Symposium NT4 : Emerging Non-Graphene 2D Materials

Electrons in two-dimensional crystals with a honeycomb lattice structure possess a valley degree of freedom (DOF) in addition to charge and spin. These systems are predicted to exhibit an anomalous Hall effect whose sign depends on the valley index. Here, we report the observation of this so-called valley Hall effect (VHE). Monolayer MoS₂ transistors are illuminated with circularly polarized light, which preferentially excites electrons into a specific valley, causing a finite anomalous Hall voltage whose sign is controlled by the helicity of the light. No anomalous Hall effect is observed in bilayer devices, which have crystal inversion symmetry. Our observation of the VHE opens up new possibilities for using the valley DOF as an information carrier in next-generation electronics and optoelectronics. Time permitting, I will also discuss the properties of other novel 2D materials under study in our laboratory.

Two-dimensional (2D) transition metal dichalcogenides (TMDCs) such as monolayer MoS₂ and WS₂ are promising for developing truly atomic-scale light emission devices. Despite perfect surface passivation and strong exciton binding energy, the luminescence efficiencies of substrate-supported monolayers, however, are grossly poor compared to their suspended counterparts. Here we demonstrate that the luminescence efficiency of these monolayers is strongly dependent on the type of substrates and can be improved by orders of magnitude through substrate engineering. The origins for the substrate effects mainly lie in doping the monolayers as well as affecting their exciton dynamics, including facilitating the non-radiative recombination and slowing down the radiative decay. Using proper substrates like mica and Teflon can minimize the adverse effect of doping, and their luminescence efficiency can be largely tuned at room temperatures. These results provide useful guidance for the rational design of high-performance 2D TMDC light emission devices.

Energy transfer between excitonic systems separated by a small barrier layer plays an important role in determining the exciton decay dynamics in such systems. The ability to manage the direction of energy transfer in an intelligently engineered structure is of fundamental interest in semiconductor optoelectrnic devices. A vertical stack of semiconducting 2D crystals is effectively a coupled quantum well system that should allow exciton energy transfer across a sub-nanometer van der Waals gap via near field interaction. Previous studies on hetero-bilayers of group 6 transition metal dichalcogenides (TMDs) suggest that the interlayer charge transfer processes dominate exciton decay dynamics over energy transfer. In this talk, I will discuss our experimental observation of fast interlayer energy transfer in MoSe₂/WS₂ hetero-bilayer using photoluminescence excitation (PLE) spectroscopy. We find that the energy transfer is Förster-type involving excitons in the WS₂ layer resonantly exciting free electron-hole pairs in the MoSe₂ layer. Our results indicate that energy transfer competes with interlayer charge transfer with efficiencies exceeding 80% despite the type-II band alignment.

Atomically-thin transition metal dichalcogenide (TMD) semiconductors have generated great interest recently due to their remarkable physical properties. In particular, reduced screening in 2D has been predicted to result in dramatically enhanced Coulomb interactions that should cause giant bandgap renormalization and excitonic effects in single-layer TMD semiconductors. Here we present direct experimental observation of extraordinarily high exciton binding energy and band structure renormalization in a single-layer of semiconducting TMD [1]. We have determined the binding energy of correlated electron-hole excitations in monolayer MoSe₂ grown via molecular beam epitaxy [2] on bilayer graphene (BLG) by using a combination of high-resolution scanning tunneling spectroscopy and photoluminescence spectroscopy. We have measured both the quasiparticle electronic bandgap and the optical transition energy of monolayer MoSe₂/BLG, thus enabling us to obtain an exciton binding energy of 0.55 eV for this system, a value that is orders of magnitude larger than what is seen in conventional 3D semiconductors. In a more highly screened environment (on top of graphite), we find that single layers of MoSe₂ show a strong 51% reduction in the exciton binding energy and an 11% reduction in the quasiparticle electronic gap, without significantly changing the optical gap. We have corroborated these experimental findings through ab-initio GW and Bethe-Salpeter equation calculations, which show that the large exciton binding energy arises from enhanced Coulomb interactions that lead to blue-shifting of the quasiparticle bandgap. We have also studied the role of interlayer coupling and layer-dependent carrier screening on the electronic structure [3] of few layer MoSe₂. We find that the electronic quasiparticle bandgap decreases by nearly 1 eV when going from one layer to three. Our findings paint a clear picture of the evolution of the electronic wave function hybridization in the valleys of both the valence and conduction bands as the number of layers is changed. These results are of fundamental importance for the design and evaluation of room-temperature electronic and optoelectronic nanodevices involving single layers of semiconducting TMDs as well as more complex layered heterostructures.

References:

[1] M. M. Ugeda, A. J. Bradley, et al., Nature Materials 13, 1091 (2014).
[2] Y. Zhang, M. M. Ugeda, et al., Submitted (2015).
[3] A. J. Bradley, M. M. Ugeda, et al., Nano Letters 15, 2594 (2015).

Two-dimensional (2D) semiconductors, such as transitional-metal-dichalcogenide (TMD) monolayers, have aroused great attention because of the underlying quasiparticle physics and the promising optoelectronic applications such as atomically-thin light-emitting diodes and photodetectors. Even though considerable progress has been achieved recently, intrinsic and extrinsic excitonic states of TMD monolayers are still under discussion and require direct experimental exploration. Here, we report observations of unconventional excitonic emission states of monolayer WS₂ and MoS₂ under electrical doping and photoexcitation at low temperature. Through electrical and optical injection of charge carriers, tunable excitonic emission has been realized due to interplay of various excitonic states, and the crucial binding energies of trions, biexcitons and bound excitons have been obtained. Particularly, the superlinear emission and the unusual linear polarization behaviour of its photon energy opposite to that of exciton emission have been clearly observed, which is assigned to the presence of biexciton states, evidencing the strong Coulomb interactions in such 2D semiconductors. Extrinsic emission bands in monolayer WS₂ and MoS₂ have also been revealed and attributed to specific localized states resulting from impurities and structural defects, where the exact origin of such states has been discussed according to our experimental observations and theoretical calculations. In view of applications, we have developed a highly practical strategy to electrically manipulate the diverse excitonic emission in atomically thin semiconductors, which opens up many possibilities for design and improvement of 2D TMD-based optoelectronic devices.

Energy transfer is a well-known phenomenon commonly observed in molecular complexes, biological systems, and semiconductor heterostructures. In van der Waals heterostructures of 2D semiconductors, Förster-type energy transfer (ET) is expected to be efficient due to the parallel orientation and sub-nanometer separation of excitons in the neighboring layers. Here, we report an experimental observation of fast interlayer ET in MoSe₂/WS₂ heterostructures using photoluminescence excitation spectroscopy. We find that the ET is Förster-type involving excitons in the WS₂ layer resonantly exciting higher-order excitons in the MoSe₂ layer. Our results indicate that ET competes with interlayer charge transfer with efficiency exceeding 60% despite the type-II band alignment. Efficient ET in these systems offers prospects for optical amplification and energy harvesting through intelligent layer engineering. We anticipate that the ability to manage energy via ET at the sub-nanometer length scales in carefully designed heterostructures will allow realization of novel optical and optoelectronic device concepts.

Atomically thin transition metal dichalcogenides (TMDCs) have attracted great interests as a new class of two-dimensional (2D) direct bandgap semiconducting materials. The controllable modulation of optical and electrical properties of TMDCs is of fundamental importance to enable a wide range of future optoelectronic devices. Here we investigate the modulation of optoelectronic properties of 2D TMDCs, including MoS2, MoSe2, and WSe2, by interfacing them with two metal-centered phthalocyanine (MPc) molecules, nickel Pc (NiPc) and magnesium Pc (MgPc). Through energetically favorable electron transfer from the photo-excited TMDCs to the MPc, the photoluminescence (PL) emission characteristics can be selectively and reversibly engineered. NiPc molecules, whose reduction potential is positioned below the conduction band minimum (CBM) of MoSe2 and WSe2 monolayers, but higher than that of MoS2, quench the PL signatures of MoSe2 and WSe2, but not MoS2. Similarly, MgPc quenches only WSe2 as its reduction potential is situated below the CBM of WSe2, but above that of MoS2 and MoSe2. The quenched PL emission can be fully recovered when MPc molecules are removed from the TMDC surfaces. We also find that photocurrents from TMDCs, probed by photoconductive atomic force microscopy (PC-AFM), are promoted when MPcs quench the PL, further supporting the photoinduced charge transfer mechanism. Our results will benefit design strategies for 2D inorganic-organic optoelectronic devices and systems with tunable properties and improved performances.

Two dimensional (2D) monolayer transition metal dichalcogenide (ML-TMDC) semiconductors are ideal building blocks for atomically thin, flexible optoelectronic and catalytic devices. Although the tightly bound exciton states lie at the heart of the optoelectronic properties of ML-TMDCs, surprising voids in the understanding of the manifold of these states persists. Here, we move beyond ground state excitonics, unifying somewhat contradictory observations by combining PLE spectroscopy, hyperspectral imaging, and DFT calculations of ML-MoS2, revealing anomalous optoelectronic behavior. Signatures of the 1S and 2S states of the A and B excitons are resolved and a reduction in PL quantum yield with increasing excitation energy is observed, confirming previous PLE studies. In contrast to the previous works however, we also observe subtle features in the PLE spectrum that are assigned to the quasiparticle bandgap and resulting onset of interband transitions, which is supported by complimentary density functional theory (DFT) calculations. Additionally, an “anomalous” increase in the experimental PLE spectrum at energies >2.6 eV diverges from previous results and expectations for a 2D quantum well with substantial many-body effects. This feature corresponds nicely to the step-like feature observed in the previous PCE measurements and likely arises from efficient coupling of excitations created in the band-nesting region of the Brillouin zone to the photoemissive direct gap excitons at the K-point. We find that these high-energy excitations can enhance emission from lower-energy states, such as trions or biexcitons. From hyperspectral PLE imaging of the single flakes of ML-MoS2, we find that the Stokes shift between excitation and emission resonances, a common metric for disorder in conventional 2D quantum wells, is significantly more uniform than the exciton energies, implying internal relaxation processes remain substantially more ordered than the exciton energy. Finally, the spatially resolved PLE spectroscopy reveals that spatial dispersion of the anomalous excitation resonances is correlated to the A and B exciton states, suggesting that these high-energy states are similarly excitonic in nature.

Black phosphorous (BP) is known as a most stable allotrope of phosphorous with its unique orthorhombic structure. Recent studies have shown that BP is an elemental layered material with strong in-plane bonds and a weak van der Waals interlayer interaction that make it possible to obtain layered structures through mechanical cleavage. The energy band gap of BP is reported to be 0.33 eV in the bulk and around 1.8 eV for a monolayer, and unlike transition metal dichalcogenides (TMD) such as MoS₂ and WSe₂, the energy band gap of BP retains its direct band gap character when the layer thickness varies. Moreover, the carrier mobility of BP can theoretically be as high as 20000 cm²/Vs. These fascinating properties mean that layer-structured BP is a promising semiconducting channel material for future nanoscale electronic devices. However, there are several technological obstacles that must be overcome to enable the actual implementation and integration of BP devices. The mechanical cleavage with tape of BP from the bulk is not as easily realized as for graphene and TMD materials making it difficult to control the layer thickness of the exfoliated film. Furthermore BP exfoliated from the bulk is not environmentally stable because it reacts with oxygen and water molecules and forms a rough surface containing impurities which act as charge trapping centers and carrier mobility scattering centers. As a result, considerable degradation of transistor performances, including decreases in mobility and threshold voltage shifts have been observed.
In an attempt to resolve these two issues, we tested in this study a plasma etching process in which the thickness of the layered BP film can be controlled by modulating the plasma etching time at the thinning rate of 2 layers /min. Not only does the plasma treatment control the thickness of the BP film, it also removes the chemical degradation of the exposed oxidized BP surface and that the remaining few-layer BP films exhibit excellent morphology and crystalline structure, which results in enhanced field-effect transistor (FET) performance. Our fabricated BP FETs were passivated with polymethyl methacrylate (PMMA) immediately after the plasma etching process to protect the prepared BP films from water and oxygen molecules in air. With these techniques, a high field-effect mobility was achieved, 1150 cm²/Vs, with an I_on/I_off ratio of ~10⁵ at room temperature. Furthermore, a fabricated FET with plasma-treated few-layer BP that was passivated with PMMA was found to retain its I-V characteristics and thus to exhibit excellent environmental stability over several weeks.

Gallium chalcogenides are new class of 2D materials with the chemical formula GaX where X are chalcogen atoms. In the bulk form, GaSe and GaTe are direct bandgap semiconductors displaying strong photoluminescence (PL) but become indirect gap semiconductors in the monolayer form. The direct to indirect transition happens because the valence band renormalizes itself to a “W” shape around the Γ point so that the valence band maximum (VBM) shifts from the Γ point. However, our recent experiments show that for few-layered GaX, the PL comes back at low temperatures and increases by many orders of magnitude as down to 4 K. This finding indicates the 2D GaX can still behave like a direct bandgap semiconductor at low temperatures due to dramatically weakened exciton-phonon interactions. To further study the exciton complexes in these materials, power dependent PL is measured at 5 K. The result shows that PL intensity is well fitted according to I=AP^α, where α varies in terms of different excitonic processes. Interestingly, in 2D GaX α is found to be different from its bulk form as well as 2D MoS₂, indicating a novel exciton recombination process. Based on the abnormal thermal quenching and power dependence of the PL, a three particle recombination process involving one electron and two holes is proposed for 2D GaX.

By combining electrical measurements, scanning Kelvin probe microscopy, and numerical electrical simulations, we find significant current crowding effect on two dimensional black phosphorus field-effect transistors. This current crowding can lead to localized Joule heating close to the metal contacts, and it is consistent with the features of the device failure observed in this study. Importantly, we find that the commonly used transmission-line model, in general, significantly underestimates the extent of the current crowding by 1 – 2 orders of magnitude, compared to our results. In addition, our approach allows us to determine the black phosphorus layer resistance and the specific contact resistance, an important parameter for evaluating electrical contacts, whereas the simple relation for calculating the specific contact resistance is invalid due to the current crowding. We find that the layer resistance increases significantly when the layer thickness decreases towards about 5 – 6 nm and remains relatively constant for thicker layers. It turns out that the specific contact resistance of Titanium contacts on black phosphorus field-effect transistors is about 0.3 Ωcm² in average when the layer thickness is larger than 7 nm, and significantly increases when the layer thickness is below 7 nm. The specific contact resistance also increases when the back gate voltage sweeps from -9 V to 9 V. These findings, which are likely to be relevant in other 2D materials, suggest the need to take into account the current crowding effect in designing novel 2D devices.

Two-dimensional materials are subject to intrinsic and dynamic rippling that modulates their optoelectronic and electromechanical properties. Here, we directly visualize the dynamics of these processes within monolayer transition metal dichalcogenide MoS₂ using femtosecond electron scattering techniques as a real-time probe with atomic-scale resolution. We show that optical excitation induces large-amplitude in-plane displacements and ultrafast wrinkling of the monolayer on nanometer length-scales, developing on picosecond time-scales. These deformations are associated with several percent peak strains that are fully reversible over tens of millions of cycles. Direct measurements of electron-phonon coupling times and the subsequent interfacial thermal heat flow between the monolayer and substrate are also obtained. These measurements, coupled with first principles modeling, provide a new understanding of the dynamic structural processes that underlie the functionality of two-dimensional materials and open up new opportunities for ultrafast strain engineering using all-optical methods. The pump-probe techniques described here also allow for studies of small, thermally-induced structural changes under quasi-equilibrium conditions, difficult to resolve with static heating approaches, including direct measurements of thermal boundary resistances across buried interfaces.

Atom-thick 2D materials such as Graphene and monolayer Molybdenum Disulphide (MoS₂) make very sensitive biosensors and detectors of gaseous organic analytes. Every atom on the surface is exposed, which maximizes the material’s sensitivity to molecules in the external environment. Through an efficient reproducible growth process that yields a high density of monolayer single crystal MoS₂ flakes, we are able to manufacture thousands of FET devices with a yield of 95% and average mobility of 36cm²V^-1s^-1 at room temperature in vacuum. With this method we were able to successfully perform bio and vapor sensing on our samples.
For biosensing, atomic length nickel-mediated linker chemistry is used to enable target binding events close to the MoS₂ surface to maximize sensitivity. A computationally redesigned μ-opioid receptor (MOR) is linked to the surface of the MoS₂, from which we were able to read out the concentrations of a synthetic opioid peptide with high MOR specificity. The data indicates binding affinities that match the values determined using traditional techniques.
For vapor sensing, we exposed an array of MoS₂ FETs to various analytes. Huge responses, up to 94% increase in the electrical conductivity, were observed upon exposure to 33vol% saturated dimethyl methylphosphonate (DMMP) vapor. These large responses were caused by the analyte molecules interacting with either the MoS₂ flake or the contacts between gold electrodes and the MoS₂ flake. Using Transmission Line Measurements (TLM), we studied the resistance of the MoS₂ flake and the resistance of the gold-MoS₂ contact as we exposed the sample to different vapor concentrations. This study strives to determine whether the responses to the analytes are driven by the MoS₂ flake or gold-MoS₂ contacts.
The combination of scalable array fabrication and rapid, precise readout enabled by the MoS₂ transistor offers the prospect of a solid-state drug testing platform for rapid readout of the interactions between novel drugs and their intended protein targets. Additionally, the large responses exhibited by MoS₂ devices upon exposure to various vapor analytes indicate a highly sensitive surface, which could potentially enable detection of small traces of dangerous compounds in the environment.

Two-dimensional (2D) transition metal dichalcogenides (TMDCs) have generated significant research excitement due to their unique thickness-dependent electronic and optical properties, particularly their semiconducting direct bandgap that leads to photoluminescence, making them attractive for applications in next-generation electronics, optics, and chemical and biological sensing. Chemical functionalization is crucial for altering and tuning the properties of nanostructured materials such as graphene and carbon nanotubes in these types of applications, but is in its relative infancy for the TMDCs. In this work, we demonstrate robust biofunctionalization of pristine MoS₂ monolayers and bilayers with the fluorescent proteins GFP and mCherry. The fluorescent properties of both proteins are maintained and are simultaneously localized to the MoS₂ surface as shown by confocal fluorescence microscopy. Atomic force microscopy shows a uniform coverage of protein molecules, and Raman and PL mapping show that MoS₂ also retains its optical properties. The biofunctionalization process here relies on Ni-NTA tethering of polyhistidine-tagged proteins, as well as direct covalent functionalization of the pristine MoS₂ basal plane via aryl diazonium salts, which we also show for the first time. Previous methods of covalent modification of MoS₂ require harsh treatments such as lithium intercalation, defect engineering, or phase engineering, which disrupt many of the useful semiconducting and photoluminescent properties of the TMDCs. Instead, we demonstrate that diazonium functionalization of the pristine MoS₂ surface maintains and even enhances the MoS₂ photoluminescence (PL) two-fold. We confirm C–S bond formation by x-ray photoelectron spectroscopy (XPS) and we conduct detailed spatial mapping of the optical emission and surface coverage of reacted sites of the MoS₂ at time points ranging from 2 s to 6 h to derive a kinetic model of the covalent functionalization, which we find can be modeled by surface-limited second order reaction kinetics.

Atomically thin transition metal dichalcogenides (TMDCs) are two-dimension (2D) materials. Their unique electrical and optical properties make them an ideal platform for both fundamental studies as well as practical applications. In many cases, the ability to manipulate the optical properties of TMDCs via external modulation is highly desirable.
We have studied the photoluminescence (PL) response of both monolayer and bilayer tungsten disulfide (WS₂) under lateral electric field.¹ We demonstrate that monolayer and bilayer WS₂ responded oppositely towards the lateral electric field. PL of monolayer WS₂ can be substantially quenched under lateral electric fields while the emission of bilayer WS₂ was enhanced. In both cases, the spectral shifts were negligible. Temperature dependent measurements showed distinctly different behavior from that of the electric field, indicating that joule heating was not the dominant mechanism. Direct ionization and impact ionization were ruled out by comparing the theoretical threshold electric field strength with the experimental value. Considering the multi-valley band structure of tungsten disulfide, we attribute the field induced PL variations to photo-excited electron transfer from one conduction band extremum to another, resulting in the modification of recombination pathways. Such a physical process can only be observed in 2D TMDCs due to their exceptionally large exciton binding energy and the small energy difference between two conduction band extrema.
Laser treatment can also tune the emission spectrum of monolayer tungsten disulfide. We show that strong laser irradiation (>30 kW/cm²) can introduce additional luminescent features at 77 K. Before laser irradiation, the PL spectrum of monolayer WS₂ under low excitation power (~3 W/cm²)was dominated by exciton, trion and localized states (LS). However, after laser treatment, the emission intensity of LS was quenched substantially while a new PL peak with emission energy between LS and trion emerged under low excitation power (~3 W/cm²). However, this behavior can only be observed in monolayer WS₂ aged in ambient conditions, and thus is associated with defects. We found that the resultant PL shape evolved with laser irradiation time. It was a reversible process where PL spectrum can reverse back to original shape if thermal cycle was applied.
[1] Z. He, Y. Sheng, Y. Rong, G. Lee, J. Li, J. Warner. ACS Nano 2015, 9, 2740-2748.

Monolayer WS₂ is a direct band gap semiconductor in the visible spectral region and as part of the monolayer transition metal dichalcogenides (TMDs) family it is opening up new avenues for electronic and optoelectronic devices. Creating alternating layers of two-dimensional (2D) materials forms vertical heterostructures with diverse electronic and optoelectronic properties that could revolutionize future ultrathin devices. These 2D heterostructures can be realized by either transfer or direct growth.

One important aspect for atomically thin TMDs is its flexibility, which is better suited for strain engineering compared with other bulk semiconductors. Monolayer WS₂ grown by chemical vapor deposition (CVD) can have inbuilt strain due to interactions with the substrate. The strain modifies the band structure and properties of monolayer WS₂ and can be exploited in a wide range of applications. It has been predicted that tunable photonic devices and solar cells with better energy band matching can be realized by building multilayer 2D heterostructures based on stained TMDs. However, building layer-by-layer vertical heterostructures containing a strained TMD layer has been challenging due to the standard aqueous based transfer processes washing off the TMDs from the growth substrate. It is important to develop a non-aqueous transfer approach that allows other 2D crystals to be assembled on top of the strained TMDs to build multi-layered vertical heterostructures.

We demonstrate a non-aqueous transfer method for creating vertical stacks of mixed 2D layers containing a strained monolayer of WS₂, with Boron Nitride and Graphene. The 2D materials are all grown by CVD enabling large area vertical heterostructures to be formed. WS₂ monolayers grown by CVD directly on Si substrates with SiO₂ surface are easily washed off by water and this makes aqueous based transfer methods challenging for creating vertical stacks on the growth substrate. 2D hexagonal Boron Nitride films are used to provide an insulating layer that limits interactions with a top graphene layer and preserve the strong photoluminescence from the WS₂. We show that stronger interactions between graphene and monolayer WS₂ occur when using non-aqueous based transfer, associated with sharper atomic interfaces from less contamination. This transfer method is suitable for layer-by layer control of 2D material vertical stacks and is shown to be possible for all CVD grown samples, which opens up pathways for the rapid large scale fabrication of vertical heterostructure systems with atomic thickness depth control and large area coverage.

Hexagonal boron nitride (hBN) is a dielectric insulator with a two-dimensional layered structure, having a wide band gap energy of 5.6–6.0 eV. Due to its novel properties, which include a high mechanical strength, electric resistivity, thermal conductivity, and chemical inertness, as well as flexibility and transparency, hBN can provide many advantages to next-generation electronics and optoelectronics such as flexible and transparent transistors or transferrable light-emitting devices. While hBN layers can be prepared by mechanical cleavage methods from bulk crystals, their limited sizes and random thicknesses and orientations may not be suitable for many device applications. On the other hand, large-area hBN layers can be prepared by chemical vapor deposition (CVD); however, they usually exhibit polycrystalline structures with a typical average grain size of several microns. Grain boundaries or dislocations in hBN can affect its electronic or mechanical properties; several theoretical calculations have discussed the lower band gap energy and poorer mechanical strength of polycrystalline hBN. Accordingly, large-area single-crystal hBN layers are required to fully realize the potential merits of hBN. Here, we report the synthesis of single-crystal hBN (SC-hBN) layers by CVD.
The SC-hBN was grown on a 20 × 20 mm² Ni (111) substrate from a single ammonia-borane precursor by atmospheric pressure chemical vapor deposition (APCVD). A grown layer was transferred to arbitrary substrates by an electrochemical delamination technique, and the remaining Ni (111) substrate was repeatedly re-used. Raman spectroscopy, X-ray photoelectron spectroscopy (XPS) and UV-Vis-IR spectroscopy confirmed that the grown films exhibited the characteristics of hexagonal boron nitride layers. Microstructural analysis by transmission electron microscopy (TEM) revealed that the grown layers were single-crystalline over their entire areas, presumably due to heteroepitaxial growth on the Ni (111) substrate. Atomic force microscopy (AFM) and electron backscatter diffraction (EBSD) studies on the Ni (111) substrate proved that there was no degradation after repeated growth and transfer cycles. The effect of growth parameters on the crystal structure of the grown hBN layers were investigated using reflection high-energy electron diffraction (RHEED).

The emerging two-dimensional (2D) layered materials are promising candidates for high-performance and energy efficient, flexible transparent electronics. Graphene is widely studied for transistors, sensors, and optoelectronic applications but is limited because of its lack of band gap. Beyond graphene, transition metal dichalcogenides (TMDCs) such as MoS₂ and WSe₂ are attractive because alloying, thickness, or strain engineering can be used to tune their band gap. Here, piezoresistive strain sensing with transparent and flexible atomically thin molybdenum disulfide (MoS₂) field effect transistors (FETs) made from highly uniform large-area synthesized films is demonstrated. The change in the MoS₂ current-voltage characteristics with mechanical strain is observed to be consistent with a change in band gap as measured using optical reflection spectroscopy. The magnitude of the band gap change agrees with previous theoretical and spectroscopic studies. In addition, the gauge factor or the sensitivity of the MoS₂ strain sensor is found to be adjustable via gate biasing, which changes the Fermi level of the MoS₂. The gate-tunable piezoresistivity is useful to adjust the sensitivity in practical sensing applications. The results show that MoS₂ can lead to a new generation of transparent and flexible strain and force sensors.

Transition metal dichalcogenides such as MoSe₂ have attracted great attention due to their desirable properties for electronic and optoelectronic applications. MoSe₂ field-effect transistors (FETs) showed promising device characteristics such as high on/off ratio (10⁶) and field-effect mobility (50-200 cm²/Vs). However, strategies to optimize device performance have not been established yet. In this presentation, we explore the possibility of improving MoSe₂ FET performance using atomic-layer-deposited Al₂O₃ passivation layers on MoSe₂ channels. By comparing the effect of Al₂O₃ layers on the device characteristics of bottom-gate multilayer MoSe₂ FETs, we observe the consistent enhancement of field-effect mobility and current-voltage hysteresis. These results suggest that the deposition of Al₂O₃ passivation layers on top of MoSe₂ channel can be an efficient method of optimizing the device performance of MoSe₂ FETs.

The fabrication of a top-gated transition metal dichalcogenide (TMD) FET requires a uniform and pinhole free gate dielectric. However the direct deposition of gate dielectrics on TMD materials is challenging due to the absence of surface dangling bonds. To deposit dielectrics with low leakage and low equivalent oxide thickness (EOT) on WSe₂, a flat-lying titanyl phthalocyanine (TiOPc) monolayer, deposited via molecular beam epitaxy (MBE), was used as a seed layer for atomic layer deposition (ALD) on the WSe₂ surface. ALD pulses of trimethyl aluminum (TMA) and H₂O resulted in the uniform deposition of Al₂O₃ on the TiOPc/WSe₂, confirmed by AFM and cross-sectional TEM. After depositing 50 cycles of Al₂O₃ onto the TiOPc/multilayer WSe₂, a top-gated FET was fabricated. The top gate leakage current was measured as 0.046 pA/µm2 at 1 V gate bias with 2.2 nm EOT, which is lower leakage current than all previous results on TMD with similar EOT. A subthreshold swing of 80 mV/dec was measured. While the TiOPc/WSe₂ was deposited by MBE, other phthalocyanines can be readily deposited from solution, making the functionalization technique more accessible. For example, to make a Cobalt Crown ether Phthalocyanine (CoCrPc) monolayer on WSe₂, 3 uM of CoCrPc/benzene-ethanol (9:1 v/v) solution was prepared. Two droplets of the solution were deposited onto a freshly cleaved, bulk WSe₂ substrate. After the solvent was completely evaporated, the sample was annealed at 200 C for 30 min in a N₂ filled glove box. An Al₂O₃ layer was deposited by 50 cycles of ALD using TMA and water at 120 C. The surface of the Al₂O₃ film was analyzed with ambient Atomic Force Microscopy (AFM), showing that the film is dense and uniform and has no pinholes detectable by AFM.

Two-dimensional (2D) materials have attracted a great attention since the rise of graphene. Recently, black phosphorus (BP), its monolayer known as phosphorene, has been highlighted as a promising 2D material that has properties comparable to that of graphene and other 2D materials. Beside its excellent electronic properties, BP has an advantage over graphene in the presence of direct bandgap. As each layer of phosphorene is held to each other by van der Waals interaction, BP flakes of high crystallinity can be easily transferred using mechanical exfoliation method. However, preparing BP film with desired thickness by mechanical exfoliation is still an issue. Fine control of BP thickness is essential for the fabrication of BP-based electronic devices because the characteristics of BP, such as bandgap, is highly sensitive to its thickness. In our experiments, we employed UV-ozone treatment on exfoliated BP flake to control its thickness by etching each BP layer. BP flakes were prepared on SiO₂/Si substrates using mechanical exfoliation method. The in situ optical characteristics of BP flakes were observed with Raman spectroscopy as UV-illumination time increased. Likewise, the thickness-dependent electrical properties of BP-based field-effect transistors were studied in situ. We investigated the changes in various electrical properties, including mobility and on/off ratio, as the BP thickness varied. The obtained results would be reproducible and controllable using our suggested method. Further results and discussion will be presented in detail.

Composites offer a facile means to tailor the properties of hybrid, dissimilar material systems for applications ranging from optoelectronics to strain sensors. Two-dimensional (2D) layered materials have attracted significant interest in recent years with the isolation of graphene through the mechanical exfoliation of graphite more than a decade ago. The 2D materials family is also rich with diverse compositions that offer semiconducting characteristics. In this work, we have explored the prospects of MoS₂ and WS₂, both of which are semiconducting 2D materials, for potential composite applications. In order to form 2D materials composites we have to first disperse them chemically in solution. MoS₂ and WS₂ powders were oversaturated in N-Methyl-2-pyrrolidone (NMP) solution at 37.5 mg/mL and sonicated at room temperature (RT) for sonication times ranging from 30 minutes to close to 24 hours. After solution processing, the samples with the 2D flakes were transferred to an Isopropyl Alcohol (IPA) bath for particle size distribution analysis.
We have observed significant changes in particle size distribution spanning two orders of magnitude as a function of the sonication conditions. Specifically, the observed changes in particle size distribution for MoS₂ and WS₂ powders ranged from 44 microns down to 0.409 microns, and 148 microns down to 0.409, respectively, as compared to the untreated materials. Structural analysis was conducted using the SEM and X-Ray diffraction. The structural analysis using the SEM revealed morphological signatures between the two materials, where the MoS₂ flakes had a randomly oriented distribution with occasional triangular flakes. In the case of the WS_2, regardless of the sonication conditions, the flakes seemed to have a characteristic 120° angular distribution at the vertices, representing a rhombus with concave edges. The XRD analysis showed a minute shift in the characteristic peaks that maybe due to strain-induced effects as a result of the solution processing. Optical characterization of the materials was also conducted using Raman Spectroscopy to validate the average layer number resulting from the solution dispersions and the spatial and compositional uniformity of the two material samples. After the material characterization studies on the parent MoS₂ and WS₂ were conducted, we integrated these materials with a polymer matrix to form a hybrid composite system. The optical, mechanical and electrical properties of the 2D-polymer composites were investigated for their potential utility in optoelectronic and strain sensing devices.

Surface migration of foreign atoms is expected to be promoted for 2D materials, which are virtually all-surface materials. An accompanying issue is chemical reactivity of these migrating atoms, which can dramatically alter the atomic structure and properties of 2D materials. Unfortunately, very little is known about these phenomena and the role of temperature and current density. In this study, we examine freestanding WSe₂-graphene heterostructure as well as multi-layer graphene for the effect of heat and charge flow by performing insitu experiments inside the transmission electron microscope. We observe selenium decompose and then silicon atoms migrate from the silicon micromachined test structure to the freestanding specimen. At high temperatures, this initiate rapid degradation through the formation of tungsten disilicide and silicon carbide formation. The role of current flow is to enhance the migration of silicon from the sample holder and to knock-out the selenium atoms. Experimental results show that purely thermal loading requires about 150 C higher temperature than that under combined electrical and thermal loading. We also present evidence of moderate current density, when accompanied with high temperature, promoting migration of foreign atoms on the surface of multi-layer graphene. Our in situ transmission electron microscope experiments show migration of silicon atoms at temperatures above 800 C and current density around 4.2x10⁷ A/cm². The silicon atoms are observed to react with the carbon atoms in the multi-layer graphene to produce silicon carbide at 900-1000 C temperature. In absence of electrical current, there is no migration of silicon and only pyrolysis

In order to realize the potential applications of layered TMDs, TMD materials have to be exposed to ambient air for device fabrication and characterization. Therefore, it is critical to understand the effect of air exposure on layered TMD materials. The effect of air exposure on MBE 2H-WSe₂/HOPG was elucidated on the atomic scale via scanning tunneling microscopy and spectroscopy. The 2H-WSe₂ was grown by molecular beam epitaxy (MBE) on HOPG. The WSe₂ nucleation is initiated at both the step edges and the terraces of HOPG, consistent with mixing of step flow growth mode and Volmer-Weber growth mode. High resolution STM images revealed nearly zero vacancy defects or dislocations in WSe₂ and the band gap of a WSe₂ monolayer was determined to be 2.18 ± 0.03 eV by STS. After exposure of air to WSe₂/HOPG, only the edges of the WSe₂ layer were oxidized, while the terraces of the WSe₂ layer were nearly intact. ; Iin the air exposed WSe₂ terraces, zero air induced defects were observed and the band gap was nearly identical to that of as-decapped WSe₂ ML (2.07 ± 0.05 eV by STS). Conversely, exposure of air to WSe₂/HOPG induced topological transitions from smooth edges of WSe₂ to bright protrusions along the edges of WSe₂ and a significant decrease in tunneling current at the edges of WSe₂ ML compared to the edge of as-decapped WSe₂ ML. It is hypothesized that the oxidation of the WSe₂ edges originates from the dangling bonds on the atoms at the WSe₂ edges. ; Iin high resolution STM images of as-decapped WSe₂, a bright brim was observed along the WSe₂ edge, originating from excess charge states of dangling bonds at the WSe₂ edge. In the STS of the as-decapped WSe₂ edge, a very narrow band gap was observed, as well as a shifted Fermi level towards the valence bands. However, this electronic edge state is passivated by the oxidation of edges in ambient air. The data is consistent with oxidation passivating defect states at the edges of WSe₂ which will be beneficial for device fabricating.Consequently, the present study can serve as a milestone to develop processes for device fabrication based on TMD materials, and elucidate the effects of the presence of electronic edges states on device performance.

One of the key challenges in the controlled chemical vapor deposition (CVD) of two-dimensional (2D) tungsten dichalcogenide crystals is achieving a steady flow of tungsten source in the vapor phase. Due to the high sublimation point of tungsten oxide precursors, CVD of tungsten dichalcogenides can be achieved only at high temperatures (>1000 ^oC) or at low pressures. We demonstrate atmospheric pressure CVD of WSe₂ and WS₂ monolayers at moderate temperatures (700 ~ 850 ^oC) using alkali metal halides (MX where M= Na or K and X=Cl, Br or I) as the growth promoters. Chemical analysis of the precursor after a growth run reveals that alkali metal halides react with tungsten oxide at elevated temperatures to form volatile tungsten oxyhalide species during growth, which leads to efficient delivery of the precursor to the growth substrates. The monolayer crystals obtained in this method were found to be highly symmetric, large in size (up to 200 μm), and free of unintentional doping due to alkali metal and halogen atoms. The monolayer crystals also exhibited excellent field-effect transistor (FET) performances with tunable polarity, large current on/off ratios (~10⁷), and high field-effect mobilities (electron and hole mobilities of 102 and 26 cm² V^-1 s^-1 for WSe₂ and electron mobility of ~14 cm² V^-1 s^-1 for WS₂ devices), further indicating their high quality.

Reference：Li, S. et al., Applied Materials Today 1 (2015) 60–66

Two-dimensional transition metal dichalcogenides (TMDC) have recently been under intense investigation due to their direct bandgaps and potential uses in high speed electronic devices. Despite being the most thoroughly studied TMDC, the mechanisms dominating the carrier dynamics in monolayer MoS₂ remain poorly understood. Much work to date has been performed on multilayer films, which have an indirect bandgap and reduced Coulombic interactions as compared to monolayer MoS₂. Other studies have focused on monolayer exfoliated flakes, which are not scalable for industrial applications.

In this work, we present transient optical absorption and time-resolved terahertz spectroscopy (TRTS) measurements of large area (~ 1 cm²) monolayer films of MoS₂ deposited on dielectric substrates via chemical vapor deposition. We report excited state dynamics that are surprisingly similar to those previously reported for multilayer MoS₂ in that they can be described by two decay components: one operative on sub-picosecond times scales and the other in 10's of picoseconds. Combining these two techniques allow for the contributions to the two-component excited state dynamics to be identified.

Terahertz conductivity measurements reveal that charge carriers, and not excitons, are responsible for the sub-picosecond dynamics. The frequency-dependent dielectric function is characteristic of localized charges in a polycrystalline material with sub-micron grains. Sub-picosecond lifetime photo-induced conductivity suggests that charges are rapidly trapped. This process complicates the excited state spectrum, which shows changes in line-width and spectral shifts associated with bandgap renormalization in addition to simple state-filling effects. We also report conductivity originating from sub-optical gap excitation, which reveals the existence of mid-gap trap states that may originate from surface defects or grain boundaries. These dynamics are largely insensitive to excitation density, wavelength, temperature and choice of substrate. We assign the slow excited state dynamics to the interband exciton lifetime, which lengthens from 50ps in monolayer films to 150ps in multilayer films due to the transition from direct to indirect bandgap. We see no signature from trions, suggesting that surface adsorbates neutralize the intrinsic n-type behavior.

Our results help add clarity to several conflicting reports of the optoelectronic properties of MoS₂, providing explanations for the sub-picosecond excited state dynamics as well as the absence of trions. Our findings suggest that trapping dominates the transient optical properties of monolayer MoS₂. They also suggest that CVD grown films hold potential for high-speed optoelectronic applications.

Black phosphorus (BP) is the most thermodynamically stable allotrope of phosphorus with several unique properties that has recently been exfoliated to atomically thin flakes. On the downside, BP atomic layers are observed to irreversibly degrade into phosphorus oxoacids upon exposure to humid air. However, in more robust configurations such as films, composites, and embedded structures, BP can potentially be utilized in many applications. In this study, we fabricated films of stacked black phosphorus nanoflakes made by liquid exfoliation and explored their chemical sensing characteristics upon exposure to different families of chemicals such as ketons, benzenes, alcohols, and water vapor. Interestingly, an ultra-sensitive and selective response was observed toward water vapor which provided for a trace-level moisture detection capability. The drain current of the sensors is observed to modulate by nearly 4 orders of magnitude when the humidity level changes from 10% to 85%. Evidenced by our spectroscopic and control experiments, we propose that the operation principle of the BP film sensors is based on a change in the leakage ionic current caused by auto-ionization of water molecules and ionic solvation of the phosphorus oxoacids produced on moist BP surfaces. Concurrently, our stability tests reveal that the response of the BP film sensors remain quite stable after prolonged exposures to ambient conditions (tested up to 3 months). This study opens up the route for utilizing films of BP stacked flakes in many potential sensing applications as well as energy conversion devices.

After discovery of graphene, there have been increasing interest in other two-dimensional (2D) materials that possess unique electrical and optical properties. The characteristics of 2D materials include both layered structures and heterostructures of transition metal dichalcogenides, silicone, germanene, and black phosphorous (BP). Among them BP is considered to be an interesting material due to its intrinsic band gap nature, strong in-plane anisotropy, and high electron mobility. These properties make BP as a promising material for applications in electronics, optoelectronics and clean energy. However, there are two main issue with BP, i.e., method of synthesis and degradation under moisture. In the present study, we report the facile synthesis method of few layer BP flakes by treating red phosphorous and further utilized BP flakes to make hetero-structure on single layer graphene. The structural and chemical properties of BP-graphene hetero-structure were evaluated using Raman spectroscopy. The electrical performance of heterojunctions will be discussed.

The structural, electronic and magnetic properties of MoS₂ single layer doped with 4f rare earth metal (RE) atoms Sm, Eu, Gd, Tb and Dy in the form of Mo₁₁S₂₄RE supercells were investigated using Density Functional Theory implemented within ADF-BAND. Our results with Pedrew Burke Ernzerhof (PBE) exchange correlational functional used within generalized gradient approximation (GGA) showed that variation of bond length is maximum around impurity atom in 2D Eu:MoS₂ and they are minimum near Sm impurity. The results indicated the appearance of impurity levels in the band gap of MoS₂ that lead to band gap narrowing. Analyzing the difference of spin charge densities, the magnetic moment of 3.3 µ_B, 8.1 µ_B, 8.5 µ_B, 6.8 µ_B and 6.4 µ_B was seen for Sm-, Eu-, Gd-, Tb- and Dy-doped structures, respectively. The findings of the work point out that the band gap and magnetic properties of 2D MoS₂ structures can be tuned by RE 4f doping to make it p-type degenerate semiconductor for device grade applications.

Graphene being a zero band-gap semimetal [1] intrinsically possesses very high mobility [1,2] but with very low on/off ratio [2], whereas two-dimensional MoS₂ being a direct band-gap semiconductor of around 1.9 eV [3] has low mobility [4,5], but can reach much higher on/off ratio [5]. Therefore, they are expected to complement each other in graphene/MoS₂ heterostructures with high enough mobility and reasonable on/off ratio [6]. Here we present a simple device model based on the Dirac equation for carriers in graphene and effective-mass model for carriers in MoS₂ to explain the experimental results for IV curves of graphene/MoS₂ heterostructures theoretically. The carrier transfer from graphene to MoS₂ is described by the relaxation process involving electron-phonon scattering. The IV curves are simulated according to the transport equations in semiconductors with the use of quasi-equilibrium distribution of carriers in leads. The dependence of transport characteristics on temperature, bias, and gate-voltage with and without optical pumping is simulated. The theoretical results are compared with available experimental data and good agreement is obtained with the use of a few physical parameters, which all fall into reasonable range.
[1] M. I. Katsnelson and K. S. Novoselov, Graphene: new bridge between condensed matter physics and quantum electrodynamics, Solid State Commun. 143, 3 (2007).
[2] J.-H. Chen et al., Intrinsic and extrinsic performance limits of graphene devices on SiO₂, Nat. Nanotechnol. 3, 206 (2008); F. Chen et al., Dielectric Screening Enhanced Performance in Graphene FET, Nano Lett. 9, 2571 (2009).
[3] K. F. Mak et al., Atomically thin MoS₂: a new direct-gap semiconductor, Phys. Rev. Lett. 105, 136805 (2010); A. Splendiani et al., Emerging photoluminescence in monolayer MoS2, Nano Lett. 10, 1271 (2010).
[4] B. Radisavljevic et al., Single-layer MoS₂ transistors, Nat. Nanotechnol, 6, 147 (2011); Y. Zhang et al., Ambipolar MoS₂ Thin Flake Transistors, Nano Lett. 12 (3), 1136 (2012).
[5] B. Radisavljevic and A. Kis, Mobility engineering and a metal-insulator transition in monolayer MoS₂, Nat. Mater. 12, 815 (2013); K.-K. Liu et al., Growth of Large-Area and Highly Crystalline MoS2 Thin Layers on Insulating Substrates, Nano Lett. 12, 1538 (2012); C. R. Wu et al., Multilayer MoS₂ Prepared by One-time and Repeated Chemical Vapor Depositions: Anomalous Raman Shifts and Transistors with High ON/OFF Ratio, J. of Phys. D: Appl. Phy. 48 (43) 435101 (2015).
[6] H. Tian et al., Novel field-effect Schottky barrier transistors based on graphene-MoS₂ heterojunctions, Sci. Rep., 4, 5951 (2014); M. Y. Lin et al., Toward Epitaxially Grown Two-Dimensional Crystal Hetero-Structures: Single and Double MoS₂/Graphene Hetero-Structures by Chemical Vapor Depositions, Appl. Phys. Lett. 105 (29), 073501 (2014).

Piezoelectric properties of transition metal dichalcogenide monolayers are important properties for the application of this kind of 2D material in various fields and nonlinear piezoelectric coefficients are vital since large strains are often observed for 2D materials. In this works, first principle’s calculation in the frameworks of density functional theory were performed to investigate the piezoelectric properties of TMX₂ (TM=Mo,W, X=S,Se,Te) monolayers. Polarization of the 1H phases of TMX₂ were obtained for a series of strains from -5% to 5% and a nonlinear relationship was found. Piezoelectric coefficients were obtained by fitting the simulated data and nonlinear terms of piezoelectric coefficients must be included to fit the simulated polarization results. The piezoelectric coefficients e₁₁ are 370.5, 410.8 and 537.0 pC/m for MoS₂, MoSe₂ and MoTe₂, and 267.7, 309.7 and 407.9 pC/m for WS₂, WSe₂ and WTe₂, respectively. The nonlinear terms e₁₁₁ and e₁₂₂ are -817 and -509 pC/m for MoS₂, -813 and -367 pC/m for MoSe₂, -1206 and -540 pC/m for MoTe₂, respectively. The coefficient e₁₁ and e₁₂ have almost the same value but opposite sign, but e₁₁₁ and e₁₂₂ are independent.

Fabrication of electrical and opto-electronic devices with vertically stacked transition metal dichalcogenides (TMDCs), leads to interfaces that are misoriented. Prior experimental and theoretical studies of misorientation in graphene bilayers demonstrated that a few degrees of misorientation is sufficient to decouple the low energy states of the individual layers. Experimental and ab-initio calculations have shown the bandgap of misoriented bilayer MoS₂ remains indirect. Some specific trends of the interlayer distance and band gaps have also been researched. However, the transport properties across the misoriented interface of the bilayer TMDCs are currently unknown. The coherent interlayer transmission across two stacks of MoS₂ are calculated for unrotated and rotated MoS₂ bilayers using ab-initio calculations and non-equilibrium Greens functions. The energy dependence of the interlayer transmission is analyzed. The effect of rotation angle on the electron and hole interlayer resistance is determined.

Two-dimensional (2D) heterostructures hold the promise for future atomically-thin electronics and optoelectronics due to their diverse functionalities. While heterostructures consisting of different transition metal dichalcogenide monolayers with well-matched lattices and novel physical properties have been successfully fabricated via van der Waals (vdW) or edge epitaxy, constructing heterostructures from monolayers of layered semiconductors with large lattice misfi