanical strength and high electronic mobility two dimensional (2D) materials represent the ultimate limit in size of both mechanical atomic membranes and molecular electronics. Bringing these capabilities together make 2D materials and heterostructures promising for emerging technologies like origami robotics, stretchable electronics and mechanically reconfigurable nanoelectromechanical systems. Additionally, many of the most interesting properties of 2D materials and new functionality arise from the interfaces between layers and in engineering multilayer heterostructures. Building an understanding of the mechanics of the van der Waals interface is critical to these next generation devices. In this presentation, we will examine the role of the van der Waals interface in determining the mechanics of deformation in atomic membranes under bending, stretching, and shear and the impact on the behavior of 2D electromechanical systems.
First, we use aberration-corrected scanning transmission electron microscopy to image multilayer graphene and 2D heterostructures draped over a series of atomically sharp hexagonal boron nitride steps and extract. Combining the measurements with atomistic simulations and energy conservation models we extract the bending stiffness over a range of thicknesses and step heights. We find an bending angle and interface orientation dependent bending stiffness. At high bending angles, the bending stiffness to scale linearly with the number of layers due to free slip between layers. At low angles, materials made from commensurate interfaces become stiffer, converging to the cubic scaling with thickness predicted from continuum mechanics. In contrast, heterostructures with an incommensurate interface maintain the linear scaling at all bending angles due to the low interfacial friction. By tailoring the ordering of layers within the heterostructure, we demonstrate tuning of the bending stiffness. These results provide a united description for bending in 2D materials which resolve several years of contradictory measurements and enable the design of bending stiffness in 2D heterostructures with mixed commensurate and incommensurate interfaces.
Next, we engineer electromechanical drumhead resonators from 2D membranes such as twisted bilayer graphene, commensurate bilayer graphene and graphene/MoS₂ bimorphs. In the commensurate bilayer graphene, we observe discrete jumps in frequency which do not appear in monolayer resonators. We find the magnitude of the frequency jumps follows a simple model for changes in stress due to the dynamics of solitons at the interface. Moreover, we find enhanced structural dissipation in twisted 2D interfaces due to the inelastic sliding between layers. Finally, we examine the evolution of buckle and fold instabilities in 2D membranes and heterostructures under compression, and examine the behavior of stretchable electronics made from crumpled 2D heterostructure devices.
Taken together, these experiments show that interfacial slip strongly impacts the mechanics of 2D materials and heterostructures and leads to membranes which are orders of magnitude more deformable than conventional 3D materials.

A major drawback in shrinking down mechanical components to the nanoscale is the excessive dissipation coming from friction at contacting surfaces. A physical regime where this could be substantially avoided is structural superlubricity, wherein an atomically flat and incommensurate van der Waals interface will lead to ultra-low coefficients of friction of <0.001, less than 1% of conventional interfaces. Superlubricity has been experimentally observed via scanning probe measurements of 2D material surfaces, as well as in static sliding of twisted bilayer graphene [1,2] and heterostructure interfaces [3,4]. Moreover, many theoretical studies have predicted and proposed superlubricity in a variety of systems. Yet, nearly three decades after being predicted, superlubric behavior has not been demonstrated in a nanoelectromechanical system, nor has the energy dissipation from superlubric friction ever been directly measured.
In this study, we directly probe the dissipation due to incommensurate interfaces by engineering and comparing mechanical drumhead resonators made from twisted (incommensurate) bilayer, Bernal-stacked (commensurate) bilayer, and monolayer graphene. Because of the ultra-low mass of graphene, the total energy in the resonators is orders of magnitude less than in conventional MEMS systems, making their response extraordinarily sensitive to changes in the energy dissipation. We find that the incommensurate bilayer graphene resonators have damping coefficients more than twice than that of the commensurate counterparts. The extra dissipation in the twisted interface is well described by a structural damping model common in macroscale vibrating components arising from small slips at mechanical joints [5,6]. In the twisted bilayer graphene, the damping arises from inelastic sliding between layers during vibration. From this structural damping model, we extract an upper limit for the interlayer friction stress to be 0.86 mN/m, in good agreement with the reported value obtained in static measurements of interlayer sliding [2]. Additionally, by measuring the mechanical quality factor versus temperature, we observe that friction decreases with decreasing temperature, down to a value < 0.1 mN/m, in agreement with theoretical predictions of a phonon limiting mechanism for superlubric friction. These results unveil the important role of the superlubric interface in mechanical dissipation and provides a principle for designing nanoelectromechanical systems with 2D heterostructures.


[1] M.Dienwiebel, et al. Superlubricity of graphite. Phys. Rev. Lett., 92:126101, 2004.
[2] E. Koren, et al. Adhesion and friction in mesoscopic graphite contacts. Science, 348(6235):679–683, 2015.
[3] S. Kawai, et al. Superlubricity of graphene nanoribbons on gold surfaces. Science, 351(6276):957–961, 2016.
[4] C.M. Harris and A.G. Piersol, Harris’ shock and vibration handbook, McGraw-Hill handbooks, McGraw-Hill, 2002.
[5] Yiming Song, et al. Robust microscale superlubricity in graphite/hexagonal boron nitride layered heterojunctions. Nature Materials, 17(10):894–899, 2018.
[6] W. Chen and X. Deng. Structural damping caused by micro-slip along frictional interfaces. International Journal of Mechanical Sciences, 47(8):1191 – 1211, 2005

Switching mechanism in transistors based on transformation of transition metal dichalcogenides (TMD) between semiconducting and semi-metallic phases may offer superior performance by eliminating static and dynamic power consumption problems associated with scaling of conventional field effect transistors. It has been shown recently that the transformation between semi-metallic TMD 1T’-MoTe₂ and semiconducting MoTe₂ using nanoscale strain engineering results in such switching behavior at device scale. Since strain is the key factor in controlling such electrical properties in these multilayer materials, we use a combination of modelling and experimental approaches to examine and quantify the efficiency of strain transfer in multilayer TMDs and how strain is tied to electrical properties. In this study, molecular dynamics (MD) models are used to predict properties of both monolayer and multilayer structured TMDs under different mechanical loading such as uniaxial, biaxial strain, and nanoindentation. Using the reactive empirical bond order (REBO) potential available in literature along with local structural identification methods such as polyhedral template matching, phase transformation behavior is studied under mechanical loading. Additionally, nanoindentation is used for experimental measurement of properties of different structure types of TMDs (2H and 1T’) for verifying, as well as developing, interlayer van der Waals interaction parameters for atomistic models. Results from the atomistic model coupled with experimental observations using optical microscopy and Raman spectroscopy reveals the mechanism of strain transfer at device scale and its efficiency in various setups.