Symposium FF01 : Beyond Graphene 2D Materials—Synthesis, Properties and Device Applications

Hexagonal boron nitride (hBN) has many unique properties such as a strong interaction with thermal neutrons, a wide energy bandgap (>6.0 eV), and surface phonon polaritons. These properties make it a good candidate for compact, efficient, and low-cost neutron detectors, deep UV light emitters and detectors, and infrared nanophotonic devices. Thus, high quality hBN single crystals are needed for national defense (detection and monitoring of nuclear weapons by neutron detectors), health (sterilization of surfaces by radiation from deep UV LEDs), and microscopy (deep sub-wavelength optical imaging).

Molten metal solutions in nitrogen at atmospheric pressure have proven effective in growing large, monoisotopic, high purity, and high quality hBN single crystals. Single crystals have been grown that are up to 3mm across, hundreds of microns thick, and with etch pit densities of 10⁷ to 10⁸ cm^-2. Monoisotopic ¹⁰B and ¹¹B enriched hBN crystals were also grown simply by using isotopically pure boron as the source material. While epitaxial growth techniques (CVD and MBE) have also proven effective at growing hBN, these methods require more complex apparatus and processes than solution growth and require a substrate which introduces lattice constant and thermal expansion coefficient mismatches. Furthermore, they cannot produce crystals as thick as solution growth, nor readily produce monoisotopic hBN, as they require specialized chemical sources, which are rarely available and quite expensive. However, to become commercially viable, solution growth must be further refined to produce much larger crystals with lower defect densities, and lower residual impurity concentrations.

One of the primary obstacles to optimizing this process is the lack of a cohesive understanding of how the melt composition impacts hBN crystal growth. For example, several different combinations of metals (Ni, Ni/Cr, Fe, Fe/Cr), two different boron sources (pure boron powder and hot-pressed boron nitride), and other additives such as silicon have been used in varying concentrations within the melt. In each case, the crystal nucleation density, grain size area, thickness, and other characteristics have been affected by the particular composition used. However, how the melt composition affects these characteristics and what specific properties are responsible for these effects has not been thoroughly analyzed. Thus, it is difficult to explain why certain compositions of boron, metals, and nitrogen yield excellent quality crystals while others do not.

In this work, processes employing various combinations of Ni, Cr, Fe, and other additives with a boron source were systematically analyzed according to the crystal size (obtained from micrographs), defect density (determined via defect-selective etching), and quality (determined via a variety of spectrographic tools including XRD and Raman spectroscopy). Specific features of the melt were analyzed with respect to these criteria to pinpoint exactly which properties were most important in the process. For example, pure Ni and Fe both have high boron solubilities, but nitrogen solubility is negligible in pure Ni. While alloying with Cr greatly increases the nitrogen solubility, it also increases the nucleation density. Thus, relatively large hBN single crystals can be produced in pure Fe, but not pure Ni. Thick hBN layers can be produced in Ni/Cr solutions, but with small crystal grain sizes. In addition, residual impurities present in the source materials, intentionally added impurities, and the interaction of the melt with the furnace and crucible materials significantly affects the process and the crystal properties. This project gives new insights into the interplay between melt composition and hBN crystal growth and reveal new, direct pathways toward optimizing the process.

Layered bulk materials have attracted much attention for various technological applications including tribology and catalysis [1]. One of the most recent applications is in their study in the two-dimensional limit i.e. atomically thick materials such as graphene. Inorganic graphene analogues such as few layer molybdenum disulphide, black phosphorus and main group chalcogenides have also been the subjects of much intense research as two-dimensional semiconductors that would be complementary to graphene in a future electronics industry [2]. We have devised top-down routes to 2D black phosphorus [3] and tin(II) sulphide [4] both of which have layer dependent band gaps. However, these top-down processes suffer from a major drawback: often commercially available layered bulk crystals are used as feedstock and this inherently limits the palette of two-dimensional materials available for study without resorting to expensive processing via CVD.

As a solution to this, we have devised new routes based on solventless thermolysis of metal-organic precursors that can produce layered materials in bulk at relatively low temperatures and in relatively high purity. Because the precursors used are on the molecular scale, we can control the extent of doping or alloying of materials and this then widens range of layered materials available for exfoliation. I will present the synthesis of layered transition metal oxides and chalcogenides [5] and main group chalcogenide alloys [6], the latter also studied as post-processed 2D materials as an example of a new route to 2D materials with wide applicability.

[1] Tedstone et al. Chem. Mater. 2016, 28, 1965.
[2] Tedstone et al. Nanoscience Vol 4 (RSC Specialist Periodical Reports), 2017, 4, 108.
[3] Brent et al. Chem Commun. 2014, 50, 13338.
[4] Brent et al. J. Am Chem. Soc. 2015, 126, 9413.
[5] Zeng et al. Chem. Commun. 2019 ,55, 99.
[6] Norton et al. Chem. Sci. 2019, 10, 1035.

Given their established robustness and favorable optoelectronic properties, the semiconducting transition metal dichalcogenides (TMDs, e.g. MoS₂ and WSe₂) are attractive for optoelectronic applications including solar energy conversion (photovoltaic and photoelectrochemical solar fuel production).[1] Recent advances in the liquid-phase exfoliation (LPE) of semiconducting TMDs into mono- or few-layered 2D nanoflake dispersions suggests that inexpensive roll-to-roll processing can be used to prepare TMD-based devices inexpensively over large area.[2] However, the high concentration of defects in these materials act as recombination sites for photogenerated carriers and limit the performance. In this presentation the challenges with charge transport, separation, recombination, and interfacial transfer in LPE TMD nanoflake thin film devices will be discussed with respect to the 2D flake size and defect passivation/charge extraction treatments.[3] Our results give insight into the roles of both edge and internal defects and suggest routes for improvement. Overall it is shown that LPE semiconducting TMDs with suitable defect mitigation can achieve internal quantum efficiency for photon harvesting similar to bulk single crystal samples. Specifically, we show that WSe₂ nanoflake thin films achieve absorbed-photon-to-current efficiency over 50% and photocurrent densities for solar water reduction at 4 mA cm^–2 under standard testing conditions.[4]

[1] X. Yu, K. Sivula, ACS Energy Lett. 2016, 1, 315.
[2] X. Yu, M. S. Prevot, N. Guijarro, K. Sivula, Nat. Commun. 2015, 6, 7596.
[3] X. Yu, K. Sivula, Chem. Mater. 2017, 29, 6863.
[4] X. Yu, N. Guijarro, M. Johnson, K. Sivula, Nano Lett. 2018, 18, 215.

Graphene and 2-dimensional (2D) materials such as molybdenum disulfide (MoS₂) have attracted much attention in a wide range of unique electrical and optical properties. These high-quality 2D materials are obtained by means of mechanical or chemical exfoliation from natural crystals. But they are basically very small flakes and difficult to control thickness. In comparison with the exfoliation method, chemical vapor deposition has a better potential for fabricating large area 2D materials, but the synthesis is still a challenge. Furthermore, MoS₂ films have been prepared by dip-coating ammonium tetrathiomolybdates ((NH₄)₂MoS₄) followed by the annealing process in the atmosphere of Ar and S in order to make up for S deficiencies [1]. The purpose of our study is to synthesize large-area high-quality MoS₂ on SiO₂/Si substrate by thermolysis of (NH₄)₂MoS₄. By means of employing thioacetamide (CH₃CSNH₂), we are trying to compensate for deficiency of S in MoS₂ thin films and to improve the crystallinity. Moreover, we are performing Raman scattering measurements in order to explore the molar concentration dependence of the chemical precursors used for the growth of MoS₂.
Aiming for chemically synthesizing MoS₂ we employed the aqueous thermolysis method. (NH₄)₂MoS₄ and CH₃CSNH₂ with the same molar concentration were mixed in deionized water and the mixture was heated at 95 degree C for 5 hours. The solution containing the synthesized MoS₂ was spin-coated onto the 90-nm SiO₂/Si substrates. We confirmed that for the thin films synthesized with the molar concentration less than or equal to 10 mM, Raman double peaks E¹_2g and A_1g appeared at about 383 cm^-1 and 407 cm^-1, respectively. The E¹_2g and A_1g peak positions shift to a lower wavenumber with increasing the molar concentration. The Raman peak energy difference can be used to identify the number of MoS₂ layers. We found the thickness of our MoS₂ is as thick as that of MoS₂ natural crystal. On the other hand, for the films synthesized with the larger molar concentration E¹_2g and A_1g peaks did not appear and another two Raman peaks strongly appeared at around 810 and 1000 cm^-1 originating in MoO₃ [2]. We found a synthesis of MoS₂ using aqueous thermolysis method should be performed under the condition of a lower molar concentration of the chemical precursors.
Furthermore, as for the sample synthesized with the lower molar concentration of (NH₄)₂MoS₄ and CH₃CSNH₂ we performed Raman scattering measurement again after a week. As the result, Raman peak strongly appeared at around 810 cm^-1 originating in MoO₃. It is conceivable that oxidation of Mo is caused by a reaction with the residue of water after the synthesis, therefore we consider the residue of water should be evaporated completely to prevent the oxidation of MoS₂. For this purpose, we coated the solution directly on a SiO₂/Si substrate and heated at 200 degree C for 3 hours. Even in a week after the synthesis by means of this improved process, we confirmed E¹_2g and A_1g double peaks appeared and Raman peaks derived from MoO₃ did not. Now we are evaluating the electrical properties of these MoS₂ thin films. Moreover, we will stack the MoS₂ thin films and graphene for the application of gas sensors, which transduces gas-molecules adsorption on the surface of these 2D materials and at the MoS₂/graphene interface into a change of the resistance.
[1] K-K. Liu et al., Nano Lett. 12 1538-1544 (2012).
[2] K. Koike, et al., Jpn. J. Appl. Phys. 53, 05FJ02 (2014).

After the demonstration of a variety of inorganic two-dimensional (2D) materials (graphene, hBN, MoS₂, etc.), molecular 2D materials have attracted a significant research interest as well. However, the direct synthesis of these materials is an exceptionally challenging task for material scientists. In this contribution, a simple and robust physical method for the synthesis of molecular 2D materials is presented based on low-energy electron irradiation induced chemical reactions in aromatic molecular layers. In this way, ultrathin (~1 nm) molecular nanosheets with adjustable chemical and physical properties called Carbon Nanomembranes (CNM) can be prepared. Moreover, the method enables the synthesis of various other 2D organic-inorganic hybrids (e.g., MoS₂-CNM, graphene-CNM lateral heterostructures, etc.) [2] or ~20 nm thick nanosheets of organic semiconductors [3]. Functional properties of these molecular 2D materials including their chemical functionalization and engineering of hybrid hierarchical structures for application in electronic and energy conversion devices will be discussed [4-5].

[1] A. Turchanin, A. Gölzhäuser, Adv. Mater. 28 (2016) 6075.
[2] A. Winter, A. Turchanin et al., Carbon 128 (2018) 106.
[3] S. J. Noever, B. Nickel, A. Turchanin et al., Adv. Mater. 29 (2017) 1606283.
[4] C. Neumann, A. Turchanin et al., ACS Nano (2019) doi: 10.1021/acsnano.9b03475.
[5] A. Winter, A. Turchanin et al., 2D Materials 6 (2019) 021002

We report wafer-scale growth of atomically thin, three-dimensional (3D) van der Waals (vdW) semiconductor membranes. By controlling the growth kinetics in the near-equilibrium limit during metalorganic chemical vapor depositions of MoS₂ and WS₂ monolayer (ML) crystals, we have achieved conformal ML coverage on diverse 3D texture substrates, such as periodic arrays of nanoscale needles and trenches on quartz and SiO₂/Si substrates. The ML semiconductor properties, such as channel resistivity and photoluminescence, are verified to be seamlessly uniform over the 3D textures, and are scalable to wafer-scale. Additionally, we demonstrated that these 3D films can be easily delaminated from the growth substrates to form suspended 3D semiconductor membranes. Our work suggests that vdW ML semiconductor films can be useful platforms for patchable membrane electronics with atomic precision, yet in large-areas, on arbitrary substrates.

Two-dimensional (2D) materials own many interesting properties due to their atom-thick characteristics and versatile heterostructures built layer-by-layer. The capability of assembling both organic and inorganic 2D materials into heterostructures will lead to the formation of artificial van der Waals solids that takes advantage of molecular functionalities brought by the organic 2D materials. Towards this end, the biggest hurdle has been the lack of a general method to synthesize organic monolayers at macroscopic dimensions and integrate them with monolayer precision. Herein, we report the general synthesis of monolayer 2D polymers, the molecular analogs of 2D atomic crystals, with wafer-scale homogeneity and the fabrication of the hybrid superlattices. For this, we develop a new interfacial synthesis technique compatible with various molecular building blocks and polymerization chemistries. This approach incorporates key features necessary for scalable and facile processing, including large-area synthesis, ambient growth conditions, and compatibility with established patterning and integration methods. Enabled by those characteristics, hybrid organic-inorganic superlattices of monolayer 2D polymers and 2D atomic crystals were fabricated with molecular precision. The employment of versatile 2D polymers allows the incorporation of functional molecular moieties into two-dimensional electronic circuits. These materials show promising applications in multifunctional 2D electronic devices.

Utilisation of hexagonal boron nitride (h-BN) in applications requires further advancements in the production, processing, integration methods and also their cost reduction. We focus here on catalytic chemical vapour deposition (CVD) and discuss two very different catalysts for monolayer h-BN growth, iron and platinum, examples of a high and low precursor solubility transition metals. For Fe, we systematically explore the role of bulk dissolved species, and find that a simple pre-growth step enables us to tailor a scalable CVD process to give mm-sized h-BN domains,[1] among the largest reported to date. For Pt, we developed sequential step growth to enable independent control of h-BN nucleation and domain expansion to also give large (> 0.5 mm) h-BN domains and continuous films.[2] We show that targeted h-BN transfer methods are required for the different catalysts or even tweaked to the processing of the catalyst, whereby we develop a moisture oxidation and acid release method for the iron and a dry peeling approach for platinum. The utilisation of h-BN with graphene is demonstrated as an encapsulating layer for integration with common ALD dielectrics and in FET devices.[1] Additionally, we demonstrate the use of monolayer CVD h-BN as an active material for room temperature single photon emission.[3], [4]

References
[1] Babenko et al, submitted (2019).
[2] Wang et al. ACS Nano 13, 2114 (2019).
[3] Comtet, J. et al., Nano Lett. 19, 4, 2516 (2019)
[4] Stern, H. L. et al. ACS Nano 13, 4538 (2019).

2D materials have attracted wide interest because of their potential performance and diversity of function as electronic and optoelectronic materials. Growth of transition metal dichalcogenides (TMDs) such as MoS₂ and WSe₂ typically requires high temperatures (>700 ^oC) for large domain size and epitaxy. This restricts the applications of these materials, especially considering the growing interest in 2D materials integration with silicon in back-end-of-line (BEOL) applications that require processing temperature < 450 ^oC. In contrast, group III metal chalcogenides (indium selenide and gallium selenide) possess lower melting temperatures than TMDs and thus high crystal quality films are anticipated at lower growth temperature. In addition, depending on their stoichiometry and phase, they also have a layered structure that offers interesting physical, electronic, and piezoelectric properties down to the monolayer limit. In the case of indium selenide, there is interest in γ-InSe due to its high carrier mobility (>1000 cm²/Vs) and β-In₂Se₃ and a-In₂Se₃ which are ferroelectric phases. Despite the intriguing properties, there have been few studies thus far aimed at investigating the epitaxial growth and properties of indium selenide films.
In this study, we demonstrate the growth of β-In₂Se₃ thin films on various substrates in a vertical cold-wall metalorganic chemical vapor deposition (MOCVD) system at 400 ^oC using trimethylindium (TMIn) and hydrogen selenide (H₂Se) in a H₂ carrier gas. The In₂Se₃ films were grown epitaxially on c-plane sapphire and Si (111) surfaces. The films were identified as β-In₂Se₃ by both Raman and XRD. A low reactor pressure (100 Torr) and high total gas flow rate were required to suppress gas-phase reactions between TMIn and H2Se based on their Lewis acid and base properties. β-In₂Se₃ films were formed on both c-plane sapphire and Si (111) surfaces in this work, however, γ-In₂Se₃ films were synthesized on amorphous SiO₂/Si substrates indicating the importance of substrate type determining the crystal structure of the films. Top-gated thin film transistors (TFTs) fabricated on β-In₂Se₃ thin films reasonable mobility and on/off ratio and therefore offer potential applications in electronic devices.

Monolayer molybdenum disulfide (MoS₂) crystals grown on amorphous substrates such as SiO₂ are randomly oriented. However, when MoS₂ is grown on crystalline substrates, the crystal shapes and orientations are also influenced by their epitaxial interaction with the substrate. In this work, we present the results from chemical vapor deposition growth of MoS₂ on three different terminations of single crystal strontium titanate (SrTiO₃) substrates. On SrTiO₃(111), the monolayer MoS₂ crystals form equilateral triangles with two main orientations, in which they align their 〈2-1-10〉-type directions (i.e., the sulfur-terminated edge directions) with the 〈1-10〉-type directions on SrTiO₃. This arrangement allows near-perfect coincidence epitaxy between seven MoS₂ unit cells and four SrTiO₃ unit cells. On SrTiO₃(110), the MoS₂ crystals tend to align their edges with both the 〈1-10〉 and 〈1-1-2〉 directions on SrTiO₃ because these both provide favorable coincidence lattice registry. This distorts the crystal shapes and introduces an additional strain detectable by photoluminescence. When triangular MoS₂ crystals are grown on SrTiO₃(001), they again show a preference to align their edges with the 〈1-10〉 directions on SrTiO₃. Our observations can be explained if the interfacial van der Waals (vdW) bonding between MoS₂ monolayers and SrTiO₃ is greatest when maximum commensuration between the lattices is achieved. Therefore, a key finding of this work is that the vdW interaction between MoS₂ and SrTiO₃ substrates determines the supported crystal shapes and orientations by epitaxial relations. Controlled crystal orientations make the growth of large sheets of MoS₂ possible when there are multiple nucleation sites. This minimizes the number of grain boundaries and optimizes the electronic properties of the material, e.g., charge mobility, which is crucial for the application of monolayer MoS₂ in next-generation nanoelectronics devices.

Transition metal dichalcogenides (TMDs) are ideal for exploring fundamental physics and applied optics as they are semiconductors with a direct bandgap in the monolayer (ML) limit. Therefore, it is very important to have access to high quality and also large surface area films. Here we show that chemical vapour deposition (CVD) growth can yield high quality MoS₂ monolayers on SiO₂/Si substrates.

We show high structural quality of CVD grown MoS₂ MLs on SiO₂ [1] in 4k x 4k high-resolution transmission electron microscopy (HRTEM) images using the chromatic and spherical aberration-corrected low-voltage TEM instrument operated at 60 kV. We determine a defect concentration of 10¹³/cm² in as grown MoS₂ MLs. But we observe broad optical transitions in as-grown samples (50 meV FWHM at T = 4K) on SiO₂ that are not expected for high structural quality, therefore indicating detrimental ML-substrate interactions. Therefore, we lift off the CVD grown layers from the growth substrate and encapsulate them in hBN flakes, which give us access to the intrinsic optical quality of the MoS₂ MLs.

We compare the optical quality of MoS₂ MLs in three different structures: CVD grown MoS₂ films on SiO₂, exfoliated MoS₂ ML from bulk MoS₂ in exfoliated hBN and most importantly MoS₂ ML grown by CVD encapsulated in hBN crystals prepared by mechanical exfoliation. For the latter structure, we show (i) in photoluminescence the neutral A-exciton emission linewidth reduced to 5 meV at T = 4 K, as compare to 50 meV linewidth in the as grown CVD sample and (ii) in absorption well separated optical transitions A:2s stemming from excited states of the A-exciton Rydberg series, indicating comparable quality of our CVD MLs to exfoliated MoS₂ material. We optically generate valley coherence and valley polarization in our CVD grown MoS₂ layers, showing the possibility for studying spin and valley physics in CVD samples of large surface area.

[1] A. George, C. Neumann, D. Kaiser, R. Mupparapu, T. Lehnert, U. Hübner, Z. Tang, A. Winter, U. Kaiser, I. Staude, A. Turchanin, J. Phys.: Mater. 2 (2019) 016001

Hexagonal Boron Nitride (hBN) is a 2D, III-nitride wide bandgap semiconductor that has a structure very similar to graphene. Due to its extraordinary physical properties, such as high resistivity, high thermal conductivity, stability in air up to 1000°C, large bandgap (Eg ∼ 5.9 eV), hBN appears to be a promising material for emerging applications, including deep UV (DUV) optoelectronics, electron emitters and neutron detectors. There is also significant interest in monolayer and few-layer hBN as an encapsulating layer for 2D devices based on graphene and transition metal dichalcogenides.
While high quality hBN bulk crystals are available for exfoliation, the size of the crystals is limited, consequently there is continued interest in the epitaxial growth of large area hBN films using chemical vapor deposition (CVD). The low temperature (500-1000°C) CVD growth of hBN on metallic substrates such as Cu, Ni, Pt, Ru etc. has been widely studied using ammonia borane (H₃N-BH₃) and borazine (B₃N₃H₆) as single-source precursors. However, NH₃ has been used in some cases to control the N/B ratio. In addition, borazine is thermally unstable and readily reacts forming low volatility polymers. While growth on metal substrates is a viable approach, there is significant interest in the growth of hBN on non-metallic surfaces for device applications. The growth of hBN on a sapphire substrate was initially reported by Nakamura et. al. in 1986 followed by Kobayashi et. al in 2008 and others at a growth temperature greater than 1200°C. The precursors used for these growths includes triethyl boron (B(C₂H₅)₃, TEB), trimethyl boron (B(CH₃)₃, TMB) and NH₃ as the source of N. It is well known, however, that organic sources such as TEB result in carbon incorporation which is the biggest concern for the growth of epitaxial high purity BN films.
In this work, we are studying the growth of crystalline sp² BN in a vertical cold wall MOCVD reactor using a mixture of diborane (5% B₂H₆ in H₂) and NH₃ as precursors for B and N, respectively using H₂ as a carrier gas. The growths are performed on nitrided c-plane sapphire substrates at temperatures ranging from 1100-1500°C. Diborane reacts with NH₃ at temperatures as low as room temperature to form H₃N-BH₃ and other volatile B-N species, consequently, gas phase chemistry plays an important role in this deposition process. The influence of temperature, N/B inlet gas ratio and precursor flow method (pulsed versus continuous) on the material quality, characteristics and deposition rate were studied. In the lower temperature range (<1200°C), stoichiometric sp² BN films were obtained on sapphire using a N/B ratio of ~600 under continuous precursor flow giving a growth rate of ~10 nm/hr. Sequential pulsing of the B₂H₆ and NH₃ precursors resulted in a significant increase in the growth rate (~18nm/hr) due to a reduction in the extent of gas phase reaction. Additional studies are underway to investigate growth at higher temperatures and over a wider parameter space to ascertain the growth mechanism as a function of conditions to provide additional insights into the role of gas phase chemistry on the properties of highly crystalline sp² BN films on sapphire.

Despite an abundance of reported growth methods, a key challenge for both current research and future technologies is the efficient and scalable growth and device integration of “electronic-grade” 2D layers. Metal organic chemical vapour deposition (MOCVD) has emerged as highly suitable particularly for transition metal dichalcogenide (TMDC) layer growth [1]. However, the understanding of the MOCVD process and control over detailed TMDC micro-structure remains currently limited, and carbon contamination and long growth times are of concerns. Our motivation is to advance the understanding of the growth mechanism of TMDCs with the focus on developing improved, scalable process technology. We employ here a “deconstructed” MOCVD process for WS₂, based on a simple sequential exposure pattern using precursors with low toxicity, namely tungsten hexacarbonyl and dimethylsulfide [2]. Combined with the use of a cold-walled reactor set-up, this minimises precursor pre-reactions and warrants a substrate surface bound reaction path. We use this as model system to systematically explore each growth aspect, including different substrates and their pre-treatment. For Au support, we find a significant catalytic effect that not only allows a significant reduction of the carbon contamination but also a self-limiting behaviour to mono-layer WS₂ with reasonable crystallinity, and full coverage within 10min exposure times. We discuss in this context of overlap between MOCVD and ALD processes, and, in parallel to other 2D materials, of suitable device integration strategies [3].

References
[1] Briggs et al. 2D Mater. 6, 022001 (2019)
[2] Fan et al. submitted (2019)
[3] Wang et al. ACS Nano 13, 2114 (2019).

The ability to synthesize large-area and high quality atomic films is a prerequisite for their successful integration into a wide variety of novel and existing technologies. Here we show the growth of transition metal dichalcogenides (MoS₂, WS₂ and WSe₂) via modified chemical vapor deposition (CVD) methods using volatile precursors [1,2]. The use of high vapor pressure precursors allows for the controlled delivery to the growth sample [3], and therefore, suitable for homogeneous and large-scale synthesis, as required for many applications. However, one of the problems with these precursors is the small domain size usually obtained [3]. In this scheme, a modified approach in which the metal and chalcogen precursors are delivered in a pulsed fashion is demonstrated. This approach allows to achieve a ten-fold increase in the domain size, from ~500 nm (or below) to ~10s of microns. Moreover, we demonstrate that the growth kinetics is highly dependent on the surface chemistry and by controlling it, the growth of ad-layers is inhibited and thus, more than 95% monolayer films are obtained. Another advantage of using volatile precursors is the ability to control the lateral and vertical heterostructures formation, and will be described as well. Following this work and in order to expand our growth capabilities, the growth of monochalcogenides using lessons learned while growing TMDs will be briefly described. Systematic structural, chemical, spectroscopic and electrical characterization will be shown for both systems as well. This work presents a big step towards the controlled growth of 2D materials.

[1] Hod O., Urbakh M., Naveh D., Bar-Sadan M. and Ismach A. Adv. Mat., 2018, 30.
[2] Radovsky G, Shalev T and Ismach A., J. Mater. Sci., 2019, 54, 7768-79.
[3] Cohen A., Ranganathan K., Patsha A. and Ismach A. In Preparation.

Although the existence of thousands of new 2D materials are being theoretically predicted [1], only a handful can be experimentally synthesized with the level of control required for their large-scale integration in future innovative devices and technologies. This is mainly due to the insufficient control over the nucleation at vdW interfaces, to the difficulty of implementing self-limiting monolayer growth in many systems and to the lack of experimental data regarding the van der Waals growth dynamics. In this talk, we will present in situ Low-Energy Electron Microscopy (LEEM) studies of the van der Waals growth dynamics of novel group V 2D materials. By combining LEEM and LEED measurements of the growth dynamics of antimonene on germanium and graphene substrates [2, 3] with semi-empirical modelling of the growth kinetics, we are able to determine the mechanisms governing the lateral and vertical growth modes of this new 2D material and to achieve a better control over the growth morphology. Moreover, we will present experimental evidence of the synthesis of a new 2D As_xSb_1-x alloy with tunable As content. These results lay the groundwork for the implementation of novel group V 2D materials and heterostructures in electronic, optoelectronic and quantum devices.

[1] N. Mounet, M. Gibertini, P. Schwaller, D. Campi, A. Merkys, A. Marrazzo, ... & N. Marzari., Nat. Nanotechnol. 2018, 13(3), 246.
[2] M. Fortin-Deschênes, O. Waller, T. O. Mentes, A. Locatelli, S. Mukherjee, F. Genuzio, P. L. Levesque, A. Hebert, R. Martel, O. Moutanabbir., Nano Lett. 2017, 17(8), 4970–4975.
[3] M. Fortin-Deschênes, R. M. Jacobberger, C. A. Deslauriers, O. Waller, É. Bouthillier, M. S. Arnold, O. Moutanabbir., Advanced Materials. 2019, 31(21), 1900569.

Two-dimensional elemental materials (X-enes) have attracted increasing research attention due to tunable bandgap in between graphene and TMDs. As an emerging 2D X-ene, bismuthene, is expected to yield descent carrier mobility, large spin-orbit coupling, and topological states. Currently, bismuthene or bismuth thin films can be grown via chemical vapor deposition (CVD), pulsed laser deposition (PLD) and molecular beam epitaxy (MBE). MBE can provide high quality bismuthene with sophisticated facilities, whereas PLD or CVD method struggles for precise thickness control and uniformity. It is of great importance to develop a cost-effective method to produce high quality bismuthene. Herein, we report our pioneer work on 2D bismuthene growth via e-beam evaporation and its device study showing comparable properties to MBE counterpart for potential innovative nanoelectronics.
2D bismuth thin film is grown on Si (111) or SiO₂/Si substrates in an e-beam evaporator. Before the deposition, Si (111) and SiO2/Si substrates were cleaned by ultrasonic in acetone for 5 min and rinsed by isopropyl alcohol and deionized water. Deposition of bismuth was initiated from a 99.999% pure Bi source at a rate of 0.1Å/s under 5.0×10^-4 Pa pressure. The thickness of our Bi thin films can vary from 1-50 nm. Annealing effect is also studied via a tube furnace at 50-260^oC in Ar atmosphere. Raman spectroscopy confirmed the successful growth of 2D bismuth with signatures at 70.4 cm⁻¹ (E_g) and 96.1 cm⁻¹ (A_1g) _,which have a blue shift as thickness increases. X-ray diffraction on evaporated 20-nm thick Bi films observed expected hexagonal (001) orientation - a good crystallinity. A typical surface roughness is ~1nm indicated by atomic force microscope imaging over a 10×10μm² area, indicating a smooth surface for following device fabrication. The evaporated 2D bismuth, as the channel material, on SiO₂/Si or Si (111) substrates, was then subjected to a field effect transistor fabrication process via lithography and Ti/Au electrodes evaporation. Electrical characterization, such as I_d-V_d curve for different thickness 2D bismuthene has been characterized on a Cascade® EPS150 probe station. A linear I_d-V_d curve reveals a good ohmic contact with conductivity ~ 3.65×10⁴ S/m for 20-nm 2D Bi and ~6.27×10⁴ S/m for 30-nm. Experimental research on photoelectric and thermoelectric behavior of e-beam evaporated 2D bismuthene has been carried out, which exhibited similar properties comparable to MBE bismuthene samples we prepared under the same material and device configuration in this study. Photoelectricity is observed under a wide range (visible to IR) laser exposure with photo responsibility ~10 A/W at unit power exposing to 800-nm laser. The Seebeck coefficient of our 2D bismuth was measured as -17.8μV/k (N type) at room temperature, which is greater than graphene.
This study explored e-beam evaporation of 2D bismuth film and its material/device characterization comparable to MBE counterpart. This cost-effective method to prepare good quality bismuthene and its device holds great promise for innovative photoelectric and thermoelectric applications.

2D layered materials reside in a class of crystals consisting of weakly bonded atomic layers by van der Waals attraction. In particular, this distinct structure feature renders 2D layered materials exceptional physical and chemical properties inaccessible in bulk crystals, e.g., quantum confinement and surface effects, which offer 2D materials a wide range of applications in electronics, optoelectronics, energy storage, catalyst, and chemical sensing fields. One of the remaining challenges in the 2D material arena is to explore novel layered materials with breaking new properties. Rather than searching for new materials, we pursued developing reliable chemical routes, intercalation/deintercalation of metals, to tailor the original 2D materials with tunable chemical and physical properties. This talk will cover a series of innovative chemical intercalation/deintercalation of metals in layered crystals (chalcogenides and oxides), followed by examples of unusual phenomena introduced by intercalation, including polytypic phase transitions in metal intercalated Bi₂Se₃, and the chemo-chromism/thermochromism of MoO₃ intercalated with metal.

Quantum materials are prime candidates for next-generation energy-efficient technologies, such as topological quantum computing, quantum sensing, and neuromorphic computing. While van der Waals 2D materials exhibit a compellingly wide range of exotic and potentially useful properties such as charge density waves, topological insulator edges, and superconductivity, one can also realize these properties by stabilizing new 2D allotropes of traditionally 3D superconductors and magnets. In this talk I will discuss our pioneering work in confinement heteroepitaxy (CHet) that enables the creation of 2D forms of 3D materials (e.g. 2D-Ga, In, Sn, Ag, Au, etc.) and decouples the growth of the metals from other 2D layers, thereby enabling a new platform for creating artificial quantum lattices with atomically sharp interfaces and designed properties. As a specfic example, we synthesize plasmonic layers that exhibit >2000x improvement in nonlinear optical properties, and 2D-superconductors combined with topological insulators as the building block of next generation “2D” topological superconductors. Confinement heteroepitaxy opens up avenues for enabling a virtual “legoland” of hybrid quantum materials.

As a two-dimensional analog of graphene, silicene has recently been the subject of extensive research interest. The easy oxidation of silicene in air limits its applications in device design and fabrication. In this talk we report the experimental fabrication of silicene protected by a graphene overlayer. The graphene layer is grown first on a Ru(0001) substrate and silicene is grown under it by Si intercalation. By controlling the amount of silicon, ordered arrays of pseudomorphic silicene nano flakes as well as single layers and multilayers of silicene can be successfully fabricated under graphene that is epitaxially grown on Ru(0001). Density-functional-theory calculations show weak interactions between graphene and silicene layers, indicating that the fabricated structures are silicene/graphene van der Waals heterostructures. The as-prepared silicene-based structures show no observable damage after air exposure for two weeks. The vertical graphene/silicene/Ru heterostructure shows rectification behavior with an ideality factor of ~1.5 [1].

References:
[1] G. Li et al., Adv. Mater., 30, 1804650 (2018)

Two-dimensional (2D) van der Waals (vdW) materials fabricated by stacking of homogeneous or heterogeneous monolayer “building blocks” are attracting increased attention, since their unique physical and chemical properties offer great promise for use in atomically thin electronic and optoelectronic devices. For stacks of homogeneous 2D vdW materials, the number of layers plays a crucial role in the ultimate optoelectronic properties. Since the change in properties from monolayer to bilayer to trilateral is typically more significant than that resulting from additional layers, there is great interest in achieving precise control over bilayer/trilayer formation, including the stacking geometry. In this talk I will discuss how we have made progress in this area through new methods of controlling the catalytic growth substrate, in the case of graphene, and growth promoting compounds in the case of TMD materials.

Two-dimensional (2D) transition metal dichalcogenides (TMDs) exhibit an intriguing array of layer-, topology-, and morphology-dependent properties, which have demonstrated enormous potential in optoelectronics, energy conversion, and quantum information science. Although “top-down” approaches can be used to define crystal morphologies and dimensions, explicit synthetic control of the physical properties of TMD crystals is desirable, but also a major challenge. We demonstrate a gas-phase synthesis method that significantly transforms the structure and dimensionality of MoS₂ and other TMD crystals without lithography. Synthesis of MoS₂ on Si (001) surfaces pre-treated with phosphine (PH₃) furnishes high aspect ratio nanoribbons of crystalline 2H phase MoS₂. The widths of these nominally 1D crystals are exceptionally uniform and their edges are sharper than those of TMD crystals prepared by most other means. Notably, the widths of the MoS₂ nanoribbons can be systematically controlled between 70 nm and 500 nm by varying the concentration of PH₃ gas introduced during the Si surface treatment step. Detailed kinetic studies supported by cluster expansion and DFT calculations indicate that the structure and concentration of the underlying Si–phosphide moieties is instrumental in transforming growth of MoS₂ from a conventional 2D triangular to nominally 1D morphology. Notably, far-field and near-field photoluminescence (PL) studies of 1D MoS₂ crystals show that they exhibit an emission peak which is blue-shifted by 50 meV relative to that of 2D MoS₂ nanocrystals. Our efforts highlight future opportunities for development of designer substrates that could mediate the synthesis of new low-dimensional crystals with prescribed structures and properties.

In the family of two dimensional (2D) materials, molybdenum disulfide (MoS₂) has received immerse attention owing to its unique electrical, catalytic, biological and mechanical properties. Herein, we report a robust and scalable approach to synthesize MoS₂ nanoribbons with controlled dimensions. Advanced characterizations confirm the chemistry, morphology, and crystalline structures of materials obtained at different reaction stages when synthesizing MoS₂, including MoO₃, MoS₂/MoO₂ hybrid, and MoS₂ nanoribbons. With the electric tweezers based on combined AC and DC electric fields, the MoS₂ nanoribbons can be readily manipulated with desired orientations along arbitrary trajectories, e.g. a cat drawing. Moreover, it is found that the electromechanical behaviors of the particles obtained at different stages strongly correlate with their electric properties, the demonstration of which could be utilized to monitor and understand the synthesis of targeted nanomaterials. This work could lead to a new paradigm in the large-scale fabrication of 2D materials with designed dimensions and inspire their innovative applications in nanorobotics, micro/nanoelectromechanical system devices (MEMS/NEMS), as well as molecule delivery and release.

The ability to manipulate the twisting topology of van der Waals (vdW) structures enables a new degree of freedom to tailor their electrical and optical properties. In particular, the twist angle has been shown to strongly affect the electronic states, excitons and phonons of the twisted structures via interlayer coupling, giving rise to exotic optical, electric and spintronic behaviors.In twisted bilayer graphene, at certain twist angles, long-range periodicity associated with Moiré pattern introduces flat bands and highly localized electronic states, resulting in Mott insulating behavior and superconductivity.Beyond bilayer graphene, recent theoretical studies suggest these twist-induced localization and the formation of flat bands are phenomena common to various layered materials such as transition metal dichalcogenides and black phosphorus.Twisted vdW structures are usually created using transfer-stacking method, where atomic layers are mechanically exfoliated and transferred to stack with controlled twist angles in a layer-by-layer configuration. This approach, however, is not suitable to create twisted structures of a variety of layered materials with relatively strong interlayer binding. In contrast, facile bottom-up growth methods could provide an alternative means to create twisted vdW structures. Herein, we demonstrate that the Eshelby twist, associated with a screw dislocation, a chiral topological defect can drive the formation of twisted vdW structures on multiple scales ranging from nanoscale to mesoscale. In our synthesis method, axial screw dislocation was first introduced into vdW nanowires with growth direction along the vdW stacking direction (cross-plane direction), yielding vdW nanostructures with continuous twisting. Further growth on the dislocated nanowires adhered to the substrate results in the discretization of the Eshelby twist in the mesoscale structures, yielding unique twisting structures consisting of a helical assembly of nanoplates. In these mesoscale helical crystals, atomically sharp interfaces with various twist angles are created, providing an intriguing tunable platform to explore various interlayer coupling effects. Our theoretical modeling suggests that the formation of the discretized twisting results from the interplay between the interfacial energies associated with the twist interfaces and the elastic strain energy defined by the Eshelby twist. We further show that the twisting topology can be tailored by controlling the radial size of the structure.

Since the discovery of graphene [1], other 2D materials such as silicene, germanene, borophene, phospherene, and various transition metal dichalcogenides (TMDs) have been explored in a great deal due to their novel structure and tunable properties. In contrast, group VI elements such as tellurium or selenium have yet to be stabilized in purely 2D forms, which may be due to its stable covalently bonded Te atomic chains that spiral along its c-axes. In this case, a typical exfoliation technique to produce 2D monolayers would not work. However, the density functional theory calculations by Zhu et al. predict that 2D atomic layers of tellurium might exist in stable 1T-like structure (a-Te), and the metastable tetragonal (b-Te) and 2H-like (g-Te) structures [2].
Here, we report the first experimental realization of a 2D group VI tellurium chalcogen, namely tellurene. Ultra-thin flakes of tellurium with controlled thickness were produced by substrate-free solution process. Each 2D tellurene flakes have thicknesses ranging from 15 to 100 nm. Tellurene flakes with a thickness smaller than 10 nm were also produced through a solvent-assisted post-growth thinning process [3]. A single atomic layer 2D tellurium was also synthesized by a wafer-bonding assisted self-assembly process, a new approach to synthesize and stabilize the low-dimensional structure. Each monolayer Te atom has four nearest neighbors within the plane with inter-atomic distances of 0.31 – 0.32 nm. Atomic-resolution chemical mapping and structural characterization via scanning transmission electron microscopy (STEM) leading to the definitive identification of single atomic layer tellurene. First-principles calculations found tellurene to be metallic, with electronic band structures containing Dirac-cone-like features and exhibiting significant asymmetric spin-orbit band splitting. These findings enable further research into the suitability of tellurene for device applications, such as spintronics and quantum computing.

[1] K.S. Novoselov et al., Science 306, 666 (2004).
[2] Z. Zhu et al., Phys. Rev. Lett. 119, 106101 (2017).
[3] Y. Wang et al., Nature Electronics 1, 228 (2018).

*This research was supported in part by the U.S. Department of Energy through the EERE-Sunshot BRIDGE (DE-EE0005956) and PVRD (DE-EE 0007545) programs, Global Research and Development Center Program (2018K1A4A3A01064272) and Brain Pool Program (2019H1D3A2A01061938) through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT.

Two-dimensional (2D) layered halide perovskites are highly promising semiconductors for optoelectronic applications ranging from solar cells, light emitting diodes and detectors. This class of 2D materials gaining significance as they offer high synthetic versatility and allow for more specialized device implementation due to the directional nature of the crystal structure. A remarkable advantage of the 2D perovskites is the readily tunable functionality by incorporating a wide array of organic cations into the 2D framework and by controlling the slab thickness, in contrast to the 3D analogues which have limited scope for structural engineering. We present the new homologous series, (C(NH₂)₃)(CH₃NH₃)_nPb_nI_3n+1 (n = 1, 2, 3), of layered 2D perovskites which is different from Ruddlesden-Popper type. These compounds adopt an unprecedented structure type which is stabilized by the alternating ordering of the guanidinium and methylammonium cations in the interlayer space (ACI). The these 2D perovskites combine structural characteristics from both Dion-Jacobson (DJ) and Ruddlesden-Popper (RP) structure archetypes. We also report the first examples of hybrid DJ hybrid 2D lead iodide perovskites which consist of thick perovskite slabs (n>1) with layer number (n) ranging from 1 to 5. We describe two new DJ perovskite series based on bivalent (+2) spacer cations deriving from a piperidinium (C₅NH₁₂) organic backbone. The new DJ perovskites are built from 3AMP (3AMP = 3-(aminomethyl)piperidinium) and 4AMP (4AMP = 4-(aminomethyl)piperidinium) spacers and methylammonium (MA) perovskitizers cations to form A’(MA)_n-1Pb_nI_3n+1 (A’ = 3AMP or 4AMP, n = 1-4) homologous series acting as spacers. Surprisingly, a slight difference in the position of the-CH₂NH₃⁺ group on the piperidine chair (3- and 4- position with respect to the piperidine nitrogen) exerts a strong influence on the crystal structure, which is reflected on the distortion of the inorganic layers.

The exponentially improving performance of conventional digital computers has slowed in recent years due to the speed and power consumption issues that are largely attributable to the von Neumann bottleneck (i.e., the need to transfer data between spatially separate processor and memory blocks). In contrast, neuromorphic (i.e., brain-like) computing aims to circumvent the limitations of von Neumann architectures by spatially co-locating processor and memory blocks or even combining logic and data storage functions within the same device. In addition to reducing power consumption in conventional computing, neuromorphic devices also provide efficient architectures for emerging applications such as image recognition, machine learning, and artificial intelligence. With this motivation in mind, this talk will explore the opportunities for two-dimensional materials in neuromorphic devices. For example, by combining p-type black phosphorus with n-type transition metal dichalcogenides, gate-tunable diodes have been realized, which show anti-ambipolar transfer characteristics that are suitable for artificial neurons, competitive learning, and spiking circuits [1]. In addition, by exploiting field-driven defect motion mediated by grain boundaries in monolayer MoS2, gate-tunable memristive phenomena have been achieved, which enable hybrid memristor/transistor devices (i.e., “memtransistors”) that concurrently provide logic and data storage functions [2]. The planar geometry of memtransistors further allows multiple contacts to the channel region that mimic the behavior of biological neurons such as heterosynaptic responses [3]. Overall, this work introduces new foundational circuit elements for neuromorphic computing in addition to providing alternative pathways for studying and utilizing the unique charge transport characteristics of two-dimensional materials [4].

[1] V. K. Sangwan, et al., Nano Letters, 18, 1421 (2018).
[2] V. K. Sangwan, et al., Nature Nanotechnology, 10, 403 (2015).
[3] V. K. Sangwan, et al., Nature, 554, 500 (2018).
[4] D. Jariwala, et al., Nature Materials, 16, 170 (2017).

A current artificial intelligence (AI) system consumes massive computing resources from supercomputers to handle computing traffic inside deep neural networks (DNN). Although interesting application-specific integrated circuits (ASIC) solutions in CMOS have been introduced, we are still suffering limitations on memory footprint, in-memory computing, and on-/off-chip communication. In this regard, a novel computing architecture based on non-volatile resistive switching devices called RRAM has shown a great potential as the RRAM device can continuously tune its analog conductance under moderate voltage stimulation, and thus each device can efficiently execute the “synaptic weight” in the neural network. In addition, the RRAM based computing can accomplish considerably higher integration density compared with the current CMOS technologies and highly efficient parallel computing. Therefore, people have put intensive effort in development of the RRAM array-based system beyond the current supercomputing system [1].
Here, we propose a novel approach to demonstrate stable RRAM array based on 2D heterostructure, which enables real neuromorphic computing for the first time. 2D materials have been known to have atomically thin feature for high integration density and anisotropic ionic transport characteristic in the in-plane and the out-of-plane direction for stable swtiching filament formation. It makes them promising as a switching medium for neuromorphic computing. Despite of this great potential, there was no study for real neuromorphic computing demonstration using 2D materials based memristors because limited approaches exist for large scale, uniform 2D materials demonstration. Last year, we reported an interesting method for large-scale, uniform 2D materials which ensures pristine interface at interface of heterostructures [2]. With the method, we, for the first time, have realized a 2D heterostructure memristor array having large memory window over 10⁴ and stable filament formation even after 100 times set-reset cycling test, which successfully functions to implement image recognition simulation. We strongly believe that this innovative approach can open a new opportunity for the neuromorphic computing.

[1] M. Jaidan et al, Nature Electronics, 1, 22-29 (2018)
[2] J Shim, S. Bae et al, Science, 09, 362, 6415, 665-670 (2018)

A non-volatile, solid-state, one-transistor (1T) memory is demonstrated based on electric double layer (EDL) gating of a WSe₂ field-effect transistor (FET) using an electrolyte that is a single molecular layer thick. The “monolayer electrolyte” consists of cobalt crown ether phthalocyanine and lithium ions, which are positioned by field-effect at either the surface of the WSe₂ channel or a h-BN capping layer to achieve ‘1’ or ‘0’, respectively. Bistability in the monolayer electrolyte memory is significantly improved by the h-BN cap, with density functional theory (DFT) calculations showing an enhanced trapping of Li⁺ near h-BN due to a ~1.34 eV increase in the absolute value of the adsorption energy compared to vacuum. The threshold voltage shift between the two states corresponds to a change in charge density of ~ 2.5 × 10¹² cm^-2, and an on-off ratio exceeding 10⁴ at a back gate voltage of 0 V. The on-off ratio remains stable after 1000 cycles and the retention time for each state exceeds 6 hours (max measured). When the write time approaches 1 ms, the on-off ratio remains > 10², showing that monolayer electrolyte-gated, 2D non-volatile FET can respond on timescales similar to existing flash memory. The data suggest that faster switching times and lower switching voltages could be feasible by top gating.

The research is supported by the National Science Foundation (NSF, U.S.) under Grant No. ECCS-GOALI-1408425 and DMR-CAREER-1847808. M. Wu and W. Wang acknowledge support from the National Key Research and Development Program of China with No. 2016YFB0901600 and National Natural Science Foundation of China with No. 11874223 and 51871121.

Diverse forms of two-dimensional (2D) materials have been attracting a lot of attention as emerging materials for low-power and high-performance electronic or optoelectronic devices¹. Layered double hydroxide(LDH) is a kind of 2D layers consisting of a positively charged metal hydroxide layer and intercalated charge-balancing anions, expressed by [M^{2 +} _1-xM^{3 +} _x (OH) ₂]^{x +} (A^n-) _{x / n} . mH₂O². In this study, we synthesized three LDH layers using three divalent cations such as Zn²⁺, Ni²⁺, and Co²⁺ and controlled each packing density, which all were confirmed by XRD and SEM analysis. Then, we analyzed and performed the electrical properties for each LDH in a form of two-terminal vertical junction structures using Pt and FTO electrode. Interestingly, only densely-packed ZnAl-LDH based device exhibited unipolar switching behavior with 10³ ON-OFF ratio and acceptable stability, while all others did not show any switching feature. The potential reason for the switching success or failure might be attributed to the morphology difference in LDH layers induced by different ion components and concentrations during LDH synthesis. A potential switching mechanism can be suggested by space-charge-limited conduction (SCLC) transport and directional motion of ionic vacancy by the electric field. Our result indicates the importance of 2D LDH stacking morphology for the implementation of the nonvolatile resistive switching device.


[1] Cho, J. H. et al. Probing Out-of-Plane Charge Transport in Black Phosphorus with Graphene-Contacted Vertical Field-Effect Transistors. Nano Lett., 2016, 16 (4), 2580-2585
[2] Zhao, Y. et al. Tunable Electronic Transport Properties of 2D Layered Double Hydroxide Crystalline Microsheets with Varied Chemical Compositions. Small, 2016, 12 (33), 4471-4476

The recent progress in memristor technology has made it an appealing competitor for future computing architectures and neuromorphic computing systems. However, the scale-down ability and uncertain switching mechanism are still the stumbling blocks to the development of memristor. The emerging transition metal dichalcogenides (TMDCs) has been recognized as a promising candidate for the nanoscale memristive devices owing to their flexibility, transparency and especially scale-down ability. In this work, we demonstrate a type of metal-semiconductor-metal (MSM) memristor architecture based on transfer-free WS₂/MoS₂ heterojunction, completely different from the conventional metal-insulator-metal (MIM) architectures using insulating metal oxides as the function layer. This MSM device is forming-free and shows a reproducible bipolar memristive behavior with R_ON/R_OFF ratio up to 10⁴. Moreover, the programming voltages of this device is around 1.4 V with currents lower than 10 µA and the thickness of the function layer down to 10 nm. Finally, we discuss the switching mechanism in the WS₂/MoS₂ heterojunction. Different from the filament-based mechanism, the resistance switching of the WS₂/MoS₂ 2D heterojunction memristor is mainly ascribed to the electric-field modulation of the band structure of the WS₂/MoS₂ heterojunction, which is promising for the development of high-speed and low-power memristors.

Two dimensional (2D) layered materials are crystalline materials with layered structures, including Graphene, h-BN, and Transition Metal Di-chalcogenides (TMD’s). We have studied 2D layered materials in two directions.
For the near future application, we have been investigated 2D layered materials for Si technology to enhance their performance. To overcome performance limit for the long term downscaling of Si technology, we have focused 2D layered materials as interface materials due to the chemical inertness and their atomically thin nature. We investigated Graphene/metal hybrid interconnect. Although high quality graphene can be produced on catalyst metals, Si technologies are limited by the high temperature growth and the post transfer process. W/nanocrystalline graphene (ncG)/TiN is realized by direct growth on noncatalytic TiN, up to 12" wafer, at a low temperature of 560 °C which is below the semiconductor integration temperature. ncG acts in the interconnect not only a diffusion barrier to metal silicide formation but also a promoter of the preferential grain growth of the W layer. Overall, a significant reduction (27%) in the resistance of the interconnect is achieved by the insertion of ncG between W and TiN. This work points to the possibility of practical graphene applications via direct ncG growth that is compatible with current Si technology.[1] Also, We demonstrated that 2D layered materials are good candidates for interface materials between metal and Si to reduce the Schottky barrier heights and contact resistance in source and drain, which is one of the most critical issues for scaling down.[2,3,4] We found that 2D materials may change the pinning point of Schottky barrier and end up reducing contact resistance. We realized the lowest specific contact resistivity of 3.30 nΩcm² (n-type Si, 10¹⁵/cm³) and 1.47 nΩcm² (n-type Si, 10²¹/cm³) via Graphene and h-BN insertion in 6” wafer are approaching the theoretical limit of 1.3 nΩcm².[3]
As one of the long term applications, we are developing transistors beyond the 5 nm node for high integration / high performance / low power, and beyond Si devices for optoelectronic applications in the IR domain. In pursuit of these research directions, we are studying a large area growth and doped structure growth. Recently, we grew MoS₂ monolayer on 6 inch wafer within 10min by sequential growth method. [5] To explore 2D application, we examined triboelectric nanogenerator, which has been explored as one of the possible candidates for the auxiliary power source of portable and wearable devices. [6] We investigated the triboelectric charging behaviors of various 2D layered materials, including MoS₂, MoSe₂, WS₂, WSe₂, graphene, and graphene oxide and confirms the position of 2D layered materials in the triboelectric series. It is also demonstrated that the results are related to the effective work functions. This study provides new insights to utilize 2D layered materials in triboelectric devices, allowing thin and flexible device fabrication. [7]

[1] C.-S. Lee, et al., “Fabrication of Metal/Graphene Hybrid Interconnects by Direct Graphene Growth and Their Integration Properties”, Advanced Electronic Materials 4, 1700624 (2018)
[2] K.-E. Byun, et al., “Graphene for True Ohmic Contact at Metal-Semiconductor Junctions”, Nano Letters 13, 4001 (2013)
[3] M.-H. Lee, et al., “Two-Dimensional Materials Inserted at the Metal/Semiconductor Interface: Attractive Candidates for Semiconductor Device Contacts”, Nano Letters 18, 4878 (2018)
[4] S.-G. Nam, et al., “Barrier height control in metal/silicon contacts with atomically thin MoS₂ and WS₂ interfacial layers”, 2D Materials 5, 041004 (2018)
[5] In preparation.
[6] K.-E. Byun, et al., “Potential role of motion for enhancing maximum output energy of triboelectric nanogenerator”, APL Materials 5, 074107 (2017)
[7] M. Seol, et al., “Triboelectric Series of 2D Layered Materials”, Advanced Materials 30, 1801210 (2018)

Semiconductor p-n junction diodes and field-effect transistors are essential building blocks of electronic and optoelectronic devices. Over the past several decades, in the quest for higher performance, the dimensions of functional devices comprising 3D bulk semiconductors continue to decrease. However, the inherent scaling limitations of 3D devices such as short-channel effects, have fueled the exploration of alternative semiconductors. 2D layered materials have been a central focus of materials research for More Moore and More-than-Moore device applications because each layer of 2D materials consists of a covalently bonded and dangling-bond-free lattice and is weakly bound to neighboring layers by van der Waals interactions.^1,2 Consequently, 2D layered materials can be integrated with an array of materials of different dimensionalities to form mixed-dimensional van der Waals heterostructures without the constraints of lattice matching and processing compatibility.³ Very recently, the integration of 2D materials with 3D bulk semiconductors offers promising platforms for heterojunction device applications, which allows the combination of novel properties of 2D materials with the proven merits of well-developed 3D semiconductors.⁴
Here we have demonstrated highly gate-tunable heterojunction diodes using van der Waals assembly of 2D layered semiconductors (MoS₂, WSe₂ and InSe) and 3D bulk semiconductors (silicon and gallium nitride). The vertical stacking of these semiconductors forms a heterojunction with charge transport behavior that can be fully modulated by an applied gate electrostatic field. We observe gate-tunable current rectification with forward-to-reverse bias current ratios of up to 10⁷. More importantly, when operated as three-terminal devices, the 2D/3D heterojunction diodes exhibit a unipolar n-type transistor behavior with current on/off ratios exceeding 10⁷ sufficient for digital logic circuitry applications. These devices also deliver a large current density exceeding ~ 130A/cm² which is attributed to ohmic metal contacts as well as a conductive 2D/3D junction. Our devices present new avenues for future nanoscale electronic and optoelectronic devices for both digital logic and radio frequency applications.
References
1. Liu, Y. et al. Nature Reviews Materials 2016, 1, 16042.
2. Chhowalla, M. et al. Nature Reviews Materials 2016, 1, 16052.
3. Jariwala, D. et al. Nature Materials 2016, 16, 170-181.
4. Yang, H. et al. Science 2012, 336, 1140-1143.

The electrical and optical properties of two-dimensional (2D) crystals can be strongly influenced by strain and therefore potentially useful for electronic and optoelectronic applications. Another advantage of strain engineering in 2D crystals is the unique capacity of 2D crystals to withstand extreme strain (e.g. >10%). For application such as flexible electronics with functionality controlled by strain, a gate dielectric that can be deposited at low temperature, achieve large gate capacitance (e.g., 1~4 µF/cm²), and induce strain in the 2D materials would be desirable. In this work, we propose a new approach to induce strain via field effect modulation of a single-ion conductor (i.e., ionomer). Different from many other approaches to induce strain via global bending of the entire substrate, in our approach the strain is induced locally in each device using a voltage applied to the side gate of an electric double layer (EDL) gated, field effect transistor (FET). The anions in the single ion conductor are covalently bound to the backbone of the polymer, leaving only the cations free to move in response to an applied field. A electric double layer (EDL) is created at the negative electrode, but not at the positive electrode – instead a cation depletion layer exist. In response to this electrostatic imbalance, the single-ion conductor deform which could induce strain in the 2D material that’s in adjacent to it. The mechanism to induce strain is the same mechanism responsible for actuation in ionic polymer metal composites (IPMCs), which mimic the behavior of muscle. Modeling indicates cation densities up to 1.5 x 10¹⁴ ions/cm² at the single-ion conductor/electrode interface under 1V applied bias, which is theoretically predicted to induce several percent strain on the 2D crystal – enough to induce the phase transition in 2D transition metal dichalcogenides such as Mo- and W-dichalcogenide compounds. Experimentally, a custom synthesized ionomer is used to electrostatically gate both graphene and MoTe₂ FETs. Compared with backgating through SiO₂, transfer characteristics on both EDL gated graphene and MoTe₂ FETs reveal a similar enhancement of the n branch using both the single and a dual -ion conductor (i.e., one with mobile cations and anions). The single-ion conductor decreases the subthreshold swing in the n branch of the MoTe₂ FET from 5000 to 250 mV/dec and increases the current density and on/off ratio by two orders of magnitude. However, as expected, the ionomer quenches the p branches in both the graphene and the MoTe2 FETs, which the modeling reveals is a unique behavior of single-ion conductor gating with an asymmetrical gate to channel geometry. This is the first demonstration of an single-ion conductor gated 2D FET and it lays the groundwork for demonstrating the phase transition to create a steep subthreshold swing device as well as for flexible electronics applications. EDL response speed will also be discussed, with modeling and experimental data supporting nanosecond response times by aggressive scaling of the device geometry.

This work is supported by the NSF DMR under grant #1607935.

Two dimensional (2D) van der Waals (vdW) semiconductors have recently emerged as next-generation channel materials for field effect transistors (FETs) owing to their atomically flat surface, improved electrostatic gate control and enhanced scalability [1]. Theoretical calculations have shown that elemental vdW crystals such as 2D black phosphorous (BP) and 2D arsenic can show high mobilities, layer dependent tunable bandgap along with extremely light anisotropic carrier effective masses [2]. This makes them better candidates for FETs than transition metal dichalcogenides, which are marked with stochiometric defects and much lowered mobilities [3]. BP has already been effectively utilized to fabricate high performance MOSFETs and tunneling FETs with steep subthreshold swings, high on/off ratios and low off current [4,5]. However, for practical use, deterioration under the ambient environmental conditions is still a major bottleneck that needs to be overcome before BP based electronic devices can be realized. 2D arsenic - isostructural to BP, is much more stable in ambient conditions [6]. Moreover, recent experiments have shown that 2D arsenic has superior or comparable electronic and thermal properties than other 2D materials due to the presence of highly anisotropic effective masses along the two crystal directions [6,7]. Due to its low bandgap, 2D arsenic should show ambipolar electrical conduction, however, in the literature only p-type conduction has been reported [6]. In this work, we present the first ever demonstration of ambipolar 2D arsenic FETs and analyze the device performance and transport for both electrons and holes in this material.
Flakes (exfoliated from crystals obtained from a commercial vendor) with varying thicknesses were used to fabricate a series of devices on a 300-nm-thick SiO₂ dielectric grown on an n⁺ Si wafer. Electron dispersive spectroscopy (EDS) mapping was used to verify the composition of the material and the results showed that the obtained material was > 99% pure arsenic. Angle resolved polarized Raman spectroscopy revealed a two-fold symmetry for A¹_g (220.1 cm^-1) and A²_g (253.62 cm^-1) modes (with maximum and minimum intensities along armchair and zigzag direction respectively) further establishing the presence of the anisotropic puckered crystal structure [8]. In FET devices, it was found that the subthreshold slope and off-state leakage current improved with decrease in thickness or decrease in source/drain voltage bias which is indicative of the layer dependent bandgap of 2D arsenic. At room temperature, on/off ratios as high as 2.5 x 10⁵ and 3 x 10² were obtained for hole and electron conduction, respectively, with respective mobility values (for thickest to thinnest samples) ranging from 150 to 40 cm²/Vs for holes and 175 to 10 cm²/Vs for electrons. For thick flakes, hole and electron drive currents obtained were 30 μA/μm and 10 μA/μm respectively at |V_DS| = 1 V. Low temperature transfer and output characteristics were taken, and a thermionic transport model was used to extract the Schottky barrier height at the metal/semiconductor contact. The tunability of the barrier height for both electron and hole transport was further explored by changing the applied backgate bias.
This work was supported by the NSF (Award # ECCS-1708769). Portions of this work also used shared facilities at the University of Minnesota, supported through the NSF NNCI (Award # ECCS-1542202) and MRSEC (Award # DMR-1420013) programs.
[1] M. Chhowalla, et al., Nat. Rev. Mater., 2016, 1, 16052; [2] F. Xia, et al, Nat. Commun., 2014, 5, 4458; [3] Z. Yu, et al., Adv. Func. Mater., 2017, 27, 1604093; [4] N. Haratipour, et al., IEEE Elect. Dev. Lett., 2016, 37, 103-106; [5] M. C. Robbins, et al., IEDM Tech. Dig., 2017, 15.7.1-15.7.4; [6] Y. B. Chen, et al., Adv. Mater. 2018, 30, 1800754; [7] M. Zhong, et al., Adv. Funct. Mater., 2018, 28, 1802581; [8] A. Kandemir, et al., J. Mater. Chem. C, 2019, 7, 1228-1236.

Doping and heterostructuring lie at the heart of modern semiconductor technologies, which allow for the control over the functionalities of semiconductors, and the formation of junctions that make semiconductor devices operational. Recent studies have indicated that, by substitutional doping of semiconducting 2D TMDs materials with judicious selection of dopants, a series of their physicochemical properties can be effectively modified. Furthermore, laterally stitching TMDs with dissimilar compositions enables the formation of atomically thin in-plane heterojunctions, endowing them with great potential for electronic and optoelectronic applications. In light of this, it is desired to develop a reliable doping method for TMDs that can be potentially suitable for a wide range of dopant selections, as well as an efficient route for creating in-plane heterostructures between two dissimilar TMDs. Herein, to solve this challenge, a reliable in situ substitution doping method for monolayer TMDs is reported. Specifically, a solution precursor-based chemical vapor deposition (CVD) technique is developed. By varying the composition of precursor solutions, it is demonstrated that iron (Fe)-doped WS₂ and rhenium (Re)-doped MoS₂ can be achieved by the developed technique. In addition, monolayer in-plane W_xMo_1-xS₂-Mo_xW_1-xS₂ heterostructures with molybdenum (Mo)-rich inner regions and tungsten (W)-rich outer regions is also successfully synthesized. This doping and heterostructuring method may also be extended for the incorporation of a variety of other elements into 2D TMDs. This work shed light on the design and control of the functionalities of 2D TMDs, enabling novel applications based on these 2D materials and heterojunctions.

Black Ti₂O₃ is one of titanium oxide compounds with very stable properties in air, bio, and space environments. It also shows superconductivity in the high quality thin films which can be used for the quantum technology. High quality black Ti₂O₃ layers were successfully grown using Cu-Ti very low solubility phase and fast cooling approach from near the Cu meting temperature. Here we present the properties of ultrathin black Ti₂O₃ layers on SiO₂ substrate in large scale for solid state device applications.

Strikingly black Ti₂O₃ nanoparticles are known to exhibit narrow-band gap property compared to wide-band gap TiO₂ materials, which is suitable for photothermal applications as reported by T. Wu’s group. It was one of long seeking materials for a complement and alternative material of TiO₂, which can absorb sun light beyond UV range. Mao’s group also found that disordered black TiO₂ covered around TiO₂ nanopaticles, which later turned out to be as black Ti₂O₃, effectively increase solar absorption ability. However it is a difficult task to apply Ti₂O₃ nanoparticles to solid state device applications. Wafer level Ti₂O₃ thin films are required in the standard fab processing for cost effective and mass production of sensor and bio detection devices as well as photothermal conversion applications.

Wafer size Ti₂O₃ layers of about 10 nm thickness were grown using conventional thermal furnace by employing the concept of very low soluble property of Ti-Cu layers. Raman spectra with very narrow width at 274 cm^-1 in the samples grown above the growth temperatures of 1000-1090 Celsius degree indicate that the Ti₂O₃ layers are highly ordered and PL spectra show broad weak peaks near 500 nm in addition to the strong signals at near 574 nm at room temperature which is commonly observed in different forms of TiO₂.

We started from the Ti(20nm)/Cu(600nm) template deposited on SiO₂/Si wafer using sputtering evaporation method at room temperature. It is known that the Ti-Cu system has very low solubility which is difficult to make of alloy. The maximum Ti solubility in Cu is about 8% at 885 Celsius degree that can be found in the phase diagram literature. It is well known that Cu layers with less than about 1000 nm thickness dominantly form (111) crystal phase due to low surface free energy. The deposited ultrathin Ti layers are randomly arranged in a few Ti atomic layers on Cu(111). Then a few Ti atomic layers were exposed with very small amount of O₂ at high temperature and subsequently quenching under H₂ environments. This lead to the formation of Ti₂O₃ layers because of more oxidation affinity between oxygen and Ti. The Ti-Cu system under fast cooling with a little O₂ make the most stable Ti₂O₃ surface layers removing any other TiO₂ phase structures. Actually more O₂ with slow cooling lead to anatase TiO₂ layers.

The grown Ti₂O₃ layers showed the diffraction patterns of (012), (104), (110), and (113) planes from grazing incidence of X-ray diffraction measurements, which are well agreed with Ref. [1] and [2]. Detailed atomic crystal structure was also confirmed in TEM measurements. Raman spectra with very narrow width at 274 cm^-1 in the samples indicate high quality of the crystals. Raman peaks of any other TiO₂ structure phases are completely suppressed. XPS and AFM results will be presented.

References:

[1] X. Chen, L. Liu, P. Y. Yu, S.S. Mao, Science. 331, 746 (2011).

[2] J. Wang, et al., Advanced Materials. 29, 1603736 (2017).

Two-dimensional transition metal dicalchogenides (2D-TMDs) are an outstanding class of compounds which have several fields of applications, such as: catalysis, photonics, electronic devices, sensing and energy storage. 2D-molybdenum disulfide (2D-MoS₂) is one of the most studied TMD due to its electronic, magnetic, optical and mechanical properties. For instance, 2D-MoS₂ has a high Young’s modulus (>270GPa) and low pre-tension values that makes it a great candidate for application as a nanofiller with polymeric composites. Besides the investigations on the properties of 2D-MoS₂, a major focus in recent research lies in the development of experimental methods to produce high yield nanocomposites associated with high quality nanosheets. There are several methodologies to obtain 2D-MoS₂, such as: CVD, mechanical exfoliation and ionic intercalation. One interesting approach is use liquid exfoliation and ultrasonication to obtain, in a single-step, 2D-MoS₂ nanosheets functionalized with a polymer (e.g. polybutadiene)[1]. The C-S bond on the edges of the sheets of the composite provides chemical stabilization of the 2D structure during exfoliation increasing the yield of the reaction.
In this study, we propose a novel and scalable single step synthetic method to exfoliate functionalize and polymerize MoS₂-polystyrene (MoS₂-PS) nanocomposites starting from bulk MoS₂ and the monomer (styrene). In this case, the vial containing styrene, bulk MoS₂, NMP and AIBN was introduced in the ultrasonic bath and kept for 6 days at 70°C. For a comparative study, the same experiment was carried out, but toluene was added in the mixture above. After this time, the colloidal dispersion was centrifuged at 5000rpm for 1hr and the greenish supernatant was collected for further analysis. Moreover, pure PS was synthesized with AIBN in the ultrasonic bath at 70°C.
Thin film of the MoS₂-PS was made by casting in a wafer of monocrystalline silicon. Raman spectroscopy showed the characteristics peaks of polystyrene-, C-H bands at 3000cm^-1, C-C bands at 800cm^-1, C=C bands of the aromatic ring at 1600cm^-1- indicating polymerization. The A_1g and E¹_2g bands of MoS₂ can be located at 383,9cm^-1 and 409,1, respectively, with a frequency distance of 25,2cm^-1, whilst 26,5cm^-1 value is expected for a bulk MoS₂, indicating high degree of exfoliation and the peak in 630cm^-1 represents the C-S bond confirming the functionalization of the nanocomposite. XRD of the MoS₂-PS composites showed a broad basal plane (002) peak confirming the reduction of the initial size of bulk MoS₂. Also, it is possible to notice the absence of other peaks that indicates exfoliated material except for the (103) and (105) planes due to restacking.TGA allowed quantifying the amount of 2D-MoS₂ in the film of the nanocomposite as well as to verify the thermal stability compared to the pure PS film. In this case, TGA showed a 2,33% concentration of bounded 2D-MoS₂ in the PS film and the initial degradation temperature was 185°C for the MoS₂-PS whilst 162°C was reported for pure PS. DSC showed a shift in the glass transition temperature (Tg) from 148°C of the pure PS film to 162°C in MoS₂-PS.The MoS₂ exfoliated with toluene/NMP/styrene/AIBN was not able to form thin films indicating a low degree of polymerization which can be confirmed by a 0,52% weight loss of organic matrix in TGA and showed all characteristics of crystallographic peaks of bulk MoS₂ as well as the Raman modes, confirming the low degree of exfoliation of the MoS₂. The yield of the MoS₂-PS reaction was 50% for polymerization reaction and 11% for exfoliation. Therefore, this approach opens opportunities of new scalable single step synthetic routes with high degree of exfoliation. We thank FAPESP (CDMF proc. 2013/07296-2) for the financial support (2016/20493-0) CNPq and CAPES.
[1] GONÇALVES, R. H.; FIEL, R.; SOARES, M. R. S.; SCHREINER, W. H.; SILVA, C. M. P.; LEITE, E. R.–Chem. Eur.J., v. 21, p. 15583-15588, 2015.

Weak van der Waals (vdW) interactions between the stacking layers of transition metal dichalcogenides (TMDCs) accommodate intercalation of various species from monovalent alkali, divalent alkaline earth and multivalent metal ions to organic molecules. The research over the past decades have revealed that intercalation of foreign species in the gap can drastically alter the physical properties of layered materials. Especially, controlled electrochemical intercalation can be used as a means to tune the electronic and optoelectronic properties. Among various TMDCs, MoS₂ is an important semiconductor with promising field-effect transistor (FET) applications. 2H to 1T' phase transition via chemical and electrochemical intercalations has previously been reported. However, many aspects of intercalation effects are still unclear and require further research. Here, the effects of alkali metal electrochemical intercalation on the transport properties of few-layered MoS₂ are presented. The intercalation was performed in top-contact and h-BN encapsulated edge-contact devices, which exhibit distinct behaviors. Additionally, careful magneto-transport studies were carried out on those intercalated MoS₂ flakes at various Li ion concentrations to investigate the effects of intercalants on the scattering mechanism and electron doping of MoS₂ flakes.

Quasi-2D materials have recently gathered significant interest since the discovery and application of transition metal dichalcogenides (TMDCs). Subsequently, the hunt for new families of quasi-2D materials has led us to group-IV monochalcogenides. In addition to having structures very similar to black phosphorus, they have the added advantage of showing much higher carrier mobilities. In this study, we use density functional theory (DFT) to perform structural relaxation on group-IV monochalcogenides of the form AX (where A: C, Si, Ge, Sn, Pb; X: S, Se, Te) each in 6 different prototype van der Waals-layered quasi-2D structures in an attempt to find new quasi-2D group-IV monochalcogenides. We then scan the structures that remain quasi-2D after relaxation for dynamic stability. For the dynamically stable structures, we compute the piezoelectric tensor values (e_ij). We find a total of 20 quasi-2D, dynamically stable, and piezoelectrically active structures amongst the 90 structures that are studied here. In addition to helping similar future computational searches, our study also provides guidance to experimental efforts to synthesize new quasi-2D piezoelectric materials.

Graphene nanoribbons (GNRs) are promising candidates for semiconductor logic device applications. Lithographic edge roughness limits the utility of top-down fabricated GNRs due to carrier scattering effects. Bottom-up fabricated atomically precise GNRs can circumvent this issue. We are studying atomically precise GNRs that can be fabricated on the gram scale through wet-chemical synthesis methodology [1]. A fascinating aspect of the bottom-up GNR synthesis is the ability to tailor the atomic structure of the GNR through the choice of the starting molecules. GNR edge structure and width can be controlled with atomic precision. An unexplored class of GNRs contain atomically precise built-in pores. These GNRs contain multiple pores in a single ribbon and the electronic details of the ribbon have not been reported. In this work, we use dry contact transfer (DCT) to deposit porous GNRs onto hydrogen passivated Si(100) surfaces in UHV. These GNRs have been characterized by UHV scanning tunneling microscopy (STM) and spectroscopy (STS). STM imaging confirms the expected porous structure and indicates a unique electronic feature at the graphene nanopores. Intuition would suggest nanopores appearing as valleys due to the missing carbon atoms. However, STM indicates an electronically tall peak at the pore sites. STS measurements indicate a 1.62 ± 0.11 eV bandgap in the bulk GNR and a 2.50 ±0.02 eV bandgap at the pores. These results are compared to first-principles density functional theory (DFT) simulations in which an increased local density of states at the pores is predicted. GW correction shows a 2.89 eV bandgap. Results will also be reported for GNRs with nitrogen-sulfur-nitrogen (NSN) functionalizations. These GNRs are new chemically synthesized variations for which the fabrication procedure and electronic details have not yet been reported.

References:
[1] Vo, T. H., Shekhirev, M, Kunkel, D. A., Morton, M. D., Berglund, E., Kong, L., Wilson, P. M., Dowben, P. A., Enders, A., and Sinitskii, A., Nat. Commun. 2014, 5, 3189.

Single photon emitters in hexagonal boron nitride (hBN) have shown promising electrical and optical properties for utilization in quantum light source applications, owing to the ability of emitters to show Fourier transform-limited zero phonon line emission, which is crucial in applications in which sources of indistinguishable photons are required. Here we investigate the zero phonon line broadening mechanisms of emitters located in 5-10nm thick CVD-grown hBN, films of which can be grown scalable over large areas as needed for monolithic integration of quantum emitters in future chip-based quantum photonic systems. By comparing the lifetime-limited homogeneously broadened linewidth obtained from luminescence decay lifetime measured by time-resolved photoluminescence and low temperature spectroscopy of photoluminescence we observe that even at 4K, the linewidth of the zero phonon line is broader than the Fourier limited value by 2 orders of magnitude. Furthermore, we investigate sources of ZPL broadening, such as spectral diffusion (inhomogeneous) and thermal broadening (homogeneous) by performing spectroscopy and analyzing the Voigt profile at various temperatures (4K-300K). Our studies show at 4K the main broadening source is inhomogeneous but at higher temperatures homogenous broadening mechanisms (such as thermal broadening) are more dominant. By comparing the emitter luminescence linewidth for hBN films on several different substrates and excited at varying pump intensities, we can conclude that the main broadening mechanism is spectral diffusion at low temperature. Origins of this phenomenon and possible mitigation strategies will be discussed.

While conventional epitaxy produces high-quality epitaxial layers, the limited set of commercially available wafers restricts the number of material choice that can be grown through epitaxy. To extend the set of epi-materials, heteroepitaxy has been widely used although the epi-materials have severe lattice mismatch from wafers. The problem of heteroepitaxy of the lattice mismatched systems is that introducing dislocations is unavoidable because of the strain energy built up in epitaxial layers exceeds the threshold energy value to create dislocation. As the dislocations significantly reduce the materials’ properties, people have put intensive effort into finding a new way to resolve this issue.
Here, we introduce a new strain relaxation approach which can efficiently engineer the lattice with minimizing dislocation formation. Remote epitaxy allows single-crystalline growths on graphene-coated substrates as atomic potential field guides atomic registry through graphene while weakening binding energy of epilayer to the substrate. We discovered that this weakened interface offers a new pathway for strain relaxation during epitaxy in the lattice mismatched systems. While conventional heteroepitaxy relaxes accumulated strain energy through involving misfit dislocations, heteroepitaxy on graphene-coated substrates spontaneously relaxes its strain minimizing dislocation density. Thus, at the same thickness and misfit strain, the epilayer grown on graphene-coated substrates shows full relaxation with substantially reduced dislocations while that by conventional heteroepitaxy exhibits slight relaxation with full of dislocations. It means that full of strain relaxation at much lower thickness is available through this approach. We strongly believe that our strategy allows broadening range of available semiconductors through epitaxy, which could lead high performance electronic and photonics.

Black phosphorus (BP) has shown significant promise for use in infrared photodetectors due to its high carrier mobility, tunable bandgap and anisotropic properties. To date, characterization of BP detectors has been limited to the visible and NIR, or single IR wavelengths. We will present results on the optical characterization of black phosphorus photodetectors across the entire spectral range where photoresponse occurs. These devices show broadband photodetection from <400 nm to the nearly 4 um bandgap. This full spectral characterization allows us to establish a sharp contrast between the visible and infrared behavior. In the visible, a responsivity of >6 A/W can be obtained due to a large photoconductive gain, while the infrared responsivity is nearly independent of gate voltage and incident light intensity under most conditions. We attribute this to a photogating contribution from the surface oxide. Furthermore, we observe that the polarization anisotropy in responsivity along armchair and zigzag directions can be as large as 10³ and extends from the band edge to 500 nm. This anisotropy is much larger than that in previous reports and the result of measuring in an intensity regime where photoconductive gain is absent. This opens up significant opportunity for a number of applications including IR polarimetry. The devices were fabricated in an inert atmosphere and encapsulated by Al₂O₃ providing stable operation for more than 6 months.

Exfoliated Transition Metal Dichalcogenides (TMDs, MX₂) have attracted considerable attention for infrared optical elements due to their high refractive index and extreme nonlinearities (e.g. MoS₂, TiS₂). Multiphoton absorption, nonlinear scattering, and nonlinear refraction are useful in technologies from reverse saturable absorption to optical limiting and gradient index coatings. Optimizing performance necessitates fabrication of dispersions, composites, and nanolaminates with high monolayer content, robust chemical stability, and high optical transmission. Energy intensive sonication methods to exfoliated TMDs limit scalable strategies to high quality films and coatings, due to poor oxidative stability, low concentration of TMDs, and a small choice of stabilizing solvents. The recently developed sonication-free, redox exfoliation approach address these challenges by drastically reducing reaction times for gram scale exfoliation of the entire TMD material space (Group IV – VII). This process provides TMDs dispersions in polar organic solvents (acetonitrile, acetone, alcohols) at high concentration (10% w/w) and with high monolayer content. The expansion of solvents and lack of additional stabilizers enables direct hybridization of Group VI 2H-MX₂ TMDs (e.g. MoX₂) via alkyl organometallic chemistry, as well as reduction of Group V 2H-MX₂ (e.g. VS₂, TaS₂, NbX₂) via amines and phosphines. This hybridization increases oxidative stability and tunes optoelectronic properties, including absorption and nonlinear response. The resulting access to surfactant free, monolayer populations enable design and fabrication of optical filters, GRIN optics, and non-linear absorbers.

Layered metal carbides and nitrides (MXenes) have generated substantial research interest since their discovery in 2011. Their production requires selective etching of the A layer (A = Si, Al) from bulk MAX phases, yielding a new class of layered ceramics with excellent conductivity. HF is commonly used to selectively remove the A layer; however, its acute toxicity prevents safe, scalable MXene production. The complex, environmentally-sensitive reaction mechanism, as well as degradation processes in water and oxygen, challenges MXene yield optimization and compositional control of surface termination. On top of batch-to-batch heterogeneity, this leads to mixed surface termination (-F and -OH) that limits modification of the surface and layer properties. Here, we present an alternative method to fabricate MXenes that utilizes halogens in oxygen-free non-polar solvents for selective removal of the A layer. The etch products are stabilized and dispersed in non-polar solvents (THF, CHCl₃, DCM), leaving oxidatively robust, homogeneously halogenated surfaces (Cl, Br, I). Combined, this enables access to new chemistries for surface derivatization (organometallics), which result in uniform and tunable optical, electrical and chemical properties. The process is safe, scalable, and easily transferred across the MAX material space, giving access to previously un-explored MXene’s. The development of such alternative routes for etching and delamination are crucial for large-scale manufacturing for paints, coatings, and EMI shielding applications.

Owing to their low dimensionality, two-dimensional semiconductors, such as monolayer molybdenum disulfide, have a range of properties that make them valuable in the development of nanoelectronics. For example, the electronic bandgap of these semiconductors is not an intrinsic physical parameter and can be engineered by manipulating the dielectric environment around the monolayer. Here we show that this dielectric-dependent electronic bandgap can be used to engineer a lateral heterojunction within a homogeneous MoS₂ monolayer. We visualize the heterostructure with Kelvin probe force microscopy and examine its influence on electrical transport experimentally and theoretically. We observe a lateral heterojunction with an approximately 90 meV band offset due to the differing degrees of bandgap renormalization of monolayer MoS₂ when it is placed on a substrate in which one segment is made from an amorphous fluoropolymer (Cytop) and another segment is made of hexagonal boron nitride. This heterostructure leads to a diode-like electrical transport with a strong asymmetric behaviour.

The persistent down-scaling of nanostructures, such as electronic devices, sensors, NEMS, and nanocomposites, increases the surface-to-volume ratio and introduces atomic-scale disorder at boundaries and interfaces. To avoid these issues, the nanoelectronics community has turned to intrinsically 2D materials platforms such as transition metal dichalcogenides (TMDCs), where atomic flatness and the absence of dangling bonds prevent scattering due to surface roughness (SR), which severely limits mobility in ultra-thin body 3D silicon-on-insulator field-effect transistors (FETs). Despite immunity to SR scattering, the mobility of single layer TMDs are largely deteriorated by charge impurity scattering from the substrate [1]. Researchers have demonstrated that encapsulation of TMDs with hexagonal boron nitride significantly improves mobility [2]; however, encapsulation is expensive and time-intensive. Another route to improving mobility and boosting drive current is to replace the monolayer with few-layered (FL) TMDCs. Despite recent advances on demonstrating improved electrical performance of FL TMDC FETs [3], there has been less attention towards their thermal management, which is crucial for modern nanoelectronic devices.
In this work, we develop a coupled electro-thermal model for FL-WSe₂ stacks where we simultaneously solve for the current and Joule heating by treating the stack a resistor network. The resulting rise in temperature is obtained from a FL-TBC model [4] to shed light on self-heating and heat dissipation in such devices. We found that the distribution of current and Joule heating across layers in a FL-stack is highly non-uniform and strongly dependent on the back-gate voltage (V_g), in agreement with previous studies on FL graphene [5] and MoS₂ [6]. Interestingly, we found a similar behavior with respect to the drain-source voltage (V_DS) as well. For a given V_g, the current mostly flows through the top layers for small V_DS. But when V_DS is increased, the current penetrates deeper into the stack and flows mostly through the bottom layers. We attribute the change to self-heating of these devices at large V_DS. We find that the temperature rise in the top layers is significantly larger than the bottom layers because the bottom layers have higher TBC and conduct heat more efficiently to the substrate [4,7]. The higher temperature of top layers, in turn, significantly reduces their mobility, which is strongly temperature-dependent as they are shielded by the bottom layers from substrate impurity scattering. Consequently, both the current and the dissipation hotspot move towards the bottom layers with increasing V_DS. We also uncover that, unlike monolayer FETs, a significant amount of heat is dissipated laterally through the contacts in FL devices at high V_DS due to relatively large thermal healing lengths of the top layers.
In conclusion, we have developed a robust and detailed coupled electro-thermal model to solve for the non-uniform and layer-dependent temperature rise in response to Joule heating. Our findings help develop a deeper understanding of dissipation and thermal management in FL 2D materials and vertical heterostructures.
REFERENCES: [1] Ma and Jena, Phys. Rev. X 4, 011043 (2014) [2] Lee et al., ACS Nano 9, 7019-7026 (2015) [3] Pradhan et al., Sci. Rep. 5, 8979 (2015) [4] Behranginia et al., Appl. Mater. Interfaces 10, 24892-24898 (2018) [5] Sui and Appenzeller, Nano Lett. 9, 2973-2977 (2009) [6] Das and Appenzeller, Nano Lett. 13, 3396-3402 (2013) [7] Yasaei et al., Adv. Mater. 30, 1801629, 2018

We will present our results large-area, atomically-thin metal films (Pt, Pd and Cu) can be grown epitaxially on graphene (GR) using electrodeposition. We will focus particularly on Pt films that are one to several multilayers thick (Pt_ML) epitaxially grown on graphene (Pt_ML/GR). These Pt_ML/GR 2D systems have covalent bonds at the Pt_ML/GR interface and this intimacy between the layers serves to make the GR a ‘chemically transparent’ barrier that allows catalytic chemistry to take place above it, while protecting the Pt below it from loss. We will specifically show that graphene does not restrict access of the reactants for the canonical oxygen reduction reaction (ORR) but does block Pt from dissolution or agglomeration. These architectures simultaneously achieve enhanced catalytic activity and unprecedented stability, retaining full activity for ORR beyond 1000 cycles. Using x-ray photoemission/absorption spectroscopy (XPS/XAS), high resolution TEM, AFM, Raman, and electrochemical methods, we show that Pt/GR hybrid architectures induce a compressive strain on the Pt films, thereby increasing their ORR activity. Our room-temperature, fully-wetted synthesis approach, should allow for efficient charge, strain, phonon and photon transfer, between the films and their support, impacting not just the performance of catalysts, but also those of electronic, thermoelectric and optical materials.

Doping is one of the main cornerstones of modern semiconductor technology. Recent re-discovery of 2D transition metal dichalcogenides (TMDs) attracted great interest in semiconductor doping in such low dimensional systems. Numerous efforts have been devoted to extrinsic doping of TMDs to tune optical, electronic and magnetic properties with various synthesis methods such as chemical vapor transport, powder vaporization, and plasma treatment. Establishing controlled and reliable doping techniques are indispensable for the realization of TMD based CMOS-type devices for next-generation electronic devices. However, all previously reported studies still lack a fundamental understanding of doping mechanisms and have not been extensively examined.
In this work, we present finely controlled, scalable Re doping of monolayer epitaxial WSe₂ films on c-plane sapphire and epitaxial graphene (EG) substrates via gas source metal organic chemical vapor deposition method. The MOCVD synthesis is carried out in a vertical cold-wall CVD reactor at 800 °C using hydrogen selenide (H₂Se), tungsten hexacarbonyl (W(CO)₆) and dirhenium decacarbonyl (Re₂(CO)₁₀) in a hydrogen gas atmosphere. By controlling the partial pressure of the (Re₂(CO)₁₀) in the growth chamber we demonstrate that Re dopant concentration in the epitaxial WSe₂ lattice can be tuned from 5% down to <1%. High-resolution X-ray photoelectron spectroscopy (XPS) is used to quantify Re concentration at high doping levels. Time of flight secondary ion mass spectroscopy (Tof-SIMS) is further performed to confirm a presence of Re atoms in WSe₂ at low doping concentrations. The high concentration (5%) of Re dopant atoms leads to a formation of ReSe₂ and phase segregation from WSe₂; however, the undesired phase segregation is eliminated by reducing the Re flux during the growth which is further confirmed by Raman and photoelectron spectroscopies. Back-gated field effect transistors are constructed and demonstrate a clear trend as a function of Re concentration in WSe₂ from ambipolar conduction to degenerately n-type doping. Currently, high-resolution transmission electron spectroscopy, selective area electron diffraction, electron energy loss spectroscopy, and scanning tunneling microscopy studies are under investigation to further elucidate the effect of in-situ Re doping on atomic scale structural and electronic properties of epitaxial WSe_2.

MXene (Ti₃C₂T_X) is a newly discovered family of 2D carbide material with superior electrical and electrochemical properties. Benefiting from its metal-like conductivity (6500 S/m), high volumetric capacitance (900 F/cm³) and stable dispersion in various solutions, MXene has been successfully adopted in various electronic and energy applications such as supercapacitors, batteries and electromagnetic shielding. However, the incorporation of MXene materials into emerging stretchable electronics and energy storage devices remained unexplored possibly due to their large mechanical modulus, low ultimate strain, and weak sheet interactions. The current study investigates the design, fabrication, and characterization of robust and highly stretchable MXene/RGO composite supercapacitors through wavy-shape design. The effect of RGO concentration is systematically studied.
In the fabrication of stretchable MXene electrodes, it was found that cracks formed during the prestrain-relax process leading to significant degradation of electrochemical performance of stretchable MXene electrode in the fully stretched state. To overcome this issue, Reduced Graphene Oxide (RGO) was added into the system to tune the mechanical properties of the MXene/RGO composite paper because of its lower mechanical modulus and stronger sheet interaction. It was found that MXene/RGO composite paper could maintain its mechanical integrity when the weight ratio of RGO increased to 50%. Mechanical strain tests revealed that elongation of MXene/RGO increased with RGO concentration and remained at a similar level when the RGO concentration was larger than 50%. The as-fabricated stretchable MXene/RGO composite paper (50% weight ratio) can maintain consistent resistance for more than 1000 stretch-relax cycles with uniaxial strains up to 250% which to our knowledge is the highest stretchability reported for a MXene composite. When tested as supercapacitor electrodes, the stretchable MXene/RGO (50% weight ratio) composite paper demonstrated high and identical specific capacitance in different strain states (e.g., 0%, 100%, 200% and 300%) for various charge/discharge rates. Symmetric supercapacitors made of MXene electrodes and PVA-H₂SO₄ gel electrolyte exhibit robust electrochemical performance in agreement with the electrode measurements.
The present study demonstrates a facile strategy to fabricate highly stretchable MXene/RGO composite supercapacitors. These results demonstrate the prospects of a robust, high performance and highly stretchable MXene composite in emerging applications such as wearable electronics, bioimplants, and stretchable electronics.

Magnetism in graphene and other 2D materials has long been a holy grail for modern spintronics. The conventional routes towards inducing magnetism in graphene involve either functionalization of graphene with ferromagnetic adatoms or irradiation to generate vacancies and correlated defects. In this work, we provide direct experimental evidence for the existence of localized magnetic moments at the grain boundaries (GBs) of graphene, formed as a natural consequence of the chemical vapour deposition growth. The spontaneous breaking of time reversal symmetry at the graphene GBs is observed using low frequency 1/f noise arising from universal conductance fluctuations (UCF) via the ergodic hypothesis. The gate-tunable magnetism is attributed to the formation of localized magnetic moments at the octagon-pentagon defects of the disordered GB region, further enhanced by the inbuilt lattice strain leading to the dephasing of spins.

Both thermal and wet bottom-up chemical synthesis of graphene nanoribbons (GNRs) have been shown to create ribbons with atomic precision in widths and edges. This precision is essential in maintaining band-structure uniformity across the structure. However, both these recipes for GNR synthesis result in a stochastic distribution of ribbon length, location, and orientation, when studied on a substrate. This random distribution on the surface makes it impractical for devices to be made from these GNRs as there is no feasible scheme to attach metal contacts to the ribbons. To address this challenge, we utilize dry-contact-transfer of bromoaromatic precursors to a Au(111) surface and, through the use of electron tunneling currents induced by ultra-high vacuum scanning tunneling microscopy (UHV-STM), spatially control the precursor polymerization. Here we demonstrate that at sufficiently high scanning biases, 10,10′-dibromo-9,9′-bianthracene (DBBA) polymerizes into polyanthrylene only along the path that the STM tip traverses. These results introduce the feasibility of electron-tunneling controlled bromoaromatic monomer polymerization and, in the future, high degrees of spatial control over atomically precise GNR synthesis.

Black phosphorus (BP), the most stable allotrope of phosphorus, has attracted considerable attention in recent years due to its extraordinary electrical and optical properties, which makes it an excellent candidate for applications in transistors, photodetectors, and lithium-ion batteries.¹ BP, like graphite, has a layered structure and can be fabricated into monolayers and nanosheets from its bulk form through micromechanical and liquid exfoliation.² Mechanical exfoliation (the scotch-tape method) has been predominantly used for proof-of-concept devices,³ but it is inherently unscalable and typically limited to sheets with lateral dimensions below ten micrometers.
For practical applications, scalable approaches for fabricating BP large-area films are essential. We will present our results on the development of a Langmuir-Blodgett (LB) protocol which is developed for assembling BP nanosheets into thin films. Through functionalization, we increase the electrostatic force between BP nanosheets to avoid the overlapping of BP nanosheets during compression, which allows us to achieve large-area (centimeters), smooth, and compact BP thin films. In addition, this protocol is well-adapted to assembly thin and thick BP nanosheets as well as arsenic-BP alloy nanosheets.
The optoelectronic properties of transistors and photodiodes fabricated from the BP LB film are investigated. With suitable carrier transporting materials, The BP photodiodes exhibit a good optical response from visible to infrared region. Our work highlights the great potential of BP in applications of near- and middle- infrared detection and imaging.

1. Ling, X.; Wang, H.; Huang, S.; Xia, F.; Dresselhaus, M. S., The Renaissance of Black Phosphorus. Proc. Natl. Acad. Sci. USA 2015, 112 (15), 4523-30.
2. Lei, W.; Liu, G.; Zhang, J.; Liu, M., Black Phosphorus Nanostructures: Recent Advances in Hybridization, Doping and Functionalization. Chem. Soc. Rev. 2017, 46 (12), 3492-3509.
3. Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D.; Chen, X. H.; Zhang, Y., Black Phosphorus Field-effect Transistors. Nat. Nanotechnol. 2014, 9 (5), 372-7.

Black phosphorus (BP) has begun to gain attention in 2014, when its stripping has led to 2D layers of phosphorene. The promising properties of this material have, however, been severely limited by its high instability under environmental conditions which rapidly leads to its degradation. Thus, the interest in exploring BP has been growing as well as the concern in protecting it. In this sense, we have synthesized a BP-polymeric nanocomposite, aiming to overcome BP deterioration in a novel material with synergic and enhanced properties. For this purpose, different samples have been prepared through the following steps: 1. a liquid exfoliation of BP resulting in a stable dispersion (0.025 mg mL^-1 in deaerated acetonitrile); 2. aniline polymerization in the dispersion of BP, resulting in BP/polyaniline (BP/PANI) nanocomposites; 3. thin film deposition of the BP/PANI nanocomposites through the liquid/liquid interfacial route developed in our research group. Raman spectroscopy showed the coexistence of the two components modes in the BP-PANI spectra, confirming the successful synthesis of the nanocomposite, with the polymer in its conducting form (emeraldine salt). Infrared spectra attested the unstable nature of BP, exhibiting bands related to modes resulting from the phosphorus oxidation, and suggested the protective role of PANI with the absence of these bands for BP-PANI. Thus, the stability of both bare and capped material was evaluated through their exposure to the same ambient conditions and evaluated through different techniques. Scanning electron microscopy (SEM) images evidenced the beginning of degradation of the uncapped material with only 3 days of exposure, showing the characteristic bubbles on BP surface. The same was observed through Raman, indicated by the gradual decrease in the intensity of BP bands that completely disappear with its total degradation (at about 15 days). BP-PANI nanocomposites, however, have taken up to 60 days for the same process to begin, indicating a material 2000% more stable than the bare one.
After stablishing BP chemical stability, we evaluate the electrochemical properties of the nanocomposites targeting energy storage devices. Electrochemical response and stability of BP-PANI nanocomposites were evaluated through cyclic voltammetry (CV) and charge-discharge (CD) studies (NaCl 0.5 mol L^-1 pH 3), in order to evaluate energy storage properties. Films of BP-PANI were evaluated and compared to pure PANI and BP, under air and inert atm (N_2(g)). Bare BP is degraded right after the first cycle, evidencing the expected low stability. For the nanocomposites, BP has shown a high stability even after 200 CVs in the presence of O_2(g). BP integrity was confirmed through post-characterizations (Raman spectroscopy and SEM images). Finally, aiming to evaluate the performance of the materials as electrodes for sodium ions batteries, CD studies were performed. Current densities of 0.2 A g^-1 to 3 A g^-1 were applied and the capacities reached up to 120 mAh g^-1 at 0.5 A g^-1 rate. Overall, this work not only provides the synthesis of a novel nanocomposite, but also prints to an innovative route towards the development of transferable and stable thin films, suitable to build new BP-based devices operating in water and under ambient atmosphere. Authors acknowledge UFPR, CNPq, CAPES, INCT-Nanocarbon, PhosAgro/UNESCO/IUPAC, L’Oréal-UNESCO-ABC and FAPESP (SPEC project 2012/50259-8).

Molybdenum disulfide (MoS₂) is a representative material among transition-metal dichalcogenides (TMDCs). it has a direct bandgap of ~1.8 eV and advantages of high stability, fast electrical mobility, and high current on/off ratio compared to other TMDCs. The growth of MoS₂ is relatively simple and easy by using chemical vapor deposition (CVD) which is a conventional method to ensure large-area and high-quality. Metal-organic chemical vapor deposition (MOCVD) is one of the branches among various CVD methods. One of the advantages of MOCVD use the metal-organic precursors which can be easily decomposed at low temperature. Another advantage is precise control of flow rate between two or three precursors by using the mass flow controller (MFC) system. Here, we report that monolayer MoS₂ flake and film were grown on a various substrate (SiO₂, sapphire, soda-lime-glass, and gold film/SiO₂) by using MOCVD with the modified solid bubbler system to ensure a stable supply of Mo source without the heating process of solid Mo(CO)₆. The minimum growth temperature of MoS₂ is below 400 ^oC which allowed its growth on the gold film. Also, the shapes of MoS₂ flake can be modified through the control of the ratio of flow rate between Mo and S source. Finally, the maximum electron mobility of MoS₂ field-effect transistor showed about 45.4 cm²V^-1s^-1 due to reducing strain or interaction between MoS₂ and the substrate.

Lithium atoms intercalating between layers of graphene show an energetic preferece for regions of AA stacking. Thus the triangular array of AA regions in the moiré pattern of twisted bilayer graphene allows for control over the location of intercalants in layered materials and devices. We use density functional theory to directly calculate the energetics of lithium atoms between layers of graphene in commensurate supercells with twist angles of 7.3° and 2.5°. The two cells show universal behavior in the energy profile for single Li intercalants when the location of the intercalant is measured as a fraction of the distance between AA and AB regions. This behavior can be expected to persist down to angles of about 2.0°. For the 2.5° cell we demonstrate clustering of the Li atoms in the AA regions when the average Li concentration is relatively low. The Li intercalants cause a local increase in the interlayer separation as well as charge transfer to the nearest carbon atoms. These regions of localized charge hold the potential for interesting catalytic properties. The charge donated by the Li atoms also raises the Fermi level above the Dirac point and causes qualitative changes to the band structure, especially near the Γ-point. The combination of twisting with intercalation gives the added power of spatial control over the material properties that can be tuned by intercalation.

Graphene is a 2D material with diverse applications. However, there are important applications in gas and liquid media that can benefit from the unique properties of graphene but require a 3D structure able to interact thoroughly with the medium. Thereby, the systematic and reproducible fabrication 3D Graphene would enable further applications useful both for research and industrial processes, such as catalysis, energy storage, and environmental remediation. Thereby, we developed a method to fabricate 3D Graphene Foams using the Chemical Vapor Deposition technique, acetone as carbon source, and Ni foams as scaffolds. We studied how the structural properties of the 3D Graphene Foams depend on the growth parameters, such as temperature, precursor concentration, and Ni foam thickness. Through this study, we optimized the pore size and mechanical stability of the foams produced. We are also reporting on the interaction of these 3D Graphene Foams with representative gases and liquids to understand the effective surface area and surface terminations of these foams.

The various types of 2D heterostructure prototype devices based on graphene (G) and boron nitride nanosheets (BNNS) were fabricated to study the charge tunneling phenomenon. Specifically, G/BNNS/Metal, G/SiO₂, and G/BNNS/SiO₂ heterostructures were investigated under DC-bias conditions at room temperature. Bilayer graphene and BNNS samples were grown separately on copper foil and transferred subsequently between the substrates to fabricate 2D devices architecture. The high resolution transmission electron microscopy confirmed bilayer graphene structure and few layer BNNS sheets having an interpretable hexagonal B₃-N₃ lattice. The I(V) characterizations for G/BNNS/Metal device have shown typical Schottky barrier behavior with very low forward voltage drop due to out-of-plane tunneling and low sheet resistance of G layer. A theoretical model based on current tunneling is proposed to qualitatively describe hetero-2D G/BNNS/Metal devices behavior.

Janus type of two-dimensional chalcogenide monolayers are predicted to permanently polarize in the out-of-plane direction, due to the structural asymmetry therein. The aligned dipoles, which cause the spontaneous polarization, are experimentally difficult to identify as they are confined in a Van der Waals layer, unless through the detection of electromagnetic wave-dipole interaction. We measured the refraction index of Janus monolayer MoSSe in the range of visible light. Comparison of the pattern of Janus MoSSe with MoS2 and MoSe2 reveals the existence of the built-in dipoles in the Janus monolayer. The discovery indicates the great potential of Janus type of 2D chalcogenide monolayers in optical applications.

Tin disulfide (SnS₂) is a two dimensional (2D) van der Waals material with a moderate band gap of ~ 2.1 eV, strong optical absorption and high carrier mobility that make it attractive for solar energy conversion applications. We have recently demonstrated that nanostructuring SnS₂ in the form of vertically-aligned nanoflakes is beneficial to SnS₂ photoanode performance [1]. It balances ~ 1 micron film thickness necessary for absorbing incident light with the highly exposed steps and edges suitable for photoelectrochemical applications. However, nanostructuring comes at a cost of increased carrier trapping at surface and edge states. Here we use time-resolved THz spectroscopy (TRTS) to elucidate the effect of nanostructuring on photoexcited carrier lifetime and mobility in vertically-aligned SnS₂ nanoflake array synthesized using close space sublimation method.
SnS₂ vertical nanoflakes were excited with 400 nm, 100 fs laser pulses, and transient photoconductivity was measured using a terahertz probe in the 0.25 – 1.70 THz frequency (or, equivalently, 1-7 meV) range. To delineate the effect of edges and surfaces, we have compared transient THz conductivity of nanoflakes to that of SnS₂ single crystals grown using chemical vapor transport method. Photoconductivity in both single crystal SnS₂ and vertical SnS₂ nanoflakes exhibits a multi-exponential decay. The fastest components, ~2-3 ps and 10-15 ps, are much more pronounced in the nanoflakes, and are ascribed to the carrier-carrier scattering (former) and the carrier trapping at defect and edge states (latter). Finally, a much slower component, ~ 0.25-1.30 ns, corresponds to free carrier recombination, as confirmed by the time-resolved photoluminescence measurements. We find that the photoexcited carrier mobility in single crystal SnS₂ as high as ~ 800 cm²V^-1s^-1. In vertical nanoflakes, intrinsic (intra-flake) mobility is comparable to the single crystal value at 330 cm²V^-1s^-1, while the long-range (inter-flake) mobility is reduces to 90 cm²V^-1s^-1 by the confining effect of nanoflake surfaces and edges. Based on experimental free carrier mobility and lifetime, we estimate the inter-flake photoexcited carrier diffusion length to be ~1 μm, in agreement with the nanoflake length that results in the best photoelectrochemical device performance.
K.K. and L.V.T acknowledge support from NSF DMR-1750944.
[1] B. Giri, M. Masroor, T.Yan, K. Kushnir, A. D. Carl, C. Doiron, H. Zhang, Y. Zhao, A. McClelland, G.A. Tompsett, D. Wang, R.L. Grimm, L.V. Titova, P.M. Rao, “Balancing Light Absorption and Charge Transport in Vertical SnS2 Nanoflake Photoanodes with Stepped Layers and Large Intrinsic Mobility,” submitted.

Typically, MXene polymer composites have been limited to small scales and polymers that are soluble in aqueous solutions. This is largely due to the use of MXene in its colloidal form – where concentrations are relatively low (~10-20 mg mL^-1) and stability is restricted to neutral aqueous solutions or polar solvents. Additionally, synthesis generally revolves around dissolving the polymer, mixing in colloidal MXene and then solvent casting samples. In order to make samples on an industrial scale, use of multilayer MXene and in-situ polymerization techniques is a must. Clay-reinforced nylon-6 nanocomposites (NCs), characterized by the full exfoliation of the nanofiller were introduced in the marketplace in the 1990s. Herein, we demonstrate, for the first time, a scalable technique to disperse multilayer Ti₃C₂T_z MXene into nylon-6 to synthesize melt-processable nanocomposites with excellent water barrier properties (94% reduction in water vapor permeation). To intercalate the ε-caprolactam monomer, the MXene multilayers were first treated with 12-aminolauric acid, a low-cost, nontoxic, biodegradable, and long shelf life compound. Upon heating to 250 °C, in the presence of 6- aminocaproic acid, in-situ polymerization occurred, yielding melt-processable nylon-6/MXene NCs that were in turn, studied by thermogravimetric analysis, differential scanning calorimetry, X-ray diffraction, scanning and transmission electron microscopy, infrared spectroscopy, and dynamic vapor sorption analysis. Using the latter, moisture-sorption isotherms of a neat and a 1.9 vol % NC at 60 °C, were fit to the Guggenheim, Anderson, and de Boer equation. Solubility, permeation, and diffusion coefficients of water through the NCs were measured as a function of temperature and found to be the lowest ever reported for nylon-6, despite the fact that, at ∼1.9 and 5.0 vol %, the MXene loads were relatively low. This record low diffusivity is ascribed to the very large aspect ratios ~ 500 to 1000 of Ti₃C₂T_z flakes and their dispersion. The water permeation rate is a factor of 5 lower than the best reported in the much more mature nylon/clay field, suggesting lower values can be achieved with further optimization. Lastly infrared spectroscopy spectra of neat and NC samples suggest the surface terminations of the 12-Ti₃C₂T_z flakes bind with nylon-6, limiting water adsorption sites, resulting in reduced solubility in the NC films. This facile, low cost and scalable technique allows for the complete dispersion of MXene into the polymer matrix. By building upon this knowledge and tailoring the chemistry of the surfactant, we are continuing to utilize this method in order to disperse MXene into a variety of thermoplastic and thermosetting polymeric host matrices.

The active-matrix pressure sensors as an electronic device have become of great interests as a critical component for recently emerging electronic skin and biomedical applications for artificial intelligence and human-machine interfaces. The conventional pressure sensors still has limitations such as the narrow sensing range and low pixel density, and they confines the applicability of the device to specific area.
Here, we fabricate the active-matrix arrays of pressure-sensitive monolayer MoS2 field-effect transistor (FET) with air-dielectric layer, which is supported by elastomer spacer The air-dielectric monolayer MoS₂ FET was acting as pressure sensor solely by incorporating air as the gate dielectric layer with elastomeric supporting spacer, and hysteresis of this pressure-sensitive transistor was very small because of the clean surface states between a MoS₂ channel and air. The double-layered elastomer in which the two elastomeric materials had different elastic moduli allowed a wide detectable range of pressures. By utilizing the light emission of the ML composites induced under the external stress, ZnS:Cu phosphors were implemented to the elastomer layer to induce a photocurrent effect in the MoS₂ FETs, resulting in the enhancement of the sensitivity at high pressure (> 0.5 MPa). Furthermore, the pressure on the 20 X 20 MoS₂ FETs active-matrix pressure sensor array was mapped spatially even at 10 µm of pixel resolution. Since this resolution is fine enough to measure 2D pressure distribution of the cell, we demonstrated a 2D mapping of pressure distribution during the pulsing of single cardiomyocyte in the first time. We believe this unique FET structure and its active-matrix array will contribute to future applications of prostheses, robotics and biomedical electronics.

The study of p-n junctions has been of interest in condensed matter physics and materials sciences. Especially, the fabrication of a p-n junction out of atomically thin two-dimensional crystals has drawn attraction in the field of semiconductors as it allows a high-quality interface with very high electron conductivity in the absence of strong chemical bonds at the interface. Here we report an atomically thin p-n junction prepared using a Van der Waals material, SnSe whose bulk property is a p-type semiconductor. The electron band structure of SnSe has been investigated using angle-resolved photoemission spectroscopy (ARPES). By depositing metallic atom on the surface, the valence band maximum shifts towards higher binding energy as much as 0.28 eV, leading to an n-type charge carriers in the surface. A comparison to a theoretical work suggests that this charge transfer appears only at the top-most layer, leaving the rest of the bulk to remain a p-type semiconductor. This result suggests that an atomically well-defined vertical p-n junction is fabricated in between n-type surface and p-type bulk of SnSe.

Air-stable, photosensitive copper hexadecafluorophthalocyanine (F₁₆CuPc) is a promising n-type semiconductor for organic electronics and optoelectronics. However, the performance of F₁₆CuPc-based devices is significantly limited by the poor crystallinity of thin films. In this work, the charge transport and electrical contact behavior of F₁₆CuPc nanoflakes, mechanically exfoliated from high-quality needle-like bulk single crystals, are probed by analyzing the temperature-dependent carrier mobility and conductance, where the multi-trap/release and band-like transport mechanism govern the charge transport at different temperature ranges and carrier densities. F₁₆CuPc nanoflakes based field effect transistors (FETs) exhibit high on-state current and ON/OFF ratio, one-order magnitude than those of reported F₁₆CuPc nanowires, thin films and nanoribbons. Besides, F₁₆CuPc nanoflake-based phototransistors exhibit attractive photoresponse performance in the spectral range of 300−750 nm even at quite low operating source-drain voltage (~ 1V), with maximum photoresponsivity of 19 A W^-1, detectivity of 8×10¹² Jones, and fast response speed of 36 ms, which is attributed to the single-crystalline characteristic of nanoflakes, and the resultant efficiently exciton diffusion and charge transport. Our results demonstrate that two-dimensional organic nanoflakes with single-crystalline feature will be promising candidates for flexible electronic and optoelectronic devices.

Transition-metal dichalcogenides(TMDs), such as molybdenum disulfide (MoS₂), have attracted huge interest as two-dimensional materials because of their novel electronic and optical properties.
Several methods of fabricating few-layered TMDs have been developed, however, methods for mass production of intrinsic nanoflakes of TMDs are still challenge. Mass production of two-dimensional nanoflakes is highly desirable to their practical applications.
In the present work, we present facile method for fabricating MoS₂ nanoflakes by using hydrazine-assisted ball-milling via synergetic effect of chemical intercalation and mechanical peeling.
The MoS₂ nanoflakes obtained through the hydrazine-assisted ball milling process have lateral size of 600 to 800 nm and thickness of 4 nm or less with high quality. The resultant MoS₂ nanoflakes form stable dispersion in various solutions that can be used for many applications.
For a potential application, flexible photodetectors are prepared by solution-based process using the MoS₂ nanoflakes and the optical properties evaluated.

Bismuth telluride (Bi₂Te₃) has recently attracted significant attention owing to its unique physical properties as a three-dimensional topological insulator and excellent properties as a thermoelectric material. Meanwhile, it is important to develop a synthesis process yielding high-quality single crystals over a large area to study the inherent physical properties and device applications of two-dimensional materials. However, the maturity of Bi₂Te₃ vapor-phase synthesis is not good, compared to those of other semiconductor twodimensional crystals. In this study, therefore, we report the synthesis of relatively large-area Bi₂Te₃ crystals by vapor transport method, and we investigated the key process parameters for a synthesis of relatively thin and large-area Bi₂Te₃ crystals. The most important factor determining the crystal synthesis was the temperature of the substrate. A Bi₂Te₃ device exhibited a considerable photocurrent when the laser was irradiated inside the electrode area. This indicated that the photo-thermoelectric effect was the main mechanism of generation of photocurrent. The estimated Seebeck coefficient of the device was ∼196 μV/K, which is comparable to the previously reported high Seebeck coefficient of Bi₂Te₃. This synthesis method can guide the development and applications of various types of layered crystals with the space group of R-3m.

Two-dimensional (2D) transition metal dichalcogenides (TMDCs), such as MoS₂, have attracted considerable attention owing to the unique optical and electronic properties related to its 2D ultrathin atomic layer structure ^[1]. MoS₂ is becoming prevalent in post-silicon digital electronics and in highly efficient optoelectronics due to its extremely low thickness and its tunable band gap (E_g = 1–2 eV) ^[2]. For low-power, high-performance complementary logic applications, both p- and n-type MoS₂ FETs (NFETs and PFETs) must be developed. NFETs with an electron accumulation channel can be obtained using unintentionally doped n-type MoS₂. However, the fabrication of MoS₂ FETs with complementary p-type characteristics is challenging due to the significant difficulty of injecting holes into its inversion channel ^[3]. Plasma treatments with different species (including CF₄, SF₆, O₂, and CHF₃) have also been found to achieve the desired property modifications of MoS₂ ^[4-5].
In this work, we demonstrated a p-type multilayer MoS₂ enabled by selective-area doping using CHF₃ plasma treatment. Compared with single layer MoS₂, multilayer MoS₂ can carry a higher drive current due to its lower bandgap and multiple conduction channels. Moreover, it has three times the density of states at its minimum conduction band. Large-area growth of MoS₂ films on 300 nm thick SiO₂/Si substrate are carried out by thermal decomposition of ammonium tetrathiomolybdate, (NH₄)₂MoS₄, in a tube furnace. a two-step annealing process is conducted to synthesize MoS₂ films. For the first step, the temperature is set to 280 °C for 30 min in an N₂ rich environment at 1.8 Torr. This is done to transform (NH₄)₂MoS₄ into MoS₃. To further reduce MoS₃ into MoS₂, the second step of annealing is performed. For the second step, the temperature is set to 750 °C for 30 min in a reducing atmosphere consisting of 90% Ar and 10% H₂ at 1.8 Torr. The grown MoS₂ films are subjected to out-of-plane doping by CHF₃ plasma treatment using a Dry-etching system (ULVAC original NLD-570). The radiofrequency power of this dry-etching system is set to 100 W and the pressure is set to 7.5 mTorr. The final thickness of the treated samples is obtained by etching for 30 s.
Back-gated MoS₂ PFETs were presented with an on/off current ratio in the order of 10³ and a field-effect mobility of 65.2 cm²V^-1s^-1. The MoS₂ PFETs photodetector exhibited ultraviolet (UV) photodetection capability with a rapid response time of 37 ms and exhibited modulation of the generated photocurrent by back-gate voltage. This work suggests the potential application of the mild plasma-doped p-type multilayer MoS₂ in UV photodetectors for environmental monitoring, human health monitoring, and biological analysis.

References

[1] Mak, K.F.; Lee, C.; Hone, J.; Shan, J.; Heinz, T.F. Atomically thin MoS₂: A new direct-gap semiconductor. Phys. Rev. Lett., 2010, 105, 136805.
[2] Jeong, S.H.; Liu, N.; Park, H.; Hong, Y.K.; Kim, S. Temperature-dependent electrical properties of Al₂O₃-passivated multilayer MoS₂ thin-film transistors. Appl. Sci. 2018, 8(3), 424.
[3] Yin, Z.; Li, H.; Li, H.; Jiang, L.; Shi, Y.; Sun, Y.; Lu, G.; Zhang, Q.; Chen, X.; Zhang, H. Single-layer MoS₂ phototransistors. ACS Nano, 2012, 6 (1), 74.
[4] Xue, F.; Chen, L.; Chen, J.; Liu, J.; Wang, L.; Chen, M.; Pang, Y.; Yang, X.; Gao, G.; Zhai, J.; Wang, Z.L. p-type MoS₂ and n-type ZnO diode and its performance enhancement by the piezophototronic effect. Adv Mater., 2016, 28 (17), 3391.
[5] Wi, S.; Kim, H.; Chen, M.; Nam, H.; Guo, L.J.; Meyhofer, E.; Liang, X. Enhancement of photovoltaic response in multilayer MoS₂ induced by plasma doping. ACS NANO, 2014, 8 (5), 5270.

Two-dimensional (2D) transition-metal dichalcogenides (2D TMDs) in the form of MX2 (M: transition metal, X: chalcogen) exhibit intrinsically anisotropic layered crystallinity wherein their material properties are determined by constituting M and X elements. 2D platinum diselenide (2D PtSe2) is a relatively unexplored class of 2D TMDs with noble-metal Pt as M, offering distinct advantages over conventional 2D TMDs such as higher carrier mobility and lower growth temperatures. Despite the projected promise, much of its fundamental structural and electrical properties and their interrelation have not been clarified, and so its full technological potential remains mostly unexplored. In this work, we investigate the structural evolution of large-area chemical vapor deposition (CVD)-grown 2D PtSe2 layers of tailored morphology and clarify its influence on resulting electrical properties. Specifically, we unveil the coupled transition of structural−electrical properties in 2D PtSe2 layers grown at a low temperature (i.e., 400 °C). The layer orientation of 2D PtSe2 grown by the CVD selenization of seed Pt films exhibits horizontal-to-vertical transition with increasing Pt thickness. While vertically aligned 2D PtSe2 layers present metallic transports, field-effect-transistor gate responses were observed with thin horizontally aligned 2D PtSe2 layers prepared with Pt of small thickness. Density functional theory calculation identifies the electronic structures of 2D PtSe2 layers undergoing the transition of horizontal-to-vertical layer orientation, further confirming the presence of this uniquely coupled structural-electrical transition. The advantage of low-temperature growth was further demonstrated by directly growing 2D PtSe2 layers of controlled orientation on polyimide polymeric substrates and fabricating their Kirigami structures, further strengthening the application potential of this material. Discussions on the growth mechanism behind the horizontal-to-vertical 2D layer transition are also presented.

Similar to graphene, III-V semiconductors offer extraordinary properties on a diversity of morphologies. In the last two decades, gallium nitride (GaN) and related group III-nitrides have become key materials in light emitting diodes (LEDs) and laser diodes [1,2]. Therefore, GaN-based device technologies have improved substantially, and are still under intense investigation for improved performance.
GaN crystallizes in the wurtzite structure (nonlayered). Therefore, the probability to be exfoliated with traditional methods is almost null. However, this phase of gallium nitride has been experimentally produced via encapsulation between two graphene sheets and theoretical results have predicted a stable 2D graphene–like phase for GaN, known as g-GaN in which gallium and nitrogen atoms are in the honeycomb-like hexagonal disposition.
We have studied the graphene phase of gallium nitride (g-GaN) structural and electronic properties by using first-principle calculations within the framework of the Density Functional Theory (DFT). We tried two different approaches: in the first one we considered the hydrogen adsorption at different levels of coverage over each kind of atom and in the second we introduced vacancy charged defects [3].
We considered hydrogen adsorption in a range from the diluted limit, as a 1/64 of the monolayer, up to the full coverage of the monolayer. It was found that N–N distances around the gallium atom with adsorbed hydrogen were similar to the clean distances. Contrarily, Ga-Ga distances around the N atom with adsorbed hydrogen increased as the hydrogen percentage decreased. The 2D g-GaN-H configurations are the most stable, having lower formation energy. Nonetheless, the ferromagnetic system with a hydrogen atom in the gallium position presents a nonzero magnetic moment. As it could be observed in the band structure, a defect level induced by the H adatom is the source of this magnetic moment, a phenomenon which can be exploited for spintronic applications. This ferromagnetism is mainly localized in the first and second closest N atoms around the gallium atoms where adsorption occurred [4].
It was found from the second approach that the Ga-Ga (or N-N) bond lengths around the vacancy increased for the gallium vacancies and decreased for the nitrogen vacancies with respect to the bond distance without vacancy. Vacancy-induced defect levels are responsible for the total magnetic moment in g-GaN, generating a magnetic moment located at the nitrogen atoms for gallium vacancies or delocalized in the lattice for nitrogen vacancies. Thermodynamic transition levels close to the conduction band minimum are identified for both the Ga and N vacancy. In addition, N vacancy has formation energy lower than the Ga vacancy. N vacancy is the dominant defect in g-GaN and could be responsible of the n-type behavior of GaN observed experimentally. Similarly, the w-GaN, gallium vacancies have a higher formation energy than those of nitrogen. However, gallium vacancies could be successfully grown in order to design a two dimensional p-type semiconductor.
It is expected that the results of these studies will contribute to the analysis of hydrogen adsorption on 2D g-GaN, as well as to the design of 2D H-g-GaN and 2D g-GaN-H based devices in electronic and spintronic fields.
References
[1] S. Nakamura, Ann. Phys., vol. 527, no. 5–6 (2015) 335–349.
[2] H Amano, Rev. Mod. Phys, vol. 87, (2015) 1133.
[3] R. González, W. Lopez-Perez, A. González-Garcia, MG. Moreno-Armenta, R. Gonzalez-Hernandez, Appl Surf Sci. Vol. 433 (2018) 1049-1055.
[4] R. González, O. Martinez-Castro, MG. Moreno-Armenta, A. González-Garcia, W. Lopez-Perez, R. Gonzalez-Hernandez, Phys B: Condensed Matter Vol. 569 (2019) 57-61.

Van der Waals (vdW) heterostructures formed by layering twisted monolayers of two-dimensional materials result in formation of Moiré superlattices, which leads to new physical phenomena, including emergence of flat energy bands or tunable conductivity. So far, Moiré superlattice physics has been explored in graphene-related system and various combinations of monolayer transition metal dichalcogenides. Hexagonal boron nitride (hBN) forms a similar honeycomb lattice as graphene, but in contrast to graphene, has a large band gap with luminescence in ultraviolet region. Because of its properties, hBN has been recently extensively used as a substrate or dielectric spacer in vdW heterostructures.
Here, we show that the optical properties of stacked bulk-like (hundreds of nanometers in thickness rather than monolayers) hBN layers are dominated by the innermost interface layers and their relative twist angles as a consequence of the Moiré superlattice formation. Both the luminescence intensity and energy can be tuned using the relative twist angle between stacked layers. hBN multilayers were exfoliated from an hBN single crystal and these were used to form twisted hBN multilayers by a dry transfer method. The relative twist angles between the stacked hBN layers were measured using electron diffraction in a transmission electron microscope (TEM) and the optical properties of the resulting structure were measured simultaneously using TEM-based cathodoluminescence (CL). We show the strong functional dependence of the luminescence properties as a function of the twist angle. Our results demonstrate a new way of controlling optical properties of thin-film materials, beyond conventionally used heterostructures.

The ability to assemble van der Waals heterostructures from two-dimensional (2D) crystals provides a wealth of new opportunities for novel optoelectronic devices. To date, most such devices have shown functionality in the infrared to visible range, whereas ultraviolet (UV) applications are less well explored. Here we describe electrically driven light emission from vertical heterostructure devices consisting of thin hexagonal boron nitride (hBN) between graphene electrodes, with additional hBN encapsulation. Upon illumination at low electric fields, optically excited carriers tunnel and generate interlayer photocurrent through Gr/hBN/Gr heterostructures. At high electric fields, hBN shows bright light emission with a strong peak in the near ultraviolet (NUV) regime at 394 nm (3.14 eV), and stable response over a wide voltage range. Light emission is accompanied by a rapidly increasing current and appearance of new photoluminescence peaks, indicating that it is facilitated by creation of color centers, whose potential structures are explored by density functional theory calculations. Finally, we report the generation of red-green-blue (RGB) color-mixed white light by a standard down-conversion approach using phosphors. These results demonstrate the promise of hBN-based van der Waals heterostructures for light emission in the NUV to visible regime.

All-optical integrated circuits have long been a major pursuit for scientists and engineers because they may allow ultra-fast computing with ultra-low power consumption. The actual realization of this dream has been very difficult, however, because photons rarely interact with each other and, even with the aid of nonlinear optical materials, interact very weakly. My research group tries to leverage the advances in photonics, plasmonics, optoelectronics, and quantum optics and develop new material and technology platforms for solid-state all-optical information processing that work all the way down to single quantum level. In this presentation, I will describe our recent efforts to realize solid-state quantum optoelectronic devices by combining atomically thin semiconductors and nanoscale photonic/plasmonic structures. In particular, I will discuss how we improve the optoelectronic properties of these atomically thin materials and how we leverage them to gain detailed insight into their optoelectronic properties. I will then discuss our efforts to realize atomically thin, electrically tunable mirrors and excitonic “drums” as well as solid-state optoelectronic devices that are built upon quantum mechanical principles.

Visible spectrum band-gaps with strong excitonic absorption make transition metal dichalcogenides (TMDCs) of molybdenum and tungsten attractive candidates for investigating light-material interactions and applications as absorbing media in opto-electronics.^{1, 2} Further, the excitonic features become more prominent as the layers are thinned down and dominant in the monolayer limit where the TMDCs transition into direct band-gap semiconductors with strong photoluminescence. Their excitonic character and luminescence in monolayer regime has been heavily investigated in fundamental science as well as exploited for device applications.² However, the excitonic character in TMDCs persists even for few-layer thick samples and bulk crystals which has led to recent demonstrations of some interesting photonic and light trapping phenomena. Further, TMDCs are known to have very large values of optical constants which allows strong light trapping even in ultrathin samples.³
Here, we explore the fundamental material properties of light trapping in multi-layer TMDCs (namely WS₂) when coupled to plasmonic substrates. Our calculations and demonstrations are primarily in WS2 which can be easily extended to other excitonic TMDCs. We systematically demonstrate via calculations and matching experiments that the presence of strong excitonic resonances in multilayers (< 20 nm thickness) combined with surface plasmon excitations of the nearby metals can achieve strongly coupled modes with apparent voided crossings in reflectance spectra. Further, we explore additional light confinement by patterning 1D arrays of rectangular resonators of varying widths and periods (100 nm to 500 nm). We find that there is a critical thickness of TMDC (~15 nm) beyond which we observe photonic modes emerging in the TMDC/metal hybrids that are strongly dispersive as a function of resonator width. Newer and higher order modes appear with increasing TMDC thicknesses and widths. Simulated field profiles suggest that these modes are mainly plasmonic and hybrid in nature. For longer wavelengths and thicker TMDCs, guided modes emerge as well. Further, the plasmonic modes exhibit strong dependence on 1D array grating period.
Finally, we will show the detailed multi-dimensional phase diagram of geometric parameters that lead to voided crossings between various resonant photonic modes and excitons and demonstrate sensing applications in the strong coupling regime. Our work may pave the way to novel hybrid photon-polariton states that are of interest for all-optical information processing, quantum and classical computing.
References:
1. Jariwala, D.; Davoyan, A. R.; Wong, J.; Atwater, H. A. ACS Photonics 2017, 4, 2692-2970.
2. Brar, V. W.; Sherrott, M. C.; Jariwala, D. Chemical Society Reviews 2018, 47, (17), 6824-6844.
5. Jariwala, D.; Davoyan, A. R.; Tagliabue, G.; Sherrott, M. C.; Wong, J.; Atwater, H. A. Nano Letters 2016, 16, (9), 5482-5487.

Two-dimensional materials are inherently anisotropic – with strong bonding in-plane and weak bonding out-of-plane. Consequently, their optical properties are birefringent, particularly at frequencies close to the energies of their optically active infrared phonons. At these frequencies, certain 2D materials support hyperbolic behaviour, where along one (two) crystal axes they behave as metals (negative dielectric function), and along the other two (one) behave as dielectrics (positive dielectric function). Typically, two of these axes are in the plane of the 2D layers, and the third is out of the plane, the most well studied example being hexagonal boron nitride (hBN). Materials which exhibit hyperbolicity can support polaritonic modes, which have a wavelength much shorter than that of light in free space. Such spatially compressed modes have important implications for infrared technology, as they may be utilized to create infrared optics that are much more compact than the current state of the art. In particular, 2D materials can be used to realize metasurfaces, capable of shaping both nearfield- and far field light waves for a broad range of applications.
Conventional metasurface designs use geometrically fixed structures, or materials with excessive propagation losses, thereby limiting their potential applications. Here we show how to overcome these limitations by demonstrating a reconfigurable hyperbolic metasurface comprising a heterostructure of isotopically enriched hexagonal boron nitride (hBN) in direct contact with the phase-change material (PCM), single-crystal vanadium dioxide (VO₂). Metallic and dielectric domains in VO₂ provide spatially localized changes in the local dielectric environment to tune the wavelength of hyperbolic phonon polaritons (HPhPs) supported in hBN by a factor of 1.6. In contrast, propagation was only seen above a low loss dielectric phase with non-anisotropic materials supporting surface polaritons.
Using this platform, we demonstrate the first reduction to practice of in-plane HPhP refraction, and the means for launching, reflecting and transmitting of HPhPs at the PCM domain boundaries. Ultimately, this phenomenon can be used to create devices based on refractive effects, such as lensing, but on length scales far below the diffraction limit. This relies on the unique properties of the 2D materials themselves, aided by the changing dielectric function of the PCM substrate.

The interaction between two-dimensional crystals (2DCs) and metals is ubiquitous in 2D material research. Of particular interest here, is the fact that metals can significantly affect both the electronic and optical properties of 2D semiconductors. We report how 2DC overlayers influence the recrystallization and dewetting behavior of metallic films and the subsequent opto-electronic phenomena that arise from this new heterostructured assembly. The annealing of nanocrystalline metallic films with 2DC overlayers results in a ‘surface templated epitaxial’ process where the metal films become highly textured, with micron-sized, well-faceted grains in close crystallographic registry to the 2D crystal orientation. Dewetting of the encapsulated metal film occurs through the formation of pores that can span up to 60% of the planar surface area, creating an orientated pore enabled network (OPEN) film. This process has been demonstrated for a number of mechanically-exfoliated and chemical-vapor deposited 2DCs (e.g., graphene, h-BN, and transition-metal dichalcogenides (TMDs)). In the OPEN film structure, the 2DC overlayer can remain suspended above or coat the inside of the metal pores. OPEN-TMD films exhibit appreciably distinct properties, including enhancement in photoluminescence, spatially varying surface potential, and routes for coupling surface plasmon-polaritons (SPPs) to photon emission. SPPs were found to propagate throughout the OPEN film structure following generation by either free-space laser excitation or TMD-emitted photons at the metal pore sites, enabling non-local emission at room and low temperatures. We also find a high density of single photon emitters (SPEs) across an OPEN-WSe₂ film and demonstrate non-local SPE excitation at distances of at least 17 mm with minimal loss of photon purity. Our results suggest the OPEN film geometry is a versatile platform that could facilitate the use of layered materials in quantum optics systems.

Work performed at NRL was supported through Base Programs funded by the Office of Naval Research. JJF acknowledges the NRC Research Associateship Programs for their support.

Future integrated photonic circuits need materials to perform ultra-fast and low-power optical switching. Chalcogenide phase-change materials (PCMs) and two-dimensional (2D) materials contain many promising candidates to meet this need. Among layered 2D chalcogenides, IV-VI materials are particularly interesting for their trirefringent optical properties and low-processing temperature.
Here we propose a new mechanism for optical switching in anisotropic materials, and we report results obtained for SnSe. a-SnSe has an orthorhombic crystal structure that can be visualized as a highly-distorted rock salt lattice. We present theoretical predictions that, owing to strong dielectric anisotropy, short pulses of linearly polarized light can switch between ferroelastic variants of the a-SnSe structure.^[1] Switching occurs because the light field changes the potential energy landscape and eliminates the barrier for switching. We predict a critical power density of 7.7 x W/cm² for near-IR photons for a barrier-less transition. We report experimental results obtained on crystals and mechanically-exfoliated sheets. We use spectroscopic ellipsometry to measure the full optical property tensor of trirefringent a-SnSe, to test theoretical calculations and to optimize the design of optical switching, and we report preliminary results using short laser pulses to perform ferroelastic switching.
The study presented on trirefringent a-SnSe can be generalized to a class of optical martensitic transitions that we predict to be ultra-fast, barrier-less, and diffusion-less. These transitions happen at the speed of lattice vibrations and are inherently non-thermal, and therefore low-power. 2D materials are ideal for realizing the predicted phenomena and for application because the transformation strains are lower than in 3D crystal structures. Optical martensitic transitions may therefore offer an improvement on existing PCMs for photonics in terms of switching energy, speed, and device lifecycle.
[1] J. Zhou, H. Xu, Y. Li, R. Jaramillo, and J. Li. Nano Lett. 18, 7794 (2018).

Transition metal dichalcogenides (TMD) exist in several polymorphs including 2H (usually semiconducting), and 1T (or 1T’, semi-metallic). These polymorphs are expected to have significantly different opto-electronic properties, and phase change between these polymorphs offers a new way for controlling light. This phase change is particularly interesting for near-infrared (NIR) integrated photonics, where materials with strong light-matter interaction are required.
We present an experimental and theoretical study of sulfides (MoS₂, TiS₂ and ZrS₂) and tellurides (MoTe₂ and Mo_1-xW_xTe₂) in the NIR. Our density functional theory (DFT) calculations predict a large refractive index contrast in the 1-1.5 um spectral range (relevant for telecommunications and photonics) between 2H and 1T (or 1T’) phases. We measure the complex optical constants of bulk crystals via spectroscopic ellipsometry and Fourier transform IR (FTIR) spectroscopy, and find significant differences between 2H and 1T materials. By explicitly measuring native oxide thickness using transmission electron microscopy and subsequent optical modelling, we extract true optical constants of TMDs. Further, for structural polymorphs of the same material (MoTe₂ and Mo_1-xW_xTe₂ in both 2H and 1T phases), we measure a large difference in refractive index and optical loss, in agreement with our DFT calculations. Our work lays the foundation for TMDs for use as active IR materials that can be switched optically and electrically.

Two-dimensional materials are known to exhibit an amazing range of novel electronic and optical properties, as demonstrated in many state-of-the-art experiments. First principles calculations play an important role in identifying the atomic-scale origins of these properties, as well as in predicting properties of new materials. In this talk, I will draw upon examples from recent work in my group on first principles calculations in 2D materials and their heterostructures. Comparison with experiment will be made throughout the talk. Firstly, I will show convincing evidence that oxygen-related point defects are dominant in mechanically-exfoliated and chemical-vapor-deposition (CVD)-grown WSe₂ samples, in contrast to mechanically-exfoliated MoS₂ samples, where sulfur vacancies are dominant. Furthermore, based on state-of-the-art, numerically converged, GW-BSE calculations, we show that oxygen interstitials in the WSe₂ lattice (in conjunction with strain gradients) are most likely responsible for experimentally observed single photon emission in these materials.[1] Secondly, I will present the electronic properties of covalently functionalized organic-MoS₂ heterostructures that have been experimentally synthesized. Interestingly, we find that local magnetic moments are present. Furthermore, while the organic molecules can either withdraw or donate electrons from the MoS₂ layer, depending on the electron affinity of the molecules, electrons are always the predominant carrier type in the MoS₂ layer. Thirdly, I will show how our recent theoretical developments have enabled the prediction of quasiparticle level alignment at organic-2D material interfaces, in good agreement with experiment.[2,3] Fourthly, I will show that experimentally observed Dirac cones in black phosphorus can be explained by quantum confinement and anisotropic interlayer interactions.[4] If time permits, other results, including predictions on novel materials from high throughput calculations, will be presented.

[1] ACS Nano 13, 6050 (2019)
[2] JCTC doi:10.1021/acs.jctc.9b00229
[3] ACS Nano 10. 2476 (2016)
[4] Scientific Reports 5, 11699 (2015)

Van der Waals heterostructures (vdWHs) consist of layers of dissimilar 2D materials that provide novel and unique and properties. vdWHs are commonly assembled in a myriad of combinations by stacking 2D materials using polymeric stamps. However, the properties of such heterostructures frequently are degraded by contaminants, typically of unknown composition, that are trapped between the constituent layers and impede studies of the heterostructures’ intrinsic properties, thus hindering their applications.
In this work, I will introduce the photothermal induced resonance (PTIR) technique, an atomic force microscopy (AFM)-based infrared technique that enables IR spectroscopy and imaging with nanoscale resolution. I will show that PTIR enables detection and identification of contaminants down to a few attomoles
Heterostructures comprised of WSe2, WS2, and hBN layers were found to contain significant amounts of polydimethylsiloxane (PDMS) and polycarbonate, corresponding to the stamp materials used in their construction. Notably, polymer contamination between layers occurs even when ‘dry’ fabrication methods are used. Having identified the source of the contaminants, we develop a stamp cleaning procedure that eliminates or reduces PDMS contamination within the structures below the limit of detection. Our measurements suggest that the assessment of contaminants in vdWHs should not be conducted based on AFM topographic images alone but, rather, aided by nanoscale composition-sensitive methods, such as PTIR. We believe that knowledge of the contaminant composition obtained with the methods presented here will aid better understanding of vdWH properties and guide improvements to fabrication method for producing intrinsic, contaminant-free heterostructures with precisely tuned properties.

Janus transition metal dichalcogenide (TMD) is a newborn of the two-dimensional (2D) materials family. Its structure is similar to TMDs such as MoS₂, but one layer of the chalcogen is different from the other layer, one example being MoSSe. As the first successfully synthesized Janus TMD, MoSSe has motivated a series of theoretical investigations on the Janus TMDs. Due to the unique crystal structure of Janus TMDs, unconventional phenomena have been theoretically predicted, including out-of-plane piezoelectricity and exciton disassociation by the intrinsic out-of-plane dipole moment. However, experimental investigations are extremely insufficient in terms of the principle properties of monolayer Janus TMDs and their heterostuctures with other van der Waals materials. Such properties, for example interlayer coupling and phonon properties, are essential to electronic transport and optoelectronic applications.
In this work, we study the fundamental phonon properties and interlayer coupling of Janus monolayer MoSSe and MoSSe/MoS₂ heterostructures. Interlayer breathing and shear modes of high-symmetry 2H and 3R heterostackings are probed by low frequency Raman spectroscopy. The uniform selenization from MoS₂ to MoSSe is confirmed by the disappearance and the emergence of signature Raman modes of MoS₂ and MoSSe, respectively. Low frequency Raman modes for different stacking patterns of MoSSe/MoS₂ show noticeable variations in frequencies. This observation indicates that stacking configuration leads to different layer-to-layer interactions. Unintuitively, interlayer coupling in the heterostructures is stronger than their pure MoS₂ counterparts possibly due to the compressive (tensile) strain in MoSSe (MoS₂) introduced during synthesis. Difference in high frequency modes between MoSSe/MoS₂ and pure MoS₂ is consistent with variations in the low frequency modes. Both the low frequency and high frequency Raman responses are strong evidence for the abnormally enhanced interlayer coupling. These spectroscopic features can serve as a fingerprint of stacking configurations, interlayer coupling in heterostructures, and degree of conversion in the fabrication process from TMDs to Janus TMDs.

The simplicity and versatility of optical microscopy make it from the start the workhorse technique in the characterization of 2D materials (1). Optical microscopy is used to locate and determine the thickness of the 2D material by measuring its optical contrast with respect to the Si/SiO₂ substrate (2). Although in terms of technology, the large-area growth of 2D materials is about to be mastered soon, as-grown 2D materials still host abundant and different types of defects such as vacancies, adatoms, grain boundaries (GBs), edges, and impurities, which strongly influence their properties (3). In most cases, the presence of defects is disadvantageous. However, not all defects in 2D materials are detrimental. Some 2D materials have been shown to host defects that can serve as single-photon emitters (SPEs) at cryogenic temperatures for TMDs (4-7) and room temperature for h-BN (8). This discovery has motivated the search for single-photon sources in other 2D materials and efforts that aim to engineer the defects in well-controlled locations either using strain- induced potential traps (9) or via quantum dot confinement (7). The advent of single quantum emitters in 2D materials offers new opportunities to construct a scalable quantum architecture. Transmission electron microscopy TEM, SPM or confocal microscopy techniques are not ideal for fast, high-throughput, in-situ imaging of defects in 2D materials with nanometer resolution. There is a clear demand for the development of advanced optical technology that images individual defects at better temporal, spectral and spatial resolutions. We have explored the single molecule localization microscopy to characterize defects in hexagonal boron nitride (10). In addition to precise location of the optically active defects we record as well their spectral properties using spectral SMLM (11).

1. K. S. Novoselov et al., Two-dimensional atomic crystals. Proc Natl Acad Sci U S A 102, 10451-10453 (2005).
2. M. Benameur et al., Visibility of dichalcogenide nanolayers. Nanotechnology 22, 125706 (2011).
3. F. Banhart, J. Kotakoski, A. V. Krasheninnikov, Structural defects in graphene. ACS Nano 5, 26-41 (2011).
4. D. Dumcenco et al., Large-Area Epitaxial Mono layer MoS2. Acs Nano 9, 4611-4620 (2015).
5. C. Chakraborty, L. Kinnischtzke, K. M. Goodfellow, R. Beams, A. N. Vamivakas, Voltage-controlled quantum light from an atomically thin semiconductor. Nat Nanotechnol 10, 507-511 (2015).
6. M. Koperski et al., Single photon emitters in exfoliated WSe2 structures. Nat Nanotechnol 10, 503-506 (2015).
7. A. Srivastava et al., Optically active quantum dots in monolayer WSe2. Nat Nanotechnol 10, 491-496 (2015).
8. T. T. Tran, K. Bray, M. J. Ford, M. Toth, I. Aharonovich, Quantum emission from hexagonal boron nitride monolayers. Nat. Nanotechnol. 11, 37-41 (2016).
9. G. Grosso et al., Tunable and high-purity room temperature single-photon emission from atomic defects in hexagonal boron nitride. Nat Commun 8, 705 (2017).
10. J. Feng et al., Imaging of optically active defects with nanometer resolution. Nano Lett, (2018).
11. J. Comtet et al., Wide-field spectral super-resolution mapping of optically active defects in hBN. arXiv preprint arXiv:1901.06989, (2019).

Angle-resolved photoemission spectroscopy (ARPES) is a premier technique for determining the electronic structure, state symmetry, and many body interactions (MBIs) in correlated and topological materials. These MBIs involve exchange of momentum among electrons or with excitations such as phonons, and can therefore couple to nanoscale structures. By controlling the structure at the nanoscale, we can therefore hope to control or enhance the ground state properties of materials through nanoscale engineering. This dream has motivated the development of ARPES with nanoscale spatial resolution (nanoARPES), in order to probe these effects. MAESTRO, the Microscopic and Electronic Structure Observatory, is a new synchrotron based user facility for the study of in situ prepared materials, including oxides, 2D van der Waals material, semiconductors, metals, and surfaces. With a combination of three ARPES microscopes with complementary spatial/energy/momentum resolutions, and in situ sample preparation (molecular beam epitaxy, pulsed laser deposition, and micro-mechanical sample transfer) we are able to examine the relationship between electronic structure and topology with unprecedented spatial resolution, currently around 100nm, with 50 nm performance on the horizon.

As an example, I will show the spatially resolved electronic structure of two-dimensional metal dichalcogenide heterostructures of WS2, graphene, BN and TiO2 [1, 2]. Among the findings is a striking renormalization of the spin-orbit splitting of the WS2 valence band, which can be controlled by chemical doping or by choice of substrate. This is attributed to the impact of trion (charged exciton) formation on the self-energy of carriers in WS2. We have also observed modifications in the electronic structure of “twisted” layers of WS2 on graphene. Such twisted heterostructures are garnering a lot of attention due to the possibility of engineering novel electronic structures.

[1] Katoch, J. et al (2018). “Giant spin-splitting and gap renormalization driven by trions in single-layer WS2/h-BN heterostructures.” Nature Physics, 14, 355–359. https://doi.org/10.1038/s41567-017-0033-4

[2] S. Ulstrup et al (2019), “Imaging microscopic electronic contrasts at the interface of single-layer WS2 with oxide and boron nitride substrates.” Applied Physics Letters, 114, 151601 (2019). https://aip.scitation.org/doi/10.1063/1.5088968

[3] S. Ulstrup et al (2019), “Direct observation of mini-bands in twisted heterobilayers.” arxiv.org/abs/1904.06681

Black phosphorus is one of the most promising 2D semiconductors because of its layer-dependent bandgap and high mobility carriers. Its corrugated crystal structure also creates a unique electronic anisotropy which may create opportunities for novel angle-dependent electronic and optoelectronic devices. Native defects and their impact on the electronic structure of 2D semiconductors are understudied, especially when compared to defects in 3D semiconductors. Additionally, the degradation of black phosphorus in air, which is likely exacerbated by defects, creates a major limitation to applications. This talk will present scanning tunneling microscopy and spectroscopy studies of surface structures in black phosphorus, including defects, nanoscale stripes, and islands. We will use a temperature-dependent study of the native defects in black phosphorus to compare defect hopping rates at different temperatures to theoretical expectations. Additionally, we will present data on nanometer-scale black phosphorus islands, created by partially sublimating the top layers of the crystal. These islands appear to lose their crystallinity as they shrink to 10s of nanometers in diameter. Finally, we will present newly observed periodic, 5 nm stripes, whose appearance depends strongly on temperature. The voltage dependence of these stripes suggests an electronic superstructure. We will present the data and possible interpretations of this new superstructure.

A fundamental understanding of the coupling and interfacial phenomena between two-dimensional (2D) atomic layers and their 3D bulk material adjuncts is necessary for the continued and rapid development of 2D material design and device architectures. The structure and defects at the 2D/3D interface influence properties such as contact resistance, photoresponse and high frequency performance, and offer opportunities for tuning band alignment to enable drastic changes in function and properties beyond those of the individual materials. A detailed understanding of how interface structure develops during growth and how it may be modified by post-growth processing can help guide efforts to create interfaces with specific characteristics. Here we focus on an important class of interfaces between 2D and 3D materials, the structures formed when nanoscale metal islands are deposited on 2D layers. We image the structure of these interfaces during and after growth with in situ ultra-high vacuum (UHV) transmission electron microscopy (TEM). We have examined both noble and highly reactive metals (e.g. Au, Ag, Cr, Ti, Nb) deposited on a variety of 2D van der Waals materials (Gr, hBN, MoS2, WSe2 and Bi2Te3).
Samples consisting of suspended 2D membranes or heterostacks are pre-annealed in UHV at 550 C to minimize surface contamination. Metals are then deposited by evaporation and imaged without breaking vacuum. Imaging is also carried out during post-growth annealing. This procedure allows for observations of reactive metals such as Nb that would otherwise oxidize before imaging. We find that the metals exhibit equilibrium nanoisland morphologies and dynamics that are different from those seen on ‘uncleaned’ 2D material samples. Combining spatially and temporally resolved in situ TEM movies with post-growth atomic resolution analytical TEM, we measure the crystal orientation, growth morphology and modulated periodic local strain fields of the deposited nanoislands. We will describe a systematic study of several metals under varying deposition parameters that shows the transformation from dendritic growth to compact, faceted single crystal growth, tailoring growth parameters favourable for heterostructure contacts. We also find that some metals, although not epitaxial on initial deposition, can be epitaxially templated by pre-deposition of another metal such as Au, resulting in single crystal nanoisland arrays. To further predict and optimise interface morphology, we quantify the effect of 2D defects on the nucleation sites of nanoislands using both intrinsic defect density mapping and samples that are pre-patterned using helium ion microscopy. Combining the factors of deposition conditions, templating and nucleation control greatly enhances routes towards self-assembled and tailored interface design. We describe steps towards numerical analysis that ultimately aim towards a comprehensive description of both the atomic scale growth and kinetics at these material interfaces, and their resulting strain fields, defects, and dislocations. We also describe possibilities for measuring the electronic properties of interfaces that were specifically characterized in situ. We believe that achieving control of 2D/3D interfacial structure and dynamics will facilitate systematic design of contacts, heterostructures, and coupled materials for future devices with distinct properties and applications.

To understand the structure-property relationship of 2D materials and heterostructures at the fundamental level, defects and crystalline imperfections need to be accounted for with high precision. AET allows the determination of 3D atomic structure of crystal defects and disorder systems and has recently been advanced to capture nucleation at 4D atomic resolution. However, AET has thus far been limited to metallic nanoparticles and needle-shaped samples. We have developed scanning atomic electron tomography (sAET) to localize the 3D atomic coordinates in 2D materials and heterostructures with picometer precision. Using a Re-doped MoS₂ monolayer, we demonstrated the sAET method and determined the 3D atomic coordinates with picometer precision and identified 3D crystal defects such as dopants, vacancies and atomic-scale ripples. We measured the 3D atomic displacement and the full strain tensor of the Re-doped MoS₂ and observed strong correlations between local atomic strains and Re, and S vacancy concentrations. Furthermore, the experimental 3D atomic coordinates were used as direct input to DFT calculations to correlate crystal defects with the electronic band structure at the single-atom level. We observed stark differences between the band structures obtained from the experimental and relaxed atomic models. The local atomic strains from Re-dopants were found to drastically alter the overall electronic properties and was corroborated by photoluminescence spectra measurements. We also demonstrate the capabilities of the sAET method through accurate reconstructions and 3D atomic coordinate determination of a multislice simulated MoSe₂-WSe₂ moire patterned van der Waals heterostructure. We anticipate that sAET is not only generally applicable to the determination of the 3D atomic coordinates of 2D materials, heterostructures and thin films, but also could transform ab initio calculations by using experimental atomic coordinates as direct input to reveal more realistic physical, material, chemical and electronic properties.

The stacking of heterogeneous 2D materials using transition metal dichalcogenides (TMDC) allows the development of the hybrid artificial structures for unique application of tuned physical, electrical and optical properties¹. The artificial integration of heterogeneous sheets has produced abundant functionalities in the hybrid structure through interface engineering². Hence, hetero-junction hybrids are incredibly favourable to fabricate p-n junction, tunneling FET, memory devices and other optoelectronics devices³. Some of the crucial contributions of hetero-layer devices is controlled doping, bandgap tailoring and modulation of Schottky barrier height without the implementation of conventional dopant diffusion, ion implantation and laser treatment⁴ that leads to crystal damage and surface defect⁵. Nevertheless, stacked heterogeneous structures are susceptible to mechanical strain, interfacial charge transfer and unpinning of Fermi energy level through inter-layer van der Waals interaction. Thus, it has become pivotal to decouple the contribution between the electrical and mechanical properties of the overlapped heterogenous structure to fabricate flexible electronics.
The present work demonstrates nanoscale electrical properties and interfacial characteristics of two versatile heterogeneous stacked structures beyond graphene; WS₂/MoS₂ and the graphene/MoS₂. The interactions of the monolayer MoS₂ with WS₂ and graphene is collated comprehensively for the mobility of charges, doping and work function. The topping of conductive graphene over the pyramidal arrangement of MoS₂ sheets up to three layers revealed the unpinning of Fermi energy level at different step-edges of the hybrid structure. Consequently, the work function and the amount of charge transfer was found different at each layer of the hybrid structure. The nanoscale conductivity map showed the higher values of currents/area (nA/µm²) in graphene/MoS₂ configuration at thicker MoS₂ layers. Therefore, a transition from indirect to direct band gap in the CVD MoS₂ increases the charge mobility at the interface. Finally, we deconvoluted the strain in the hybrid structure from doping by Raman correlation plots⁶ for E_2g Vs A_1g and G Vs 2D peaks for TMDC (MoS₂ and WS₂) and graphene Raman modes, respectively. This investigation will address to establish a functional method to tailor the nanoscale electrical and mechanical properties of the scalable engineered 2D heterostructure for high-performance flexible electronics.
References:
1. L. Britnell, R. M. Ribeiro, et al., Science, 2013, 340, 1311-4.
2. Y. Gong, J. Lin, et al., Nature Materials, 2014, 13, 1135.
3. B. W. Baugher, H. O. Churchill, et al., Nature Nanotechnology, 2014, 9, 262.
4. M. Tripathi, A. King, et al., ACS Omega, 2018, 3, 17000-9.
5. M. Tripathi, F. Awaja, et al., ACS Applied Materials & Interfaces, 2018, 10, 44614-23.
6. A. Michail, et al., 2D Materials, 2018, 5, 035035.

Transition Metal Dichalcogenides (TMDs) are promising materials for beyond silicon electronics due to their unique electrical, including tunable band gap in the visible regime, mechanical and thermal properties. The electronic quality of such materials is highly effected by lattice impurities, such as Sulfur vacancies and other atomic defects which generates atomic gap states in TMDs, as well as the semiconductor/oxide interface imperfection.
To-date, even though some recent progress has been reported, the study and understanding of gap states in TMDs materials and devices is very limited. First, reliable, sensitive and quantitative methods to measure the concentration and energy distribution of such states are not fully developed. Second, fundamental issues surrounding the origin of gap states in TMDs materials are not known, and relatively very little is reported about the effect of these states on electronic device properties.
In this work, we use Kelvin Probe Force Microscopy (KPFM) under inert atmosphere to measure the gap states concentration, their energy distribution and Fermi level pinning in MoS₂ monolayers and multilayer transistors. The method is based on applying back gate voltage to a working MoS₂ Field Effect Transistors in-operando. The voltage applied to the back gate induce injection of electrons/holes into the transistor channel; these charge carriers populate the gap states and consequently change the Fermi level position¹. The Fermi level energy is measured directly by the KPFM, and the gap states energy distribution is then extracted using the injected charge concentration obtained from the transistor capacitance as described in our previous works¹. We have measured the gap states of both MoS₂ monolayers and multilayers exfoliated on SiO₂ and found that in monolayers the Fermi level is unpinned throughout most of the energy gap, and the states distribution is fairly constant at a concentration of around 10¹⁸ cm^-3 eV^-1. In samples which are several multilayers thick, the states distribution decays exponentially from the conduction band edge similar to our measured bulk samples. Gap states measurements at the interfaces of MoS₂ with other dielectrics will be presented and discussed.

Tal, O., Rosenwaks, Y., Preezant, Y., Tessler, N., Chan, C. K. & Kahn, A.. Phys. Rev. Lett. 95, (2005).

Piezoelectricity is widely considered to be one of the emerging research fields owing to its promising applications in diverse areas including sensors, nanogenerators, energy harvesting, and biomedical devices. Recent studies on the piezoelectricity in two-dimensional materials show remarkable features for manipulating band structure and many important physical characteristics. Here, we demonstrate the piezo-phototronic property observed in the In_1-xSn_xSe flexible photodetector, where the performance of the device can be tuned by applying systematic mechanical strain. The piezoelectric property and changes in the band gap of the material were carefully studied with several spectroscopic analysis under strained conditions. Quite interestingly, we discover that in the fabricated In_1-xSn_xSe device, the dark and photo drain-source current can be increased by 5-fold under a bending strain of 2.7 %, which shows a great promise for the design of high performance, multifunctional device as strain sensor and photodetector. The maximum photoresponsivity and strain sensitivity are obtained as 1037 AW^-1 and 206, respectively, which outperform the flexible devices in the same class of two-dimensional materials. In addition, we show that the few-layer In_1-xSn_xSe devices can be attached on a freeform surface with high performance. Thus, our designed multifunctional device is very useful for the development of advanced applications to circumvent the requisite demand for emerging technologies.

Molybdenum disulphide (MoS₂), one of the most widely applicable transition metal dichalcogenide (TMD) is known to have high electron mobility at room temperature as well as optical property for monolayer. The band gap of single layer MoS₂ shows direct transition which values ~1.8 eV [1] is believed to be the source of its rectifying behavior. MoS₂ shows Schottky behavior when it is electrically contacted to metal by forming Schottky barrier (SB) at metal/MoS₂ interface. The potential barrier at the interface is directly related to contact resistance which should be minimized to achieve lower power consumption for high device performance. Many efforts to engineer contact resistance at the metal/TMD interface such as adopting low work function metal for contact [2], doping the channel material with molecular dopant [3], transferring conductive layer at the contact/channel interface [4] and introducing thin insulating material on top of the channel [5] have been reported. In this study, we used centimetre-scale synthesized hexagonal boron nitride (hBN) on top of the MoS₂ channel layer in order to take advantage of absence of fermi level pinning which is considered to be a main reason of forming high Schottky barrier at the interface.

[1] R. Kappera et al, Nature Mater. 13, (2014) 1128 - 1134
[2] S. Das et al, Nano Lett, 13(1) (2013)pp 100 - 105
[3] Y. Du et al, IEEE ELECTRON DEVICE LETTERS, 34(10) (2013) 1328 - 1330
[4] X. Cui et al, Nature Nanotechnol 10, (2015) 534 – 540
[5] S. Lee et al, Nano Lett, 16(1), (2016) pp 276 – 281

The development of fast and reliable doping techniques of two-dimensional (2D) van der Waals (vdW) semiconductors are required for next generation 2D circuit fabrication. Many attempts are being reported, but such techniques are still hard to find. Here we demonstrated that light illumination can induce chemical doping on channel of 2D semiconductor molybdenum ditelluride (2H-MoTe₂). Moreover, the doping carrier types are selective by controlling the light illumination conditions such as wavelength and optical intensity. We found that the channel was doped to p-type under visible light illumination, whereas to n-type under ultraviolet light illumination. This light induced doping process is fast and reliable, so it can be used to assemble various functional devices or circuit elements. As a representative example, we demonstrated that we could directly write 2D CMOS inverter consisting of n- and p-type FET channels. Moreover, this CMOS inverter can also be easily converted to a CMOS switch using laser illumination.

Hexagonal Boron Nitride (BN) exhibits its own unique sets of properties, such as high chemical and thermal stability, transparency to visible light, electrical insulativity, and high thermal conductivity. The latter two properties are unique because of their contradictory nature, which attracts great attention for thermal dissipation applications. Furthermore, combined with the fact that BN is transparent to visible light, it has great potential to be used as a thermal conductive layer in light entrance part of optoelectronic devices.
In order for BN to exhibit high thermal conductivity and transparency to visible light, it needs to be densely stacked and aligned along the horizontal direction to suppress backward scattering of phonons and photons. However, fabrication of this aligned structure faces several processing challenges. Although liquid-phase deposition methods such as ink-jet printing and spray coating offer good scalability, they usually lack uniformity and poor adhesion due to the coffee ring effect and BN nanosheets crumpling.
Herein, we propose a new fabrication method of high-performance BN films by combining the vacuum-filtration and transfer-printing. The process is facile, rapid, and applicable to various substrates having thickness scalability with higher functionality compared to conventional deposition techniques. Controlled vacuum-filtration enables BN nanosheets, which is immediately followed by transfer-printing based on solvent-evaporation-induced adhesion-switching between BN-membrane and BN-substrate interfaces. The fabricated BN films with a few micron thickness has compact, anisotropically aligned nanostructure, achieving an unusually high thermal conductivity over 23W/m.K and optical transparency over 90%. Additionally, thermal dissipation effect of the BN layer characterized with simple joule heating apparatus. Finally, we report that the transfer-printed BN has an outstanding light diffusion property that optically enhances the performance of various optoelectronic applications including LEDs, thin film solar cells and organic dyes by facilitating both light out-coupling and absorption.

MoS₂ nanostructures, i.e., nanoribbons, nano-mesh, etc., may open different prospect of applications in nano-electronic and opto-electronic devices and sensors. However, the fabrication of these complicated nanostructures can be executed by using standard nano-patterning techniques such as lithography, printing, etc. Nevertheless, these standard techniques involve affluent multistep processes to optimize scalability, form factors and accuracy in the feature size. Here, we demonstrate the fabrication of unique nano-structures on MoS₂, such as nano-ribbons and nano-mesh, by a simple one-step process of direct laser writing using 532 nm low power focused laser. The minimum power required to etch a MoS₂ layer for a 532nm laser is found to be nearly 6.95 mW and the minimum void size observed is nearly 300 nm, which is very close to the diffraction limit of the laser used. Both the experimental and computational results have shown that the voids induced by laser etching always take a hexagonal or triangular shape, which can be used to define crystal orientation of the MoS₂ flake. Investigation shows that the periphery of hexagonal voids lies on S atoms, whereas for triangular voids, it lies on Mo atoms of the MoS₂ crystal. In-depth AFM and Raman analysis show that the etching rate is tunable by controlling the laser power and the exposure time. We have also demonstrated the unique electrostatic properties of MoS₂ nanostructures, fabricated in a controlled manner of different geometries on 2D flake by using the above established technique (low power focused laser irradiation technique). Electrostatic force microscopy has been carried out on MoS₂ nanostructures by varying tip bias voltage and lift height. The analysis depicts no contrast flip in phase image of the patterned nanostructure due to the absence of free surface charges. However, prominent change in phase shift at the patterned area is observed. Such contrast changes signify the capacitive interaction between tip and nanostructures at varying tip bias voltage and lift height, irrespective of their shape and size. Such unperturbed capacitive behavior of the MoS₂ nanostructures offer modulation of capacitance in periodic array on 2D MoS₂ flake for potential application in capacitive devices.

References:
1) Controlled Formation of Nanostructures on MoS₂ Layers by Focused Laser Irradiation. R. Rani, D. Sharma, A. Kundu, N. Jena, A De Sarkar*, K. S Hazra*. Applied Physics Letters, 2017.
2) Modulating Capacitive Response of MoS₂ Flake by Controlled Nanostructuring through Focused Laser Irradiation. R. Rani, A. Kundu, M. Balal, Goutam Sheet, K. S. Hazra*, Nanotechnology, 2018

Post-synthesis alloying is an important method for tailoring properties of two-dimensional (2D) transition-metal dichalcogenides (TMDs). In such a method, a binary crystal (i.e., MX₂) serves as a host lattice in which the partial substitution of a native element (e.g., X) with a dissimilar counterpart (e.g., X’) yields a 2D alloy (i.e., MX’_2xX_2(1-x)) [1]. However, the detailed mechanism of the post-synthesis alloying in 2D TMDs is largely unknown. Here, we show that native vacancy-type defects play a major role in such an alloying scheme. In a case study, we use MoSe₂ films as host crystals in which exchanging “Se” atoms with “S” atoms yields MoS_2xSe_2(1-x) alloys. Our study reveals that Se vacancies in host MoSe₂ films (1) mediate the Se-S exchange process, and (2) further take part in the lateral diffusion of S atoms within the MoSe₂ lattice. Accordingly, we show that alloying of CVD-grown MoSe₂ with abundant Se vacancies is more efficient than that of exfoliated MoSe₂ with fewer vacancies [2]. Practical consequences of this important conclusion will also be discussed.
[1] H. Taghinejad et. al “Strain Relaxation via Formation of Cracks in Compositionally Modulated Two-Dimensional Semiconductor Alloys,” npj 2D Mat. and App. (2018).
[2] H. Taghinejad et. al “Defect-Mediated Alloying of Monolayer Transition-Metal Dichalcogenides,” ACS Nano (2018).

Following the extensive investigation of the physical properties of two dimensional (2D) graphene, attention has turned to transition metal dichalcogenides (TMDCs) as 2D-materials for fundamental physics and device engineering, owing to the wide variety of possible crystal structures due to the large number of accessible chemical combinations as well as unusually long lifetime carrier transport properties. Because of the possibility low energy phase switching, TMDCs can also be considered as promising candidates for high-performance phase-change materials in the next generation of nonvolatile memory. In particular, phase transitions between the monoclinic phase of 1T’-MoTe2 (being stable at room temperature, RT) and the orthorhombic phase Td-MoTe2 (being stable below 250K) structures may be useful as ultrafast symmetric switching [1], since their crystal structures are similar, except for distortions along the a- or b-axes. Very recently Zhang et al. examined photoinduced structural transitions in Td-MoTe2, showing subpicosecond lattice symmetry switching by means of second harmonic generation, although the related coherent lattice vibrations were not well observed at RT [2]. Here we investigate the dynamics of coherent lattice vibrations in the monoclinic 1T’ phase using femtosecond optical pump-probe spectroscopy at RT. Preliminary results show a total of seven coherent phonon modes, which are approximately consistent with the corresponding Raman active modes. Among these modes, however, the lowest frequency mode at 0.38 THz, unexpectedly observed at RT, is primarily thought to exist only in the Td phase along the shear layer direction. Moreover, unlike the other six optical phonon modes, the amplitude of the shear phonon mode decreases with increasing pump fluence. The shear phonon mode is key in breaking the lattice symmetry, which results in the formation of a topological Weyl semimetal phase. We argue that our data demonstrate a possible ultrafast structural modulation between 1T’ and Td phases upon excitation of a shear phonon mode at RT.

Ref.
[1] E. J. Sie et al., Nature Vol. 565, 61 (2018).
[2] M. Y. Zhang et al., Phys. Rev. X Vol.9, 021086 (2019).

MXenes (Ti₃C₂), a new class of the two-dimensional (2D) transition-metal carbides and nitride, are currently at the foremost of 2D materials research. MXenes are promising new materials for next generation electrodes of the other electronics, in which various properties are required, including conductivity, transparency, mechanical reliability and conformability. Among the classes of various MXene family, Ti₃C₂T_x MXene, that firstly introduced from Drexel University, exhibits metal-like conductivity and hydrophilic surface including majority terminal group (-OH). Although MXene has been widely used as an electrode in many electronic devices such as supercapacitors, sensors, and batteries, it has not been applied to light-emitting diode (LED) devices yet. Herein, we prepared MXene (Ti₃C₂) thin film electrode with transparency, high conductivity and flexible for operating Organic Light-Emitting diode. This OLED with MXene electrode are very unstable in DC (Direct Current) operation but shows stable property in AC (Alternating-Current) operation. Our AC-driven OLED based on MXene electrode shows a high performance than device made from other 2D materials capable of solution processing. Moreover, the MXene electrode prepared on the non-rigid substrate (PET) with solution-processing exhibited both physical stability and durability under mechanical deformation. Therefore, we expect that this new MXene (Ti₃C₂) electrode with solution processable 2D materials will provide tremendous potential to fabricate OLEDs toward high-performance, flexible, and transparent as well as a variety of other functionality.

MXene, an emerging category of two-dimensional (2D) transition metal carbides and nitrides, have the potential to be high-performance and low-cost electrodes in organic field-effect transistor (OFET), due to the water dispersibility, the high conductivity, and the work function tunability. In this manuscript, we demonstrated a large-scale and uniform MXene electrode array formation on a plastic substrate for high-performance OFETs. The work function of MXene electrode was also effectively modulated through the chemical doping with NH₃. The resulting OFETs with MXene electrodes exhibited the excellent device performances such as maximum carrier mobility of ~1 cm² V^-1 s^-1 and on-off current ratio of ~10⁷ for both p-type and n-type OFETs, even though all electrode and dielectric layers were fabricated using solution process onto the plastic substrate. Furthermore, the MXene electrode-based complementary logic circuits such as NOT, NAND, and NOR have been integrated using p-type and n-type OFETs. The proposed method is expected to expand the application of MXene in other OFETs-involved electronic devices such as organic light emitting display and electronic skin.

The discovery of 2D graphene has attracted many researchers to synthesize a 2D transition metal dichalcogenides (TMDCs) materials. The TMDCs represented as MX₂ (M stands for Mo or W, Nb, Ta, etc. and X= S, Se, etc.) has the covalent bonds between metal and chalcogens that form the three layers which are held together by a weak van der Waals force. The stable layered structure formations in TMDCs is achieved by the absence of out-of-plane dangling bonds. Tungsten Disulfide (WS₂) is reported as one of the promising TMDC materials due to its various application in electronics, catalysis, optical, and other fields. The synthesis of WS₂ was previously reported by multiple techniques such as mechanical exfoliation, chemical exfoliation, chemical vapor deposition, etc. The limited research in large scale and high-quality synthesis of WS₂ materials restricts the use in the widescale application.
In this work, we are reporting the synthesis of a few layers WS₂ using plasma enhanced chemical vapor deposition (PECVD) technique at relatively low temperature (300°C). The PECVD technology is utilized to grow WS₂ thin film directly on the 4-inch Si-SiO₂ wafer. The quality of a synthesized WS₂ was confirmed using Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), high-resolution transmission electron microscopy (HR-TEM), and energy dispersive spectrometry (EDS) mapping techniques. The thickness of the synthesized WS₂ film was found to be around a few nanometers and showed multilayered structure (~5-6 layers). Our unique approach shows the distinct advantage of the synthesis of WS₂ thin film with high uniformity and purity with reproducibility for mass production. Further, the WS₂ thin film was used to check the hydrogen evolution reaction (HER) catalytic performance with direct growth on the glassy carbon electrode. The PECVD synthesized WS₂ shows promising HER performance comparable with previously reported materials.

Novel two-dimensional(2D) materials in which simple molecular units are used as building blocks are currently being synthesized through bottom-up approaches. In this work, aromatic six-membered ring units, namely benzene, pyridine, phosphinine, arsinine, 1,3-diazine, 1,3,5-triazine and borazine are each utilized to build 2D graphene-like monolayers. Our density functional theory calculations revealed the stability of different 2D layers made out of heterocyclic molecular units, for the first time.
The electronic properties of the holey materials as well as band gap tuning resulting from the different stacking of consecutive layers are also examined.
This set of 2D materials presents a whole new class of layered nanostructures with great potentials for applications in areas such as catalysis, gas separation, optoelectronics, and nanoelectronics.

Two-dimensional (2D) transition metal nitrides (TMNs) are emerging members in 2D family with promising potential for a range of applications. Their applications can be further extended to electronic and optoelectronic devices through the acquisition of high crystalline and large-area thin films. However, materials that meet such requirements have not been achieved using previous methods. Here, we report the synthesis of few-nanometer thin Mo₅N₆ crystals with satisfactory area and quality via chemical conversion of layered MoS₂ crystals. The structure and quality of the ultrathin Mo₅N₆ crystal are confirmed using transmission electron microscope, Raman spectroscopy and X-ray photoelectron spectroscopy. The lateral dimensions of Mo₅N₆ crystals are inherited from the MoS₂ crystals that are used for the conversion. Atomic force microscopy characterization indicates that the thicknesses of Mo₅N₆ crystal reduce to about 1/3 of the MoS₂ crystal, matching well with the crystal structure model. Electrical measurement shows the high conductivity of Mo₅N₆ (resistivity is 108.4 Ω/sq). In addition, this chemical conversion strategy is found versatile for the synthesis of various metal nitrides including W₅N₆, and TiN using their corresponding metal sulfides. Our strategy offers a new direction for preparing 2D TMNs with desired characteristics, opening a door for future exploration of fundamental physics and devices applications.

Recent application of mechanical exfoliation method to beta-gallium oxide (β-Ga₂O₃) has allowed various device structures through integration with different two-dimensional (2D) materials. Combining the unique properties of β-Ga₂O₃ and 2D materials can yield higher performance in (opto)electronic devices. The direct ultrawide bandgap of 4.6-4.9 eV and the superior electrical breakdown field make β-Ga₂O₃ especially advantageous for solar-blind ultraviolet (UV) photodetectors and power devices. The mechanically exfoliated β-Ga₂O₃ nanobelts can also be easily stacked on to the 2D flakes to form van der Waals heterostructures without any strain issues through dry transfer techniques just like other 2D materials; an exceptionally large lattice constant along one direction in β-Ga₂O₃ crystal structure allows easy mechanical cleavage along (100) direction. Therefore, the excellent electronic properties, high transparency or the atomic thickness of 2D layered materials such as graphene, transition metal dichalcogenides (TMDC) and hexagonal boron nitride (hBN) can be used to develop β-Ga₂O₃ device structures.
In this work, a 2D material was applied as an interlayer between the electrodes and channel of β-Ga₂O₃ devices, and the UV photoresponse properties of the fabricated devices based on van der Waals heterostructure were analyzed. First, hBN can be applied to β-Ga₂O₃ metal-semiconductor-metal (MSM) type UV photodetector. hBN, which is a 2D insulating material, can be used as a thin barrier in addition to the Schottky barrier between the source electrode and the channel to lower the dark current of the photodetector. Graphene with very high electrical conductivity can also be applied as UV-transparent electrodes to enhance the photocurrent of the MSM photodetector. Secondly, a 2D TMDC such as molybdenum disulfide (MoS₂) can be used to modify the contact properties and observe the effects on the photoresponse properties of β-Ga₂O₃ devices. Inserting a 2D interlayer between the metal and β-Ga₂O₃ would reduce the Schottky barrier and thus improve the ohmic properties of the metal contacts on β-Ga₂O₃. Further results and discussion will be presented in detail.

Recently, 2D material based multi-valued logic circuits, which can reduce power consumption and complexity of integrated circuits compared to binary logic circuits, have received significant attention in an effort to overcome fundamental limits of conventional silicon technology. Based on the development of 2D material preparation processes, heterogeneous integration of 2D materials can be simply achieved for logic gate design. However, reported three-valued NOT gates (ternary inverters) based on 2D materials require limited condition and/or complex fabrication process for obtaining distinctive three logic states. Here, we demonstrate CMOS ternary inverter circuit consisting of a series connection of three-heterogeneous 2D transistors (WSe₂-MoTe₂-MoS₂). Due to the ambipolar nature of MoTe₂ channel, which can drive both n-type transistor and p-type transistor depending on the applied gate voltage range, distinctive three logic states can be obtained by complementary operation between WSe₂/MoTe₂ heterojunction transistor and MoTe₂/MoS₂ heterojunction transistor. Furthermore, a quaternary inverter based on 2D multiple heterojunction is demonstrated by controlling trans-conductance of a part of MoTe₂ channel via spatially controllable surface charge transfer doping method.

Two-dimensional (2D-2D) heterostructures combine the unique intrinsic chemical, physical and (opto)electronic properties of different types of 2D materials. Not hampered by restrictions of lattice matching, this allows for tailored heterostructure designs for (opto)electronic devices and circuits. The semiconducting transition metal dichalcogenide (TMDC) WS₂ is interesting due to its direct bandgap in the visible range in combination with a high radiative quantum efficiency. Unlike graphene, WS₂ has a large optical absorption coefficient (10⁵-10⁶ cm^-1), thus its implementation is an ideal method to increase the external quantum efficiency of e.g. graphene-based optoelectronic detectors. Similarly, the TMDC MoS₂ could complement graphene-based electronics by a transistor channel material that can be pinched off efficiently.
The main preparation strategy of such 2D-2D heterostructures has been sequential stacking of the layered materials using wet or dry transfer methods. However, those mostly complicated transfer processes generate defects and leave residues at the interfaces of the stacked materials. In recent years, the well-established technique of metalorganic vapor phase epitaxy (MOVPE) has been used to deposit several types of 2D materials. MOVPE enables the development of a reproducible and scalable deposition process for high-quality 2D-2D heterostructures. Here, we investigated key process parameters, including growth temperatures, pressures and growth times for the single-step deposition of 2D-WS₂ and -MoS₂ on CVD (chemical vapor deposition)-graphene using MOVPE.
Monolayer and bilayer/multilayer graphene films were directly grown on sapphire (0001) using an AIXTRON CCS 2D system in 19 x 2" configuration. The metal-free CVD process was carried out at 1400 °C for 6 minutes using CH₄ with N₂ as carrier gas. The deposition of MoS₂ and WS₂ on these graphene samples was performed in an AIXTRON planetary hot-wall reactor in 10 x 2" configuration. Tungsten hexacarbonyl (WCO), molybdenum hexacarbonyl (MCO) and di-tert-butyl sulfide (DTBS) were used as metalorganic precursors (avoiding toxic hydride sources) and N₂ as carrier gas. All samples were characterized using Raman spectroscopy, photoluminescence spectroscopy (PL), scanning electron microscopy (SEM) and atomic force microscopy (AFM).
WS₂ was deposited for 20 h at 845 °C and 30 hPa on the different graphene samples. Deposition on monolayer graphene results in triangular domains up to 800 nm and a surface coverage of about 64 % whereas deposition on bilayer/multilayer graphene leads to crystallites up to 1.5 µm in size and 60 % surface coverage. Extending growth to 30 h yields nearly no increase of the monolayer WS₂ domain size on neither monolayer nor bilayer/multilayer graphene, but an enhanced bilayer WS₂ nucleation. Thus, the surface coverage increases only slightly up to about 71 % for monolayer and 67 % for bilayer/multilayer graphene. A rise of the total pressure at lower growth temperature (550 °C) in 10 h processes increases the domain size from about 140 nm (30 hPa) to 660 nm (200 hPa).
Using the same initial process parameters (845 °C, 30 hPa) for MoS₂ growth for 24 h, we find an enhanced lateral growth rate compared to WS₂. This is the same trend as previously observed for growth directly on sapphire. The MoS₂ monolayers are nearly fully coalesced on all graphene samples (> 98 % surface coverage). Unlike the growth of WS₂ on graphene or both TMDC directly on H₂-desorbed sapphire, MoS₂ crystallites exhibit a distinct hexagonal structure when deposited onto graphene. Only MoS₂ crystallites from secondary nucleation, which grow on top of monolayer MoS₂ domains, are also of hexagonal shape. The hexagonal shape generally hints on a close-to-ideal stochiometry at the crystallite edges and the absence of a significant substrate influence. Unlike the deposition of MoS₂ and WS₂ on sapphire, MoS₂ and WS₂ domains on graphene do not show any preferred orientation.

Two-dimensional materials (2DM) have received great attention because of their unique electrical, mechanical and optical properties. Especially, the weak van der Waals interaction and atomically sharp interface without dangling bonds make 2DMs suitable for heterojunction devices. Therefore, 2DMs-based heterojunction devices can avoid the problems of lattice constant mismatch and show high-performance compared to those fabricated by conventional epitaxy. Studies on heterojunction-based devices based on 2DMs have mostly focused on fundamental structures such as a single p-n junction diode, but not on active devices like heterojunction bipolar transistors (HBT).
In our study, we successfully fabricated double HBTs (pnp or npn) by vertically stacking n-type or p-type 2DM flakes. P-type (Black phosphorus or WSe₂) or n-type (MoS₂) flakes were vertically stacked in order of npn or pnp on to SiO₂/Si using a micro-manipulator. Then, electrodes (Ti/Au or Pt/Au) for the emitter, base, and collector were defined using standard e-beam lithography. The interface between each flake was clean without strain and bubbles as observed in the cross-sectional TEM images of the vertically stacked heterojunction structure. The formation of the two p-n junctions in base-emitter and base-collector was experimentally confirmed by the rectifying behavior in I-V characteristics. An npn double HBT showed excellent electrical characteristics with high current gain (β = ~100) which is comparable to those of conventional bipolar junction transistors. The detail of our experiment and results will be presented at the meeting.

Thin flakes of Cr1/3NbS2 are fabricated successfully via microexfoliation techniques. Temperature-dependent and field-dependent magnetizations of thin flakes with various thicknesses are investigated. When the thickness of the flake is around several hundred nanometers, the softening and eventual disappearance of the bulk soliton peak is accompanied by the appearance of other magnetic peaks at lower magnetic fields. The emergence and annihilation of the soliton peaks are explained and simulated theoretically by the change in spin spiral number inside the soliton lattice due to dimensional confinement. Compared to the conventional magnetic states in nanoscale materials, the stability and thickness tunability of quantified spin spirals make Cr1/3NbS2 a potential candidate for spintronics nanodevices beyond Moore’s law.

Among transition metal dichalcogenides (TMDCs), rhenium disulfide (ReS₂) has unique insensitive direct bandgap from monolayer to bulk [1]. Combining with this unique property and large area film fabrication technics can overcome the limitation of TMDCs. Herein, we synthesized polycrystalline ReS₂ via solid state reaction route and then sputtering target was prepared using spark plasma sintering. Uniform ReS₂ films were deposited on 2 inch wafer by sputtering with deposition temperature and time variations. Large area ReS₂ films were characterized and its morphology and thickness can be controlled systematically by changing deposition temperature and time. Basic p-n diodes with p-Si and ReS₂ film were fabricated and it demonstrated fast response time (7 us) and typical diode behaviors. These large area ReS₂ films have potential to expand the applications of TMDCs.

Two-dimensional (2D) transition metal dichalcogenides (TMDs) exhibit unique properties which create a platform to investigate novel phenomena and point to their application in several fields, such as electronics, catalysis and energy storage [1,2]. TMDs from Ti-, V-, and Cr-groups are most easily obtained in layered crystal structures and have been extensively explored as 2D materials [3], however a large diversity of crystal structures is known for compounds with the remaining transition metals, and the exploration of the physical and chemical properties of 2D dichalcogenides based on transition metals from Fe-, Co-, Ni-, and Cu-groups has been very limited. Therefore, in order to design novel two-dimensional materials based on TMDs and explore their structural, energetic, and electronic properties, we performed an investigation of 36 MQ₂ compositions (M = Fe, Co, Ni, Cu, Ru, Rh, Pd, Ag, Os, Ir, Pt, Au; Q = S, Se, Te), by means of density functional theory calculations with semi-local and hybrid exchange-correlation functionals, and van der Waals (vdW) corrections. For each composition, a set of 11 crystal structures experimentally reported for TMDs, both with layered and non-layered frameworks, has been employed. We found that layered phases are energetically favored for Ni-group compositions, and the octahedral intralayer coordination is the most favored among the layered phases. The layered crystals of Fe- and Ni-group materials have weak interlayer binding, dominated by vdW interactions, whereas in the remaining materials a strong interlayer binding with non-vdW chemical bonds is observed. The electronic band gaps of the lowest energy phases were screened, and selecting the lowest energy layered structure for each compound, 17 semiconductors were identified among their monolayers, all based on the Fe- and Ni-groups transition metals. The conduction and valence band offsets of the semiconductor monolayers were obtained, and the trends originate primarily from the changes on the M-d and Q-p energy levels. Based on the band alignments, we applied the Anderson rule to classify the semiconductors heterojunctions formed with the combinations of these monolayers, which mainly fall in the type-II classification, although Ni-group layers are more prone to the formation of type-I junctions. To explore the potential of these materials for application in solar energy harvesting, the power conversion efficiencies of solar cells based on type-II heterostructures were estimated, and values comparable to and higher than those estimated for the widely studied Mo- and W-based TMDs were found. Our findings provide important information to guide the design and search for novel 2D TMDs and van der Waals heterostructures. We acknowledge financial support from grant 2017/11631-2, São Paulo Research Foundation (FAPESP). R. Besse is recipient of PhD fellowship, Grant 2017/09077-7, FAPESP.

[1]. R. Besse et al. J. Phys. Chem C 122, 20483-20488 (2018).
[2]. N.A.M.S. Caturello et a. J. Phys. Chem. C 122, 27059-27069 (2018).
[3]. C.M.O. Bastos, R. Besse, J.L.F. Da Silva, and G.M. Sipahi. Phys. Rev. Mater. 3, 044002 (2019).

Two-dimensional (2D) materials, such as graphene and transition metal dichalcogenide (TMDs), have activated enormous interests in exploration of van der Waals (vdW) materials. One of key features of 2D materials, which enables their high performance, is dangling bond-free surfaces, since interlayer bonds in these materials are vdW rather than covalent. However, dangling bonds on edges in 2D materials still act as scattering centers and recombination centers when scaled down to less than a few tens of nanometers. In recent years, interest in 2D vdW materials has expanded to include one-dimensional (1D) vdW structures that consist of vdW bonded molecular chains.There are two types of 1D vdW structures; a true 1D structure with a pure vdW inter-chain bond and a quasi 1D structure with additional non vdW bonds among molecular chains. Unlike the 2D vdW materials that yield 2D atomic sheets upon exfoliation, 1D vdW materials produce various 1D nanomaterials such as ribbons, wires, and molecular chains. These 1D materials could contribute to many interesting physics, such as topological insulator, charge-density waves, superconductor-insulator transitions, ultralong ballistic phonon transports, low-frequency electronic noise and molecular-scale connectivity
In this study, we synthesized a vdW1D nanomaterials on a bulk scale stacked with chain-like 1D materials. The as-grown single crystal materials contains numerous single chains linked by vdW interactions. It was confirmed that the bundle of chains can be easily separated by mechanical exfoliation. Interestingly, the isolated 1D materials exhibit a quasi-two-dimensional layered structure. The variation of the work function depends on the thickness of the layered structure, as determined by scanning Kelvin probe microscopy (SKPM) measurements. Moreover, we implemented a field effect transistor (FET) based on nanoscale vdW 2D materials stacked with chain-like single crystal 1D materials to determine proper metal electrode.

Two-dimensional (2D) transitional metal dichalcogenide (TMD) such as MoS₂ and WSe₂ has been actively researched due to superior electrical and optical properties. Heterojunction of 2D based TMD and bulk semiconductor such as Silicon (Si), Germanium (Ge), and compound semiconductors, in particular, has attracted considerable attention in recent years due to new possibilities for the next-generation nanoelectronics and optoelectronics. Ge has large absorption coefficient at near IR, widely used in optoelectronics device. Combination of TMD and Ge will have a potential to absorb a broadband light. The broadband photodiodes with photosensitivity from UV to near IR can use various applications such as color imaging, night vision and environmental monitoring. In this work, we have investigated the electrical and optical characteristics of MoS₂/p-Ge photodetector for broadband photodetection.
The fabrication starts with a deposition of 100 nm thick SiO₂ on a cleaned p-Ge substrate by PECVD method. Some portions on the substrate were exposed using a wet chemical etching process to make the junction between MoS₂ and Ge. Surface passivation of Ge was carried out in (NH₄)₂S solution (40-48 wt.% in H₂O, Sigma-Aldrich) for 30 min at 90 °C. By conventional scotch tape method, few layers of MoS₂ were mechanically exfoliated from bulk MoS₂ (Graphene Supermarket, USA), and then they were transferred onto the substrate. 5/80 nm Ti/Au was deposited on the top surface of the SiO₂ as the electrode for MoS₂ by e-beam evaporation by conventional lift-off technique. Finally, 100 nm Al was deposited on the back of Ge.
The exposed area of the MoS₂ flake is approximately 1.67 × 10^-5 cm². Electrical property of the device was characterized using a semiconductor analyzer (B1500A, Keysight, USA) in a dark box. I-V curve of the MoS₂/Ge photodiode is measured from -10 V to 10 V. The device shows the excellent rectification characteristic with a rectification ratio up to approximately 104 within ±10 V. Threshold voltage is 4.07 V, higher than expected, due to barrier of junction by band pinning. I-V curves are measured for voltage ranging from -10 V to 10 V under incident power densities of 5, 15, and 50 mW/cm² with 466 nm DPSS (Diode Pumped Solid State) laser in N₂ ambience. We calculated photocurrent (I_ph) from I_ph = I_light - I_dark, where I_light is the current with the incident laser and I_dark is the current in the dark box, which is 0.52, 1.31, and 2.63 nA under the incident power densities of 5, 15, and 50 mW/cm², respectively, at V = -10 V.
We calculated the responsivity (R) as a function of the wavelength, obtained from 400 nm to 700 nm with the voltage of -10 V using a monochromator. The responsivity is calculated as R = I_ph/P_inc, where P_inc is the incident optical power. The responsivity shows a peak at 670 nm (174.4 mA/W). External quantum efficiency (EQE) is defined by R×hc/λ×q, where h is Planck’s constant, c is the velocity of light in vacuum, λ is the wavelength of the incident light, and q is the electronic charge. The obtained EQE peak is approximately 32.3% at 670 nm.
To confirm the response behavior about the incident light, the time-dependent current was measured under pulsed DPSS laser illumination (λ = 466 nm) with 5 mW/cm^-2. To characterize the speed of the photodiode by rise time t_rise, we plotted the rise of photocurrent for turning on the laser light. The t_rise is 11 ms, which defines the time required for the current to increase from 10% to 90% of the average current for turning on the laser. At the falling edge, we confirm the fall time t_fall of 15 ms, which defines as opposed to t_rise.
In conclusion, we have fabricated the p-Ge/MoS₂ hybrid photodiode and characterized their electrical and optical properties. We observed the excellent rectification characteristic and the responsivity of wavelength from 400 nm to 700 nm. We will measure the responsivity of the wavelength above 1000 nm and present the corresponding characteristics.

The World Health Organization recognizes heavy metal toxicity arising from contaminated water supplies as a major challenge affecting millions of people worldwide, especially in rapidly developing economies of Asia. The maximum permissible limit for heavy metals has been identified as 3 ppm for Cd²⁺, 10 ppm for Pb²⁺, and 20 ppm for Ni²⁺. Low-cost methods relying on ion exchange of Cd detection include the use of chalcogenide glasses, nanomaterials, and polymers that exhibit limits of detection (LoD) ~ 0.1-3 ppm. It is thus possible to detect Cd ions at human health-relevant ppm levels reliably, but cross-sensitivities of sensing films to other ions such as Pb, Ni, etc. complicate the analysis. Further, while combinatorial sensing of different ions by using arrayed detectors is possible, followed by principal component analysis, each detector has to be independently fabricated before system-level integration and packaging that interfaces each detector to the analyte, and readout electronics. This complication can be addressed by using a common transducing layer composed of a high mobility two-dimensional transition metal dichalcogenide (TMD). The ion exchange process at the surface of the sensing layer is expected to produce changes in local charge densities due to analyte-driven changes in the local potential. These charges can be conducted laterally to contacts by the underlying TMD layer, incorporated in the nanoscale form in a spin-coated sensing film to form ion sensitive electrodes (ISEs). In this work, we have grown proof of concept nanoscale heterojunctions composed of two dimensional n-MoSe₂ with a vertically grown p-Cu₂S sensing layer using a colloidal synthesis route via defect passivation of MoSe₂. Cu₂S is chosen intentionally because the cation exchange between Cu⁺ and Cd²⁺ is known to be favored in an aqueous solution. Powder X-ray diffraction (PXRD) data reveals the growth of hexagonal 2H MoSe₂ nanosheets and a hexagonal close-packed array of Cu₂S. Transmission electron microscopy (TEM) measurements were carried out on the dried sample to obtain a film thickness of approximately 1 nm. A micro glass slide was cleaned and bottom contacts (Cr (10 nm)/Au (50 nm)) were thermally evaporated through a shadow mask at a base pressure of 3.4 x 10^-6 Torr. An ink composed of the nanoscale heterojunctions was spin-coated to form a uniform film to complete the ISE. The properties of the material have been tested by electrochemical impedance spectroscopy (EIS) measurements (with Pt as counter and Ag/AgCl as reference electrode on a PAR PMC 2000 electrochemical workstation) and open circuit potential measurements (Keithley 6514 system electrometer). A series of test solutions were prepared by spiking the electrolyte with [Cd²⁺] in the range ~[1mM, 1nM]. Charge transfer resistance values derived by fitting Randle's equivalent circuit to the Nyquist plots were found to be approximately 5.5 kΩ near the LoD. These resistance values decrease with the increasing [Cd²⁺] due to increasing cation exchange at the Cu2S/analyte interface. Open circuit potential measurements revealed that a ~20mV/decade [Cd²⁺] increase in potential was generated with respect to the Ag/AgCl reference electrode with increasing [Cd²⁺]. A linear response to a logarithmic rise in [Cd²⁺] is found for the sensing device with an estimated LoD ~0.1 ppm for this initial device. Through careful tailoring of the interface between the sensing layer and the underlying 2D TMD, the strength of the molecular interactions can be enhanced, resulting in greater sensitivity to the analyte.

The reconfigurability of the electrical heterostructure featured with external variables, such as temperature, voltage and strain, enabled electronic/optical phase transition in functional layers has great potential for future photonics, computing and adaptive circuits. VO₂ has been regarded as an archetypal phase transition building block with superior metal-insulator transition characteristics. However, reconfigurable VO₂-based heterostructure and the associated devices are rare due to the fundamental challenge in integrating high quality VO₂ in technologically important substrates. In this report, for the first time, we show the remote epitaxy of VO₂ and the demonstration of a vertical diode device in graphene/epitaxial VO₂/single crystalline BN/graphite structure with VO₂ as a reconfigurable phase change material and hexagonal boron nitride (h-BN) as an insulating layer. By diffraction and electrical transport studies, we show that the remote epitaxial VO₂ film exhibits higher structural and electrical quality than direct epitaxial one. In the reconfigurable diode, we find that the Fermi level change and spectral weight shift along with metal-insulator transition of VO₂ could modify the transport characteristics. The work suggests the feasibility of developing single crystalline VO₂-based reconfigurable heterostructure with arbitrary substrates and sheds light on designing novel adaptive photonics and electrical devices and circuits.

Air-stable, crystalline, atomically-thin metals can be realized at the epitaxial graphene (EG)/silicon carbide (SiC) interface through a method termed confinement heteroepitaxy (CHet). This EG/SiC interface plays a crucial role in the creation of crystalline, 2-dimensional (2D) metals, and it provides a unique environment in which material structures with mixed interfacial van der Waals and covalent bonding can be formed. Through this thermal evaporation-based technique, intercalation of group-III elements (Ga, In) through epitaxial graphene can be achieved to create atomically-thin layers at the EG/SiC interface.
CHet is a unique EG intercalation method as it utilizes graphene defects and high-pressure (1-700 Torr) thermal evaporation. In this intercalation method, graphene defects created through exposure to O₂ plasma serve as entry points for a metal that is vaporized at high temperatures (>700°C). The defective graphene layers that result from plasma treatment correlate with the increase of the Raman D:G peak intensity ratios seen between as-grown and plasma-treated EG. During intercalation, the engineered graphene defects are shown to heal. This can be observed in the significantly decreased Raman D peak intensity measured post-intercalation. Regardless of the metal intercalated, CHet also produces a 3-4× increase in the Raman G peak intensity. This effect is attributed to plasmonic resonance. Due to their potential for high optical sensitivity and tailorability, 2D metals are promising materials for plasmonic applications.
Developing efficient photosynthetic systems that can economically convert solar energy into chemical energy on a large scale is crucial for the realization of a sustainable economy. The production of hydrogen gas by reducing water is a promising solar energy conversion technique, and it has been extensively researched in semiconductor-based photoelectrochemical devices. Semiconductors are often the foundation for solar energy-harvesting technologies because they have the ability to effectively capture photons. Due to the reflective properties of bulk metals, they are not generally used for such applications. Metals on the nanoscale, however, have proven to be effective at trapping photons due to local surface plasmon resonance. Research involving plasmon-induced charge separation for photocatalysis without the use of semiconducting materials is limited. Here, we explore how 2D allotropes of 3D metals perform as photocatalysts and how their tunable plasmonic properties can play a role in developing reaction selectivity.
In order to test the photocatalytic performance of the 2D metals realized through CHet, a white-light source is utilized in conjunction with a three-electrode system in an electrochemical cell. Because plasmons can decay via uphill electron transfer to a chemical species in direct contact, plasmon-induced charge separation (PICS) can be utilized to convert light energy into electrochemical energy. If the induced charge separation is large enough, it can generate an electronic potential that is suitable to drive the electrochemical production of fuels such as hydrogen gas. Standard measurements including electrochemical impedance spectroscopy, cyclic voltammetry, and linear sweep voltammetry completed with and without the use of light serve to elucidate the plasmonic properties of 2D metals and the role that PICS can play in catalyzing reactions that yield carbon-neutral fuels.

Recently, two-dimensional nano-materials such as graphene and transition metal dichalcogenide (TMDC) have been explored with substantial interest due to their potential in various applications including nanoelectronics, optoelectronics, and bioelectronics. One type of TMDCs, molybdenum disulphide (MoS₂) is regarded as a promising material for optoelectronic application because MoS₂ has tuneable bandgap whose range is from 1.2 eV for monolayer to 1.8 eV for multi-layer. Despite the proper band gap energy for absorbing wide-range of light covering from visible light to ultraviolet, the phototransistors based on MoS₂ channel layer have some issues regarding relatively low photoresponsivity and photosensitivity due to their low efficiency of light absorption resulting from ultrathin channel layer. In this sense, intensive efforts have been widely devoted on improving the photo-related characteristics of MoS₂ phototransistors.
In this work, we propose a simple method to enhance the photo-detecting performance of MoS₂ phototransistors where the reduction treatment is introduced. The reducing agent used in this research is tannic acid (C₇₆H₅₂O₄₆) which is a nature-friendly material extracted from various vascular plants. The reducing reaction by tannic acid is originated from its monomer with dozens of hydroxyl groups (-OH) which forms hydrogen bonds with oxygen. The reduction treatment is simply processed by dipping the device in tannic acid solution. During the dipping process, hydroxyl groups in tannic acid make hydrogen bonds with oxygen atoms that intrinsically exist in the lattice of MoS₂ by substituting the sulfur atoms. Then, the oxygen atoms bonded with tannic acid monomers are detached from lattice of MoS₂. Those broken Mo-O bonds result in increased number of sulfur vacancies which are relevant for the formation of sub-gap states in MoS₂ band gap. Thus, sub-gap states formed by sulfur vacancies in MoS₂ induce narrower effective optical band gap energy and light absorption capability could be improved.
The multilayer MoS₂ flakes were transferred onto the thermally oxidized heavily boron-doped silicon wafers by mechanically exfoliating from a bulk MoS₂ crystal. Then, titanium and gold electrode sequentially deposited by electron-beam evaporation. After completing fabrication of MoS₂ transistor, the device is dipped in 0.3 M tannic acid solution for 10 min.
We have measured the electrical and photo-detecting performance of tannic acid treated MoS₂ phototransistor. The improvement or degradation in electrical performance was not observed in MoS₂ phototransistor after reducing treatment by tannic acid. But, the difference between dark current and photoinduced current under red, green, and blue laser illumination becomes more distinguishable because of more efficient light absorption in tannic acid treated MoS₂ layer. As a result, both photoresponsivity and photosensitivity of the tannic acid treated MoS₂ phototransistor were improved as compared to the device without treatment. With an intensity of 10 mW/mm², photoresponsivity increased from 126.85 A/W to 807.36 A/W under red light, from 109.00 A/W to 819.29 A/W under green light, and from 166.44 A/W to 928.25 A/W under blue light. Also, with same intensity of light, photosensitivity increased from 9.51 to 243.93 under red light, from 16.23 to 527.05 under green light, and from 26.51 to 1700.98 under blue light. Therefore, we proposed a novel method for improving the performance of MoS₂-based phototransistor with simple reduction method which makes it possible to be applied in various optoelectronic applications.

Optical and electronic properties of 2D nanomaterials such as Graphene, Transition metal di-chalcogenides (TMDCs), Phosphorene etc. are already emerging field of research for their wide range of applications in nano-electronics, opto-electronic, solar cell, electrochemical activity etc. We have developed a novel and easy way to create various nanostructures on wide range of 2D materials in a single step process by using simple focused laser beam of a Raman spectrometer at low power irradiation. The feature size can go down up to diffraction limit of the laser used and the shape of the etched point can be used to find the orientation of the lattice structure of the material. In this presentation we focus on various applications of such nanostructures, fabricated on 2D by simple laser irradiation technique. We will demonstrate that nanostructuring on MoS₂ flake can create periodic modulation in capacitive response. Optimized geometry of these nanostructures alongwith selective deposition of Au nanoparticles demonstrate ultrasensitive SERS. These nanostructures were capable of enhancing interestingly we have found that the edges of such nanostructures are more catalytically active for electrochemical deposition of Au nanoparticles. Using this route we were also able to create nanostructures on Phosphorene flakes and rGO flakes. Controlled formation of nano-voids array on few layer BP flake induces enhanced local electric field (hot spots) at the vicinity of the nano-voids, resulting in ~ 30% Raman intensity enhancement. Such nano-voids induced hotspots on BP flake open up a new species of metal free SERS substrate, demonstrating pronounced enhancement in Raman signal of Rhodamine B as high as of the order of ~10⁴. Nano-patterning on rGO flakes leads to the observation of radiation pressure on solid-air interface.

1) Controlled formation of Nanostructures on MoS2 Layers by Focused Laser Irradiation, R. Rani, D. Sharma, N. Jena, A. Kundu, A. D. Sarkar, K. S. Hazra, Applied Physics Letter 110, 083101, 2017
2) Graphene Oxide Demonstrates Experimental Confirmation of Abraham Pressure on Solid Surface, A. Kundu, R. Rani and K. S. Hazra, Scientific Reports, 7, 42538, 2017.
3) Modulating capacitive response of MoS2 flake by controlled nanostructuring through focused laser irradiation, Renu Rani, Anirban Kundu, Mohammad Balal, Goutam Sheet and Kiran Shankar Hazra, Nanotechnology, 29,345302, 2018

The discovery of 2D layered materials has opened new frontiers in science and technology. The implication of these 2D nanomaterials in gas sensing, has gained tremendous attention owing to their outstanding virtues such as high surface to volume ratio, unique electronic and structural properties. Further, the high surface reactivity of 2D materials, offers potential for room temperature gas sensing leading to low power consumption, which is highly desirable for practical applications. Among these 2D nanomaterials, the Transition Metal Dichalcogenides (TMDC’s) such as MoS2 and WS2, have been explored for the detection of various gases including NH3, NO, NO2, etc. However, these TMDC’s are prone to degradation on exposure to ambient environment due to the significant oxygen adsorption on its surface. This limits the practicability of a sensor in real environment. The engineering of these 2D materials by incorporating another nanomaterial can provide solution to this aforesaid problem. This strategy of forming nanohybrids can not only provide the air-stability, but also improve the overall performance of the sensor.
In this work, we demonstrate the fabrication of a novel nanohybrid of Iron oxide nanoparticles (NP)- decorated WSe2 nanosheets (NS) for air-stable room temperature detection of ammonia with very high response magnitude and lowest limit of detection. The aforesaid nanohybrid was fabricated using modified co-precipitation method. At room temperature, the sensor demonstrated significant response of 510% to 3 ppm ammonia while achieving the lowest limit of detection (llod) of 50 ppb, with a response of 2.4%. The iron oxide – Wse2 nanohybrid sensor response was dramatically higher when compared to pristine WSe2, as the sensing material which exhibited only 53.5% response towards 3 ppm of ammonia, and 300ppb as llod. In addition, we observed improved recovery of the nanohybrid sensor (148s recovery time in contrast to 436s recovery time for pristine WSe2) by incorporating iron oxide nanoparticles. These findings suggest that the decoration of nanoparticles can aid in tuning the properties of 2D layered materials and hence can open up new avenues for several applications.

We present a novel dual gate black phosphorous (BP) photodetector based on a polymer electrolyte top gate and a conventional silicon bottom gate for improved performance in terms of operating voltage and photoresponsivity in the visible and near infrared (NIR) regimes. Various photodetectors based on 2D materials and their heterostructures have been reported for detection in ultraviolet, visible and near-infrared bands, however, there is still a need for broadband low-cost photodetectors with high photocurrent and photoresponsivity of the order of hundreds of A/W and at low operating voltages for integrated photonics and optoelectronics applications.

In this work, we use polyethylene oxide and cesium perchlorate (PEO: CsClO4) solid polymer electrolyte for top gating in the electrical double layer transistor (EDLT) mode. The polymer electrolyte leads to the formation of an electric double layer at the gate electrode-polymer electrolyte interface and the polymer electrolyte-BP interface and it allows accumulation and migration of ions at these two interfaces. The polymer electrolyte gate insulator is shown to greatly reduce the operating voltage to less than 2V as compared to conventional silicon bottom gate BP photodetector that requires voltages greater than 20V. The high specific capacitance also increases the number of carriers induced in the channel at a given gate voltage. In addition to the low operating voltage, the dual gate BP photodetectors using polymer electrolyte are expected to provide high photocurrent and photoresponsivity by enhancing the carrier depletion using polymer electrolyte gating and a field tunable photoresponse for broadband photodetection.

Novel bottom-up processes allowing to synthetize a 2D sheet of Si, best known as silicene, could represent the key strategy to overcome the limits of conventional top-down technologies in the semiconductor industry. Silicene allows to reach the ultimate target for thickness scaling while combining an ultra-high carrier mobility with the unique opportunity to tune the bandgap.
However, silicene oxidizes rapidly when exposed to ambient conditions, therefore requiring a reliable passivation process. Graphene represents an ideal passivation material, due to its inertness, high flexibility and outstanding impermeability, as well as being non-Raman active in the fingerprint region of silicene, allowing a precise analysis of the structural properties of the encapsulated layer.
We report the successful passivation, for up to 48 h, of silicene by few-layers graphene (FLG) flakes.
The monolayer silicene was epitaxially grown on a well-ordered Ag(111) film on mica by exposing the substrate, kept at a temperature of 530 K, to a constant flux of Si (≈ 0.02 ML/min). After the growth, silicene formation was verified via LEED analysis.
For the passivation, natural graphite was mechanically exfoliated on top of a vacuum-compatible polyimide adhesive tape and inserted into a specially designed UHV chamber. Then, FLG flakes were mechanically transferred atop of the silicene monolayer by direct exfoliation in UHV.
The encapsulating flakes allowed us to perform a subsequent ex-situ Raman characterization of the 2D silicon layer. The analysis was performed at room temperature in a back-scattering geometry using a confocal μ-Raman setup. The collected spectrum is characterized by two clear peaks located at 216 cm^-1 and 515 cm^-1, consistent, respectively, with the predicted out-of-plane and in-plane vibrational modes for silicene. Polarization-dependent measurements show that the 216 cm^-1 peak is related to a fully symmetrical phonon mode, a clear indication of the 2-dimensionality of the passivated silicene layer.

In order to realize the industrialization of chemical vapor deposition (CVD) graphene, it is essential to develop a method capable of characterization in real time as well as mass synthesis. In this study, we report that the reflection mode of confocal laser scanning microscopy (CLSM) enables highly visible and distinct image of CVD-grown graphene on Cu compared to dark field optical microscopy (DM). Reflectance contrast, R_C, defined as the intensity ratio of light reflected from graphene grown on Cu to bare Cu, depends on the incident laser wavelength, of which maximum was obtained at 405 nm in this experiment. Remarkably, it was observed that R_C of CVD-grown graphene on Cu varies with doping and crystallinity of graphene and decreases as the intensity of D peak increases in Raman spectrum. The R_C over 1 calculated using the Fresnel’s interference formula with the optical conductivity of graphene estimated by tight-binding model is in good agreement with the measured R_C. Consequently, it is demonstrated that R_C of CVD-grown graphene on Cu is affected by the quality of graphene related with its optical conductivity. Based on these results, we suggest that the reflection mode CLSM can be applied as a powerful tool for in-situ monitoring of CVD graphene growing on Cu by analyzing R_C of graphene.

Selective conversion of CO2 into a single chemical product remains one of the greatest challenges in CO2 reduction. Transition-metal dichalcogenides have shown potential for electrocatalytic conversion of CO2 to CO with high selectivity in ionic liquid-based electrolytes, but in fully aqueous environments the evolution of hydrogen is commonly observed to be the dominant catalytic reaction.
Despite much promise, further progress in the design of layered two-dimensional materials for the CO2 reduction reaction (CO2RR) is needed to enable the sustainable synthesis of chemical fuels from CO2 feedstocks. In this work, we demonstrate th at the valence band maximum and the conduction band minimum of transition-metal dichalcogenides can be tuned via chemical alloying to form MoS2(1-x)Se2x alloys (where x ranges from 0 to 1). The band structure of few-layered transition metal dichalcogenides was tuned by changing the chemical composition of the chalcogens (sulfur, S and selenium, Se) in the resulting alloy. Systematic studies of electrochemical CO2 reduction with various MoS2(1-x)Se2x compounds demonstrate that the distribution of chemical products generated by the electrocatalyst can be controlled by the chemical composition of the alloy. Specifically, we find that the MoS1.48Se0.52 compound exhibited the best selectivity of all studied alloys, with a Faradaic efficiency for CO production of ~27% at an applied potential of –0.2V vs. RHE in fully aqueous 50mM K2CO3 electrolytes. In general, the alloy compounds show superior CO2RR performance relative to pure MoS2 or MoSe2-based electrocatalysts. Such a trend in synthesized MoS2(1-x)Se2x alloys is consistent with that observed from bulk crystals, indicating that chemical alloying of layered two-dimensional materials is a promising approach to tuning the selectivity of transition-metal dichalcogenides for CO2RR in aqueous electrolytes.

Group IV transition metal dichalcogenides (TMDs) have recently drawn attention due to their predicted ultra-high charge carrier mobilities (1-3000 cm2/Vs) and bandgap in the VIS-NIR region. These materials are considered to be highly promising for high-performant FETs and ultra-responsive photodetectors, in addition to their possible use as energy materials and in catalysis. However, very little work has been reported on the synthesis of two-dimensional few-layered Group IV-TMDs. Here, we report the growth of ZrS2 and ZrS2xSe2-2x thin films over cm2 areas on SiO2/Si and Au foil starting from ZrCl4 and sulphur powder as precursors. The interest in ZrS2xSe2-2x results form the possibility to tune a low energy band gap. The synthesis has been performed via chemical vapour deposition (CVD) using a tubular furnace where thickness control has been achieved tuning the growth time. Our ZrS2 thin film has been characterized via Raman spectroscopy, which revealed two fingerprint peaks at 250 cm-1 and at 335 cm-1 corresponding to the Eg and the A1g peak respectively. HRTEM and SAED characterizations have confirmed the hexagonal crystal structure and the atomically thin thickness of 2-3 layers of individual platelets. XPS and EDX has provided evidence of the expected stoichiometry and composition. Furthermore, the material exhibits photoluminescence (PL) signal at 1.4 eV, which matches with theoretical predictions ranging from 1.2 to 2 eV for the Γ-M indirect bandgap. Annealing the ZrS2 film with Se we can achieve the synthesis of ZrS2xSe2-2x, which shows morphological characteristics similar to the starting material and Raman peaks at 250 cm-1 and 300 cm-1.

Design and discovery of layered materials for next-generation Lithium-ion batteries require a thorough understanding of the role of structure (bonding environment) on the energetics of Lithium (Li) intercalation and the accompanying structural accommodations. In this study, density functional theory (DFT) calculations have been used to investigate the energetics and mechanisms of lithiation/delithiation in thin films of MoS₂. The work aims at presenting a comprehensive study on thermodynamics of lithiation in order to sequentially investigate the process of Li-ion intercalation between layers of MoS₂ films and the resultant structural transformations.

Simulations have been performed to investigate the energetics of Li-ion binding and diffusion in the 2H and 1T phases of MoS₂. Li binding is observed to be stronger in 1T phase and hence, the activation energy for diffusion is higher. Sequential insertion of Li suggests that intercalation of one Li atom in the spacing between two MoS₂ layers results in easier diffusion of successive Li atoms in the same gap until all of the tetrahedral binding sites are occupied. Further lithiation results in Li intercalation into successive gaps between 2H-MoS₂ layers and promotes phase transformation of adjacent 2H layers to 1T phase. Thus, partial lithiation across layers creates hybrid 2H-1T phase with a stable phase boundary perpendicular to c-direction. This 2H to 1T phase transformation of MoS₂ layers between the Li layers is attributed to a change in charge density around the transition metal following ionization of Li to Li⁺ cation.

Such phase transformations accompanying lithiation are found to be irreversible and hence, quite detrimental to the cycling stability of batteries. With the understanding gained on the binding of Li, various combinations of transition metals and chalcogenides are tested to identify layered materials which render or resist phase transformation during lithiation. The energetics of lithiation in the various phases of layered materials will be presented. These results parallel insights gained from in-situ TEM studies and can shed light on Li-ion intercalation/de-intercalation mechanisms in thin films of layered materials. This work is being supported by NSF grant No. 1820565. CINT is an Office of Science NSRC User Facility operated for the U.S. DOE.

Molybdenum disulfide (MoS₂) is a layered material that is expected to be used in various devices such as TFT due to the presence of band gap (bulk to monolayer: 1.2 to 1.8 eV[1]), high mobility (mono to multilayer: 380 to 700 cm²/Vs[1,2]), high flexibility, transparency and stability. However, establishing a film fabrication method with high quality and high productivity is a major issue for its practical applications. Therefore, we focused on cold-wall MOCVD method as a deposition technique. In this film fabrication method, utilization efficiency of the precursor is high since the reaction occurs only at the substrate. In addition, fine controllability of the film quality can be achieved by facilitating control of various deposition conditions such as supply amount of Mo and S by using the organic precursors having high vapor pressure at low temperature.
We have previously proposed i-Pr₂DADMo(CO)₃ as a novel Mo precursor. We have reported so far that i-Pr₂DADMo(CO)₃ decomposes below 300°C, and 1L MoS₂ without C and N contamination can be fabricated on SiO₂/Si substrate at low temperature of 250°C with the Mo precursor and sulfur source (t-C₄H₉)₂S₂[3]. In this research, with the aim of further improving the quality of the MoS₂ film using i-Pr₂DADMo(CO)₃, various film fabrication conditions were controlled and the influence on the film quality was investigated. XPS, AFM and Raman spectroscopy were used as evaluation techniques of the samples. It was suggested that the increase of substrate temperature contributes to the expansion of the grain size through the increase of Mo migration. High substrate temperature also increases the deposition rate by accelerating the thermochemical reaction of i-Pr₂DADMo(CO)₃ and (t-C₄H₉)₂S₂. It was also suggested that higher deposition pressure contributes to the increase in deposition rate caused by larger degree of supersaturation. Higher deposition pressure also decreases the deposition rate caused by lower concentration of the precursor in the chamber. Thus an optimum value is considered to exist for high-quality film synthesis.
This work was partly supported by JSPS KAKENHI Grant Number 18F22879 and JST CREST JPMJCR16F4
[1] B. Radisavljevic, M.B. Whitwick, and A. Kis, Appl. Phys. Lett. 101, 043103 (2012)
[2] S. Das, H.-Y. Chen, A.V. Penumatcha, and J. Appenzeller, Nano Lett. 13, 100 (2013)
[3] S. Ishihara, et al., MRS Advances 3, 379 (2018)

The alloying of transition metal dichalcogenides (TMD) and their various properties are expected to be a significant factor for the TMDs to expand in its applications. Although TMD alloys show appealing electrical, optical, and chemical properties, the synthesis method is still at its early stage. The alloy materials under research are mostly MoS_2(1-x)Se_2x, Mo_1-xW_xS₂, WS_2(1-x)Se_2x, and so on. There have not been many reports on the alloy that incorporate Te in the system. In the past reports, we have demonstrated that Mo-S and Mo-Te bonds can be formed in the product film, but physical structure evaluation was not sufficient. Thus in this study, the alloy films were evaluated with XRD in addition to XPS. The physical structure evaluation revealed that the as-sputtered samples, i.e. samples prior to thermal treatment, shows alloy formation depending on the sputter power of MoTe₂ targets. Use of compound MoS₂ target promotes layer structure formation at the sputtering stage, however, use of merely-sintered (not compound) MoTe₂ target may disrupt this behavior. It was also revealed that different chalcogen composition in MoS_2(1-x)Te_2x requires different optimal temperature in the post-deposition thermal treatment. For the sulfurization samples, the alloy showed phase segregation, MoS₂ and MoS_2(1-x)Te_2x, when the temperature is well over or under 400°C. In other words, when the sulfurization temperature is at 400°C, the alloy showed uniform distribution of Te and S throughout the film. On the other hand, for the tellurization samples, the alloy with no phase segregation was observed at 450°C. This shows that higher Te composition for MoS_2(1-x)Te_2x may require thermal treatment with higher temperature.

This work was partly supported by JST CREST Number JPMJCR16F4, Japan. This work was also partly supported by JSPS KAKENHI Grant Number 18J22879.

Monolayers of Transition Metal Dichalcogenides (TMDs) present a giant leap forward towards the realization of semiconductor devices with atomic scale thickness. As a natural consequence of their two-dimensional character, TMDs exhibit a reduced dielectric screening, leading to the formation of unusually stable excitons. Excitons dominate the optical response as well as the ultrafast dynamics in these atomically thin materials. As a result, a microscopic understanding of excitons, their formation, relaxation and decay dynamics becomes crucial for technological applications of TMDs. A detailed theoretical picture of the internal structure of excitons and their scattering channels allows for a controlled manipulation of TMD properties enabling an entire new class of light emitters, absorbers and detectors.

We provide a fully quantum mechanical description of momentum- and energy-resolved exciton dynamics in TMD mono- and hetero-bilayers [1-3] unraveling the many-particle processes governing the ultrafast dynamics of excitons in these materials. Our approach provides novel insights into exciton-phonon and exciton-photon interaction mechanisms and the impact of dark exciton states. This includes the formation of bound excitons out of a free electron-hole gas up to their eventual radiative recombination. Furthermore, we study the formation of interlayer excitons in Van-der-Waals heterostructures, i.e. spatially separated excitons composed of electrons and holes located in different layers of two stacked TMDs.

Our theoretical model allows us to predict fundamental formation and relaxation time scales as well as spectral features accessible in ultrafast pump-probe experiments. In particular, we show that the relaxation of hot excitons throughout the Rydberg series of excited exciton states gives rise to population inversions between s- and p-type excitons resulting in transient optical gain in the THz spectrum [1]. Moreover, in a joint experiment-theory study on Van-der-Waals heterostructures [3], we demonstrate how the binding energy as well as the time resolved formation of interlayer excitons can be measured using mid infrared pulses.

[1] Brem, Selig, Berghaeuser, Malic, Scientific reports 8(1), 8238 (2018)
[2] Ovesen, Brem et al., Communications Physics, 2(1), 23 (2019)
[3] Merkl, … Ovesen, Brem, Malic, Huber, Nature Materials (2019)

Semiconductor heterostructures formed by vertically stacked two-dimensional transition-metal dichalcogenides (TMDs) sustain interlayer excitons - electron-hole Coulomb bound states between electrons and holes spatially separated in different monolayers [1, 2]. In addition, heterostructures formed by vertically stacking two different monolayer TMDs (heterobilayers) can feature a spatially periodic moiré superlattice in which a periodic potential landscape for excitons is controlled by the lattice mismatch and/or the relative angle twist between the constituent monolayers. For moiré periods larger than the exciton Bohr radius, the moiré potential minima can act as smooth quantum-dot-like confining potentials [3, 4]. Bilayer TMDs offer an additional degree of freedom to engineer the properties of moiré-trapped interlayer excitons in vertical semiconductor heterostructures. In addition to the spin-valley coupling characteristic of monolayer TMDs, bilayer TMDs also present a layer pseudospin [5]. Because the bottom layer in an AB-stacked bilayer is a 180° in-plane rotation of the top layer, AB-stacked bilayer TMDs show spin-layer locking, i.e., a spin-valley configuration that is locked to the layer pseudospin.

In this work, we exploit the spin-layer locking of bilayer 2H-MoSe₂ to probe the spin-valley properties of interlayer moiré-trapped excitons in a WSe₂/MoSe₂ heterostructure. Low-temperature (4 K) magneto-optical spectroscopy reveals the coexistence of two species of moiré-trapped excitons with different spin-layer pairings for electrons in MoSe₂. Both trapped exciton species yield narrow (< 100 meV) linewidths and saturate with increasing excitation power, hallmarks of quantum dots. Although for both species of interlayer trapped excitons the hole is localized in the WSe₂ layer, the electron in bilayer MoSe₂ can be localized either in the bottom or top layer. The natural AB stacking of bilayer 2H-MoSe₂ gives rise to AA (AB) relative stacking between the WSe₂ and the bottom (top) layer of MoSe₂, respectively. Due to the spin-layer locking in bilayer MoSe₂, electrons in the bottom (top) MoSe₂ layer present parallel (antiparallel) spin and valley contributions, which results in an effective layer-locking of the Landé g-factors of the trapped excitons. Our results reveal that the trapped excitons present g-factors that are homogeneous across the sample and take only two values: -7.0 ± 0.6 and -15.76 ± 0.13 for Moiré-trapped excitons originating from the AA- and AB-stacked WSe₂/MoSe₂. Finally, helicity-resolved photoluminescence measurements confirm that the trapped excitons exhibit both strong circular and valley polarization, which indicates that the confinement potential preserves the rotational C₃ symmetry and the trapped interlayer excitons inherit the spin-valley properties of the constituent monolayers.

[1] P. Rivera et al., Nat. Commun. 6, 6242 (2015).
[2] A. Ciarrocchi et al., Nat. Photonics 13, 131 (2019).
[3] H. Yu et al., Sci. Adv. 3, e1701696 (2017).
[4] K. L. Seyler et al., Nature 567, 66 (2019).
[5] A. M. Jones et al., Nature Physics 10, 130 (2014).

In this paper we review recent developments in our understanding of the optically excited states of monolayer semiconductors and heterostructures in the family of the transition metal dichalcogenides. In particular, we will focus on the exciton character of the excited states, discussing our current understanding of the 2-, 3-, 4-, and 5-body exciton, charged exciton, biexciton, and charged biexciton species. We will also describe the presence of spin-forbidden dark excitons and how they may be observed optically. In addition to the range of different excitonic states, the 2D semiconductors feature strong electrostatic coupling to the external environment. We will describe the corresponding influence of external dielectric screening on the band- and exciton structure of the materials and how we can use this effect to tune locally the properties of 2D semiconductors. Finally, we will consider the new states that arise in 2D heterostructures and how they can be modified by electric fields and the relative crystallographic orientation of the component layers.

The ability to control the flow of excitons underlies high-speed excitonic devices and energy transport [1,2]. Two-dimensional (2D) semiconductors are well suited for these applications in that their large exciton binding energy enables room temperature operation and their enormous stretchability provides engineerable exciton potential surfaces [3]. Since excitons are hardly respond to in-plane electric field due to their charge neutrality, the control was limited to interlayer exciton with quantum confined Stark Effect in vertical heterostructure [4,5]. Strain-engineered bandgap modulation offers a new tool for monolayer excitons. Recent studies have observed excitonic flux in predetermined non-uniform geometries, an important remaining challenge concerns active and reversible controllability [6]. Here, we tune the bandgap of suspended 2D semiconductors dynamically by applying local strain gradient with a nanoscale tip. This method enables active and reversible control of the exciton drift over several hundreds of nanometer at room temperature through the energy gradient, which has been demonstrated by wide field imaging and time resolved photoluminescence measurement. Our results pave the way for broad applications including broadband photovoltaic devices and efficient excitonic devices.

[1] Butov, L. V. Excitonic devices. Superlattices Microstruct. 108, 2–26 (2017)
[2] Feng, J., Qian, X., Huang, C.-W. & Li, J. Strain-engineered artificial atom as a broad-spectrum solar energy funnel. Nat. Photonics 6, 866–872 (2012)
[3] Manzeli, S., et al., 2D transition metal dichalcogenides. Nature Reviews Materials 2, 17033 (2017)
[4] High, A. A., et al., Control of exciton fluxes in an excitonic integrated circuit. Science 321, 229–231 (2008)
[5] Unuchek, D. et al. Room-temperature electrical control of exciton flux in a van der Waals heterostructure. Nature 560, 340–344 (2018)
[6] Leon, D. F. C. et al. Exciton transport in strained monolayer WSe2. Applied Physics Letters 113, 252101 (2018)

Single photon emission, the phenomena where excited electron states emit one photon per excitation, serves as the basic building block of quantum optics technologies. In the solid state, many material systems exhibit single photon emission such as diamond, hexagonal boron nitride, and monolayer tungsten diselenide. In the first two cases, the source of the single photon emission has been positively identified as deep-defect color centers in the crystal lattice. In the latter, however, single photon emission has only been observed at cryogenic temperatures, and when the monolayer is subjected to inhomogeneous strain. It has been suggested that the quantum dot single photon emission is due confinement of excitons, which the inhomogeneous serving as the source of the confinement potential.

In this work, we find evidence that strained nanobubbles in monolayer WSe₂ host quantum dot-like confined excitons at room temperature. Using hyperspectral nano-photoluminescence mapping with below bandgap excitation, we identify a “doughnut shaped” energy profile, showing significantly lower energy emission at the edges of the nanobubbles versus the center. This observation is consistent with recent theory predicting the edges of the nanobubble induce atomic scale wrinkling, providing the highest confinement of the excitons. Additional work using time resolved photoluminescence spectroscopy shows enhanced lifetime of the exciton inside the nanobubbles, consistent with quantum confined excitons. Our observations provide evidence that strain inside nanobubbles can confine excitons, and further that this confinement is present at room temperature , raising the possibility that single photon emission may also persist as it does in colloidal quantum dots.

The development of optoelectronic devices based on transition metal dichalcogenides (TMDCs) will require a deep understanding of the spatial and temporal dynamics of their charge carriers and excitons. Whilst the properties of most TMDCs are isotropic, in a few cases a reduced crystal symmetry leads to unusual anisotropic properties, a degree of freedom that has remained hitherto largely unexplored. TMDCs are also typically technologically relevant only in the mono to few layers limit. In this respect, ReS₂ and ReSe₂ are unique members of the TMDC family. Their disordered 1T phase make them anisotropic semiconductors allowing them to support robust linear-exciton with different dipole orientations at room temperature. Besides, these excitons persist from the bulk down to the monolayer. Here, we study the anisotropic diffusion of excitons and free carriers in ReS₂ with transient absorption microscopy (TAM) for thicknesses from the monolayer to the bulk.
In TAM, a 10 fs pump pulse focused to the diffraction limit generates excitons and charge carriers, a 10 fs probe pulse in the wide field then monitors their temporal and spatial dynamics. A comparison of the spatial profile of the transient absorption signal following photoexcitation allows studying the movement of excited species with a sub-diffraction-limit resolution of 10 nm.
Varying the pump and probe linear polarisations we analyse the strongly anisotropic diffusion of the two lowest exciton populations. We observe an increase in the exciton lifetime and diffusion coefficients with the number of layers, which we attribute to a decrease in surface traps as the sample thickness increases.

Monolayer transition metal dichalcogenides have various intriguing optical properties mainly arising from room-temperature-stable excitons [1]. Recently, the exciton dynamics in monolayer vertical/lateral heterostructures or alloy monolayers have attracted much attention, because novel excitonic phenomena including the interlayer exciton [2], exciton dissociation [3], and exciton energy change [4] emerge owing to modulating electronic energy structures. These observations indicate that a combination of such structures potentially provides a platform for exploiting further novel exciton physics and applications.
Here, we report the exciton transport dynamics in a monolayer WS₂–WSe₂ lateral heterostructure with wide alloy region between the pure WS₂ and WSe₂ regions using photoluminescence (PL) spectroscopy. The lateral heterostructure was synthesized using a chemical vapor deposition method. It has the alloy region with the width of about 20 µm, where the composition ratio of WS₂ and WSe₂ gradually changes as a function of spatial coordinate. For the PL spectroscopy, we used the 2.33 eV or 2.22 eV continuous-wave (cw) lasers to excite the sample.
First, we conducted spectrally-resolved PL mapping on the lateral heterostructures across the wide alloy region in a confocal microscope at room temperature. The PL peak from the pure WS₂ region was observed at 1.94 eV and assigned to the A exciton. By scanning excitation laser spot across the alloy region of WS_2(1-x)Se_2x from the WS₂ side, the PL peak showed a gradual redshift, and finally reached to 1.70 eV that corresponds to the A exciton resonance energy in a pure monolayer WSe₂. The continuous shift of the exciton energies indicates formation of a built-in excitonic potential gradient across the wide alloy region, in contrast to the sharp potential change caused by lattice mismatch [5] and/or piezo effect [6] previously observed in various lateral heterostructures with sharp one dimensional interfaces.
Next, we examined the exciton transport properties in the wide alloy region at room temperature. We performed spatially- and spectrally-resolved PL measurements; note that these measurements are different from the confocal PL mapping described above. In this measurement, the excitation cw laser was tightly focused on the sample (spot size of c.a. 1.3 μm at FWHM), and the PL spectra at many points along the potential gradient direction (not only the directly excited spot) were simultaneously acquired. We found that the position showing the highest PL intensity deviated from the center of the laser excitation spot by c.a. 0.6 μm at maximum in the wide alloy region, and the direction of the deviation (exciton drift) was always along the excitonic potential gradient. This was contrast to the observations at the pure WS₂ and WSe₂ regions, where the highest PL intensity was always observed at the same position as the center of the laser excitation spot. We analyzed the spatial profile of the PL intensities using a drift-diffusion model taking the excitonic potential gradient into account, and confirmed that the potential gradient mainly causes the observed exciton drift. In the presentation, we will discuss the details of the exciton transport properties in the wide alloy region.

References:
[1]X. Xu et al., Nat. Phys., 10, 343, (2014).
[2]S. Mouri et al., Nanoscale, 9, 6674, (2017).
[3]D. Kozawa et al., Nano Lett., 16, 4087, (2016).
[4]Y. Chen et al., ACS Nano, 7, 4610, (2013).
[5]C. Zhang et al., Nat. Nanotech., 13, 152, (2018).
[6]Y. Kobayashi et al., Sci. Rep., 6, 31223, (2016).

Two-dimensional (2D) materials with intrinsic ferromagnetism provide unique opportunity to engineer new functionalities in nano-spintronics. One such material is CrI₃, a 2D Ising ferromagnet in monolayer with the Curie temperature (T_c) of 45 K [1]. Based on critical properties and scaling analysis, CrI₃ shows three-dimensional (3D) long-range magnetic coupling [2]. In systematic reduction of crystal thickness down to 50 nm bulk T_c of 61 K is gradually suppressed to 57 K, however, the satellite transition at T* = 45 K is observed. The T* is layer-independent and corresponds to T_c observed in the monolayer. The critical analysis around T_c reveals a crossover from 3D to 2D Ising ferromagnetism with mean field type interactions for microscale-thick crystals. This work shows that magnetic transition and critical properties can be continuously tuned on a mesoscale between monolayer and bulk crystals.

Acknowledgements

This work has been supported by the Research supported by the U.S. Department of Energy, Office of Basic Energy Sciences as part of the Computation Material Science Program (Y.L. and C. P.) and by the U.S. DOE under Contract No. DE-SC0012704 (L. W. X. T., J. L. and Y.Z.).

References
1. B. Huang et al., Nature 546, 270 (2017).
2. Y. Liu et al., Phys. Rev. B 97, 014420 (2018).
3. Y. Liu et al., submitted (2019).

Inorganic lamellar compounds are characterized by strong intra-layer chemical bonding and only weak van der Waals interaction between adjacent layers. Single layers are composed of metal ions sandwiched between chalcogenide or halide atoms, to form three- or four-atom thick slabs. Each metal is positioned at the center of a symmetric or distorted octahedron, where adjacent octahedra share edges for the formation of a single layer. Although the bulk forms of lamellar materials have been studied extensively in the previous century, transition metal di-chalcogenides [e.g. (Mo,W)Se₂] have received much attention in the past decade due to their decent conductivity, electronic band gap, in-plane stiffness, and above all, exhibiting a honeycomb Brillouin zone with distinguished K and K' symmetry points of opposing helicities, enabling potential applications in spin-based devices.
The current document proposes the investigation of metal phosphor tri-chalcogenides with the general chemical formula MPX₃ (M=metal, X=chalcogenide), closely resembling the metal di-chalcogenides, but one-third of the metal ions are replaced by a phosphor pair (P-P). A top view of a single layer reveals a honeycomb arrangement of the metal ions with P-P at the center of a hexagon. MPX₃ offers a large range of chemical compositions tunable by the M and X elements, and consequently forming semiconductors with band gap energies covering the range from UV to the near infrared. Furthermore, the metal ions within a layer produce a ferromagnetic or anti-ferromagnetic arrangement, endowing those materials with unique magnetic and magneto-optical properties. The valley degree of freedom is correlated with the antiferromagnetic ordering. The electronic band structure is dominated by the [P₂X₆]^4- units, contributing s- and p-atomic orbitals, while the metal d-orbitals hybridize with X(p) orbitals. Most common MPX₃ compounds have monoclinic or orthorhombic crystallographic structures with a space group C₂ or P2₁/c symmetry, respectively.
The talk at the MRS meeting 2019 will include description of the optical and magneto-optical properties of a vast number of MPX₃ single slabs, Moiré patterns and multiple layers. The description will focus on the influence of the internal magnetism on the optical properties of the materials, e.g., degree of circular polarization in the magneto-reflectance or magneto-photoluminescence spectra. Generally, the MPX₃ family exhibit merits beyond graphene and beyond the well-known (Mo,W)X₂ compounds, owed to the tune-ability of the electronic and magnetic properties.

Diluted magnetic semiconductors such as Mn-doped GaAs are attractive materials for gate-controlled spintronic devices but the low Curie temperature of the ferromagnetic state is far from room temperature, limiting for practical applications. Here, we report the long-range ferromagnetic order occurs above room temperature in diluted V-doped mono layer WSe₂. Our unambiguous observations are based on micro magnetic domains and the well-defined structure of V-substituted W atoms, which are characterized by magnetic force microscopy and high-resolution transmission electron microscopy, respectively. Furthermore, the modulation of the magnetic domains by an electronic gate is demonstrated, indicating the capability of gate-controlled magnetic properties of the samples. Our findings open new opportunities for using two-dimensional transition metal dichalcogenides for spintronic devices.

Most known magnetic van der Waals (vdW) materials are insulating or semiconducting. It is crucial to find materials with magnetic order and high-mobility for high-speed spintronic device making. Anti-ferromagnet GdTe₃ satisfies both down to 2D limits. The mobility reaches ~28000cm²/Vs, which is among the highest in all the known vdWs. It also exhibits an incommensurate charge density wave (CDW) and becomes superconductor under pressure. Here, we report the recent Raman study of GdTe₃, we will talk about the temperature and polarization dependence of CDW.

Conventional epitaxial ferromagnet-semiconductor multilayers have proven ripe systems for both advanced spintronics research and real-world applications in information technology. The highly ordered states inherent to ferromagnets, alongside their strong collective responses to external electromagnetic fields, makes them ideal candidates to enable control of charge and spin in closely proximitized semiconductors. Monolayers of MoSe₂ are atomically thin direct band-gap semiconductors which combine a regime of coupled spin and valley physics with an optically bright exciton ground state. As such, they are highly promising in the development of valleytronics, an analogue of spintronics which utilizes the carrier valley degree of freedom as an information carrying pseudospin. Crucially, MoSe₂ is also associated with robust chiral optical selection rules, efficiently bridging the gap between spin-valley polarization and optical addressability. By bringing MoSe₂ monolayers into contact with ferromagnetic materials, it becomes possible to bestow an additional, valley, degree of freedom onto the magnetic proximity coupled system.

Here, a monolayer of MoSe₂ is transferred directly on top of a 10nm thick film of europium sulfide (EuS), a ferromagnetic insulator with a rock-salt crystal structure [1]. Reflectivity spectra from the sample display two clear absorption resonances ascribed to attractive and repulsive Fermi-polarons, associated with spin-based interactions between photogenerated excitons and the electron Fermi sea [2, 3]. The interfacial exchange field between the EuS and MoSe₂ strongly amplifies an external magnetic field, leading to robust enhancement of the repulsive polaron valley Zeeman splitting, which reaches an effective g-factor of 15. By tuning the external magnetic field, oscillator strength may be transferred between the two polaron resonances, culminating in a complete 100% degree of circular polarization of the attractive polaron when B = 5T. This indicates total spin polarization of the 2-dimensional electron gas, owing to robust paramagnetic susceptibility of the MoSe₂ [3, 4].

Going further, we incorporate the MoSe₂ / EuS nanostructure into a zero-dimensional optical microcavity and observe valley selective strong light-matter coupling when B exceeds 5T, corresponding to the formation of Fermi polaron-polaritons in one valley only, selectable by reversing the external magnetic field orientation. In effect, swapping sample illumination between left or right circularly polarized light switches on or off the strong light-matter coupling regime. Our findings unveil the complex interplay between nanoscale magnetism, Coulomb effects, many-body physics and the spin and valley degrees of freedom in atomically thin semiconductor-ferromagnet interfaces. In particular, we expose the enormous potential of 2D material platforms for future opto-valleytronic memories and information processing technology.

[1] Zhao, et al. Nat. Nanotechnol. 12, 757-762 (2017)
[2] Sidler, et al. Nat. Phys. 13, 255-261 (2017)
[3] Back, et al. Phys. Rev. Lett. 118, 237404 (2017)
[4] Roch, et al. Nat. Nanotechnol. 14, 432-436 (2019)

Raman spectroscopy, imaging, and mapping are powerful non-contact, non-destructive optical probes of fundamental physics in graphene and other related two-dimensional (2D) materials, including layered, quantum materials that are candidates for use in the next quantum revolution. An amazing amount of information can be quantified from the Raman spectra, including layer thickness, disorder, edge and grain boundaries, doping, strain, thermal conductivity, magnetic ordering, and unique excitations such as charge density waves. Most interestingly for quantum materials is that Raman efficiently probes the evolution of the electronic structure and the electron-phonon, spin-phonon, and magnon-phonon interactions as a function of temperature, laser energy, and polarization. Our unique magneto-Raman spectroscopic capabilities will be detailed, enabling diffraction-limited, spatially-resolved Raman measurements while simultaneously varying the temperature (4K to 400 K), laser wavelength (tunability from visible to near infrared), and magnetic field (up to 9 T) to study the photo-physics of nanomaterials. Additionally, coupling to a triple grating spectrometer provides access to low-frequency (down to 6 cm^-1, or 0.75 meV) phonon and magnon modes, which are sensitive to coupling. By utilizing electrical feedthroughs, studying the strain-dependent effects on magnetic materials utilizing MEMs devices is also a novel opportunity.

Current results on intriguing quantum materials will be presented to highlight our capabilities and research directions. One example leverages the Raman spectra from α-RuCl₃ to probe this Kitaev magnet and possible quantum spin liquid. Within a single layer, the honeycomb lattice exhibits a small distortion, reducing the symmetry from hexagonal to orthorhombic. The possible effects of this broken in-plane symmetry on the Kitaev interactions are not well understood. Here, we utilize polarization-dependent Raman spectroscopy to study this distortion, including polarizations both parallel and perpendicular to the c-axis. Coupling of the phonons to a continuum was also investigated.

Another example is using Raman spectroscopy to probe magnetic phenomena in the antiferromagnetic metal phosphorus trichalcogenide family, including FePS₃, MnPS₃, and NiPS₃. Results on zone-folded phonons in FePS₃, the emergence of magnons in FePS₃ and MnPSe₃ with anomalous symmetry behaviors, and a two-magnon mode in NiPS₃ will all be discussed.

Finally, the magnetic field- and temperature-dependence of an exciting ferromagnetic 2D material, CrI₃, will be presented.

Controlling magnetism by electrical means is a key challenge to better information technology. Electrical control of magnetism has been explored in a variety of materials including dilute magnetic semiconductors, ferromagnetic metal thin films and multiferroics. The recently emerged atomically thin magnetic materials provide unprecedented opportunities to study magnetism in the 2D limit and engineer devices through van der Waals heterostructures. In particular, CrI3 is a model Ising ferromagnet with intriguing layer-dependent magnetic order: whereas monolayer CrI3 is a ferromagnet, bilayer CrI3 is an antiferromagnet with two ferromagnetic monolayers coupled antiferromagnetically. In this talk, I will present our recent results on switching the interlayer magnetic order in CrI3 bilayers by either a pure electric field or electrostatic doping and discuss possible mechanisms for the observed effects in experiment.

The ferromagnetic state in van der Waals two-dimensional (2D) materials has been reported recently in the monolayer limit. Intrinsic CrI₃ and CrGeTe₃ semiconductors reveal ferromagnetism but the Tc is still low below 60K. In contrast, monolayer VSe₂ is ferromagnetic metal with Tc above room temperature but incapable of controlling its carrier density. Moreover, the long-range ferromagnetic order in doped diluted chalcogenide semiconductors has not been demonstrated at room temperature. The key research target is to realize the long-range order ferromagnetism, Tc over room temperature, and semiconductor with gate tunability. Here, we unambiguously observe a ferromagnetic hysteresis loop together with magnetic domains above room temperature in diluted V-doped WSe₂, while maintaining the semiconducting characteristic of WSe₂ with a high on/off current ratio of five orders of magnitude.

The understanding of strongly-correlated quantum matter has challenged physicists for decades. Such difficulties have stimulated new research paradigms, such as ultra-cold atom lattices for simulating quantum materials. In this talk I will present a new platform to investigate strongly correlated physics, based on graphene moiré superlattices. In particular, I will show that when two graphene sheets are twisted by an angle close to the theoretically predicted ‘magic angle’, the resulting flat band structure near the Dirac point gives rise to a strongly-correlated electronic system. These flat bands exhibit half-filling insulating phases at zero magnetic field, which we show to be a correlated insulator arising from electrons localized in the moiré superlattice. Moreover, upon doping, we find electrically tunable superconductivity in this system, with many characteristics similar to high-temperature cuprates superconductivity. These unique properties of magic-angle twisted bilayer graphene open up a new playground for exotic many-body quantum phases in a 2D platform made of pure carbon and without magnetic field. The easy accessibility of the flat bands, the electrical tunability, and the bandwidth tunability though twist angle may pave the way towards more exotic correlated systems, such as quantum spin liquids or correlated topological insulators.

I will review our group’s work on the molecular transport properties of angstrom-scale channels fabricated by der Waals assembly of 2D crystals. These channels can be viewed as if individual atomic planes were extracted from a bulk crystal leaving behind two edge dislocations with an empty space in between. Gas, water, ion and proton transport have been studied in capillaries down to one atom in height.

Large-area, high-quality, and continuous growth technologies for CVD graphene have been developed to meet the requirements for industrial applications. However, there has been a lack of proper methods to check the quality of graphene in a fast and non-destructive way for mass-production scale. Optical microscopy (OM) has been widely explored for imaging and characterizing CVD graphene directly on Cu. Dark field (DF) OM was employed to investigate CVD graphene grown on Cu foil using the Rayleigh light scattering from Cu steps beneath graphene, which can be adopted only for metals that generate steps with considerable height beneath graphene after growing. Moreover, the DF imaging by the weak scattered light is time-consuming and requires a sample of graphene on Cu foil to be placed over a period of time under illumination. On the other hand, confocal laser scanning microscopy (CLSM) is capable of scanning a large area in a few seconds in high-resolution, which is expected to enable the faster characterization of graphene on Cu regardless of its production scale. Thus, we demonstrate that the reflection mode CLSM can be utilized for the in-situ monitoring of the CVD-grown graphene on Cu foil as a possible quality assessment tool of mass-produced CVD graphene films. The CLSM generates the high-contrast optical images of CVD-grown graphene on Cu over a large area, in which the contrast between the graphene layer and Cu is greater for shorter incident laser wavelengths. The reflectance contrast calculated by Maxwell equation is in good agreement with the experiment as the interference of reflected light at the interface between air, graphene, and Cu provides a high image contrast between the areas with and without graphene. Furthermore, we found out that the quality of CVD graphene can be characterized by CLSM because the reflectance and the optical conductivity are proportional to the defect density.

Control of nitrogen bonding configuration in hexagonal graphene lattice is considered as the promising strategy in realizing desired electrical, chemical, and optical properties of graphene. Three bonding configurations are typically detectable, including quaternary nitrogen (or graphitic nitrogen), pyridinic nitrogen, and pyrrolic nitrogen. However, cationic nitrogen is another possible bonding, which has been rarely investigated and not well understood in graphene chemistry. Therefore, study of graphene with cationic nitrogen doping is very challenging for further development towards this area. In this work, cationic N-doped graphene (CNG) has been synthesized via a novel method called solution plasma (SP) at room temperature and atmospheric pressure. The mixture of 1-ethyl-3-methylimidazolium dicyanamide (EMIM DCA) and dimethylformamide (DMF) was employed as carbon and nitrogen sources in synthesis. The originality of this work consists of two points as follows: 1. Combination of solution plasma and ionic liquid (EMIM DCA) allows us to achieve one-step, rapid, substrate-free synthesis of p-type graphene with cationic nitrogen doping at room temperature and atmospheric pressure. 2. Cationic nitrogen-doped graphene exhibited p-type semiconducting behavior with a low sheet resistance of 16 Ω sq⁻¹ and a high carrier concentration of 10¹⁹ cm⁻³. These fascinating electrical properties can be attributed to two important effects: (i) presence of cationic nitrogen in graphene lattice and (ii) preservation of planar structure.
The combination of SP and ionic liquid shows a promising strategy in design and synthesis of NG with cationic nitrogen-doping, which cannot be achieved in other existing methods. The CNG will be enable potential use in many applications in various fields, such as sensor, electronic, and energy devices.

Variously shaped transitional metal dichalcogenides (TMDC) domains are of significant interest since the electronic properties of pristine TMDC are strongly dependent on its size, shape, and edge structures. As the reactivity of TMDC is governable by the electronic structure at its edge, a number of attempts have been made to grow differently shaped TMDC domains and to define their edge structures. Of the various TMDC synthesis techniques, it is well-known that the chemical vapor deposition (CVD) is the most reasonable and appropriate method to produce large-scale and low-defect TMDC domains. And, many CVD parameters such as growth temperature, annealing time, and the amount of precursors affect TMDC growth.
In this study, we investigated CVD method in order to grow tungsten diselenide (WSe₂) domains of various shapes. By regulating the amount of Se powder as precursor and growth temperature during the CVD process, we synthesized hexagonal, square, circular, and triangular WSe₂ domains, and characterized their morphologies and physical properties using scanning electron microscopy (SEM), Raman spectroscopy, photoluminescence (PL) analyses, atomic force microscopy (AFM) and scanning tunneling microscope (STM). Based on the atomic configurations of WSe₂ domain edges and size distributions of the WSe₂ domains, we proposed the edge termination mechanism of anisotropic WSe₂ domains, which is strongly dependent on the amount of Se precursor and growth temperature during the CVD process.

Tungsten ditelluride (WTe₂) is a layered, type-II Weyl semimetal transition metal dichalcogenide (TMD) typically observed in a distorted 1T (1T’) phase with an orthorhombic crystal structure comprised planes of distorted triangular lattices of tungsten atoms sandwiched by tellurium atoms. The distortion pushes tungsten atoms closer together in the x-axis than the y or z-axis, generating quasi-one-dimensional chains of these atoms, leading to strongly anisotropic electronic properties throughout the material. It has been shown to have extraordinary physical properties, such as a high magnetoresistance, anisotropic ultra-low thermal conductivity and metal-insulator transition and it also exhibits interesting quantum phenomena such as the quantum spin-Hall effect and pressure-driven superconductivity. These unique properties make WTe₂ an exciting candidate for emerging applications, including phase change memory electrodes, magnetic field sensors, biosensors, microelectromechanical systems, hard disk drives and quantum computing.
While bulk crystals of the material have been available for many years, to date, thin films of WTe₂ have only recently been synthesized using techniques such as molecular beam epitaxy and powder source chemical vapor deposition (CVD) that utilize Te powder and W-feedstock. Metalorganic CVD (MOCVD) growth is also of interest as it enables precise delivery of precursors to the substrate at moderate growth pressures (100-700 Torr) but has not yet been studied in detail.
In this study, we investigate the use of MOCVD for the growth of WTe₂ on c-plane sapphire substrates. The studies were carried out in a vertical cold wall MOCVD reactor using tungsten hexacarbonyl (W(CO)₆) and diethyltelluride (DETe) as precursors for W and Te, respectively, in a H₂ carrier gas. Initial studies, carried out at 100 Torr reactor pressure using a W(CO)₆ flow rate of 1.3 x10^-4 sccm and Te/W ratio of ~8000 : 1, demonstrate the growth of WTe₂ as confirmed by Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). The growth rate was found to decrease with increasing temperature over the temperature range from 350-600^oC likely due to increased desorption of Te species from the substrate surface. Peaks at ~1340 cm^-1 and ~1600 cm^-1 were present in the Raman spectra of the films indicating the presence of carbon in the layers most likely due to the DETe source. Additional studies are underway to elucidate the growth mechanism and properties of MOCVD grown WTe₂. The synthesis of W(Se,Te)₂ alloy films via introduction of H₂Se during deposition will also be explored to explicate the consequences on phase control, oxidation stability, and the reduction of carbon contamination during growth.

Sensitive and flexible chemical sensors fabricated by a low-cost process are promising since they can be attached on curved, complex structures, and even the human body while providing an environmental signal to the user. For this reason, many nanomaterial-based flexible sensors have been investigated. However, they required a complicated manufacturing process leading to expensive cost and labor intensity. Furthermore, their limited chemical sensitivity and mechanical flexibility pose challenges.
Here, a highly deformable chemical sensor is reported with improved sensitivity that uses the hybrid structure of multiwalled carbon nanotubes (CNTs) and 2D transition metal dichalcogenides (TMDCs) on cellulose paper. Liquid dispersions of CNTs and TMDCs (e.g., WS₂ and MoS₂) are absorbed and dried on porous cellulose paper for sensor fabrication, which is simple, scalable, rapid, and cost-effective. Owing to the flexibility of cellulose paper, the sensor enables reversible three-dimensional folding and unfolding, bending, and twisting without any degradation. At the same time, the CNTs form a percolation network and simultaneously provide gas reactivity. Functionalization of CNTs with WS₂ greatly improves the sensing response to exposure to NO₂ molecules by more than 1.5 times. The measured sensitivity toward NO₂ is 4.57% ppm⁻¹, which is much higher than that of previous paper-based NO₂ sensors. In addition, our sensor maintains high sensitivity even under severe deformation such as heavy folding and crumpling. This approach can be a suitable way for simple fabrication of highly flexible and sensitive chemical sensor which can be potentially applicable for low-cost portable device and disposable sensor.

Tailoring the properties of two-dimensional (2D) crystals is important for both understanding the material behavior and exploring new functionality. Here we demonstrate the alteration of MoS₂ and metal-MoS₂ interfaces using a convergent Ar ion beam. Different beam energies, from 60 eV to 600 eV, are shown to have distinct effects on the optical and electrical properties of MoS₂. Defects and deformations created across different layers were investigated, revealing an unanticipated improvement in the Raman peak intensity of multilayer MoS₂ when exposed to a 60 eV Ar⁺ ion beam, and attenuation of the MoS₂ Raman peaks with a 200 eV ion beam. Scanning tunneling microscopy (STM) was used to probe the altered 4-layer MoS₂ surface, showing that the MoS₂ crystallinity still remains. This largely intact crystallinity of the 4-layer MoS₂ surface is in contrast to the significantly diminished PL peak of monolayer MoS₂, suggesting that the defects spread throughout multiple layers rather than concentrating at the surface. Using cross-sectional scanning transmission electron microscopy (STEM), alteration of the crystal structure after a 600 eV ion beam bombardment was observed for the first time, including generated defects across layers and voids in the crystal. We show that the 60 eV ion beam yields improvement in the metal-MoS₂ interface by decreasing the contact resistance from 17.5 kΩ*μm to 6 kΩ*μm at a carrier concentration of n_2D = 5.4×10¹² cm^-2. These results advance the use of low-energy ion beams to modify 2D materials and interfaces for tuning and improving performance for application as sensors, transistors, optoelectronics, and so forth.

Transition metal dichalcogenides (TMDCs) have been widely studied for a variety of applications including catalysis for hydrogen evolution reaction, flexible optoelectronics, thermoelectrics, and battery and supercapacitor applications. Recently, tellurium-based TMDCs, such as WTe₂ and MoTe₂, are heavily studied for their desirable electrical properties, which include superconductivity down to the monolayer limit,^1-2 quantum spin Hall state in monolayer,³ type II Weyl semimetallic nature,⁴ and large magnetoresistance.⁵ In recently published work⁶ we report on the synthesis of single-crystalline nanowires of TMDCs (WTe₂, WS₂, WSe₂, and MoTe₂) through conversion of oxide nanowires. The synthesized WTe₂ nanowires may be well suited to study topological superconductivity and Majorana zero modes due to the approximation as a one dimensional system.

We perform current-dependent resistance vs. temperature measurements to characterize the transport properties of our WTe₂ nanowires. We find that the nanowires exhibit a resistivity that is an order-of-magnitude higher than flakes of similar thicknesses, even after removing the contributions from a tunnel barrier at the contacts. We hypothesize that this increased resistivity results from increased surface scattering of the nanowires owing to surface oxidation and increased contribution from side surfaces rather than basal planes. We investigate in situ passivation layer to protect the nanowires from surface oxidation. Future works to develop mitigation strategies to handle scattering at the nanowire surface will allow for investigations into the topological features of WTe₂.


1: Inducing Strong Superconductivity in WTe₂ by a Proximity Effect. ACS Nano 2018, 12, 7185-7196.
2: Electrically Tunable Low-Density Superconductivity in a Monolayer Topological Insulator. Science 2018, 362, 926-929.
3: Observation of the Quantum Spin Hall Effect up to 100 Kelvin in a Monolayer Crystal. Science 2018, 359, 76-79.
4: Type-II Weyl Semimetals. Nature 2015, 527, 495-498.
5: Large, Non-Saturating Magnetoresistance in WTe₂. Nature 2014, 514, 205-208.
6: Synthesis of WTe₂ Nanowires with Increased Electron Scattering. ACS Nano 2019, DOI: 10.1021/acsnano.8b09342

The quest for transistors that operate in the THz regime is driven by the requirement for higher bandwidth to increase information bandwidth of networks. The replacement of the semiconductor transport channel with free-space is a potential path for achieving this objective. The so-called “Nano Vacuum Channel Transistors” (NVCTs) could potentially have superior performance compared to solid state devices of equivalent channel length owing to ballistic transport of electrons, shorter transit time and higher intrinsic breakdown voltage.¹ This is achievable as electrons are not scattered in free space (higher velocity) and there is no Avalanche carrier multiplication (no breakdown). Hence NVCTs have promise for very high Johnson Figure of Merit (~10¹⁴).²

NVCTs operate by injecting electrons into the channel when electrons tunnel through the barrier as the barrier is “thinned” by the application of a gate voltage with respect to the emitter tip. This barrier height is of the order of the electron affinity of Si (χ = 4.05 eV). By scaling down all the critical device dimensions by a factor of 10, very high density (10⁸ tips/cm²), self-aligned gate field emitter arrays with low turn-on voltage (8.5 V), low operating voltage (20 V), high current density (150 A/cm²) and long lifetime (>300 hours) have been demonstrated.³ However, these devices need ultra-high vacuum (UHV) for reliable operation as the field emission process is sensitive to barrier height variations induced by adsorption/desorption of gas molecules. Small changes in the barrier height lead to exponential variations in current.³ Poor vacuum also leads to generation of energetic ions that bombard the emitter tips, rendering the tips blunt and degrading electrical performance.

To overcome the need for UHV, we use atomically thin layers such as 2D materials to nano-encapsulate only the field emitter either in UHV or in a gas (e.g. helium) with high ionization energy. We exploit the higher transmission of smaller particles such as electrons through 2D materials compared to larger atoms/molecules such as gases.^4,5 By separating the electron emission region from the acceleration region (where the electrons acquire energy), electrons can be transported in a non-ideal vacuum, if not atmospheric conditions. Furthermore, the nano-encapsulation layer can either function as an ion repeller if biased positively or can be used for secondary electron emission interface. For mechanical strength, multiple layers that are transparent to electrons while impervious to gas molecules/ions are used.⁶ In this work, we investigate the structural and electrical properties of 2D materials as nano-encapsulation layers for Si based field emission sources. The goal is to realize empty state electronics capable of functioning at higher power and harsher conditions (high radiation and high temperature) compared to solid state electronics.

References:
¹ J.-W. Han, D.-I. Moon, and M. Meyyappan, Nano Lett. 17, 2146 (2017).
² S.A. Guerrera and A.I. Akinwande, in 2016 29th Int. Vac. Nanoelectron. Conf. (IEEE, 2016), pp. 1–2.
³ S.A. Guerrera and A.I. Akinwande, Nanotechnology 27, 295302 (2016).
⁴ J.S. Bunch, S.S. Verbridge, J.S. Alden, A.M. Van Der Zande, J.M. Parpia, H.G. Craighead, and P.L. Mceuen, Nano Lett. 8, 2458-2462 (2008).
⁵ C. Li, M.T. Cole, W. Lei, K. Qu, K. Ying, Y. Zhang, A.R. Robertson, J.H. Warner, S. Ding, X. Zhang, B. Wang, and W.I. Milne, Adv. Funct. Mater. 24, 1218 (2014).
⁶ L. Wang, C.M. Williams, M.S.H. Boutilier, P.R. Kidambi, and R. Karnik, Nano Lett. 17, 3081 (2017).

One of the main advantages of 2D materials is that their properties are dependent on their local environment, making them inherently controllable by effects such as electrostatic gating. Furthermore, 2D materials can be transferred onto arbitrary substrates, an excellent opportunity for device engineering. In particular, stacking 2D materials on tuneable or phase-change materials (PCMs), such as those exhibiting a metal-insulator transition (MIT), provides an avenue to further control the properties of 2D materials. Among PCMs, vanadium dioxide (VO₂) has received a great deal of attention for its easily-accessible critical temperature (T_c) near 70°C, rich phase diagram, ultrafast optically-induced phase transition, and large change in optical properties (both n and k) especially pronounced in the near- and mid-IR; properties with promise for a variety of applications, including sensors, photonic modulators and switches, passive thermal control films—and the modulation of hyperbolic phonon polaritons (HPhP) in hexagonal boron nitride (hBN). Single crystals of VO₂—while less durable and scalable than thin films—have well-defined crystal lattices, sharp phase transitions, and, when strained, striped phases of coexisting insulating and metallic domains. These are ideal for situations where sharp, lateral boundaries are required within devices. Recently, these domain walls were used to demonstrate index-based launching, reflection, and refraction of HPhPs in hBN.
Developing such applications requires robust, reproducible techniques to create VO₂ in a variety of forms and to locally adjust its phase transition. High-quality VO₂ crystals can be grown via a physical vapor transport method, where V₂O₅ precursor is vaporized, transported to a growth substrate, and reduced to VO₂. The size, shape, and structure of the resulting crystals depends heavily on growth parameters and the substrate used. Moreover, the T_c, existence of intermediate phases, and pattern of coexisting domains are sensitive to the crystal morphology and substrate interaction. While much work has focused on growing nanocrystals of various sizes and shapes, we seek to grow large, low-aspect ratio crystals to support large slabs of two-dimensional materials and to facilitate optical experiments on those heterostructures.
Specifically, we grow such VO₂ microcrystals on a variety of crystalline oxide substrates with good lattice match to VO₂. We discuss the optimization of growth parameters, and use optical microscopy, Raman spectroscopy, and XRD to analyze the effect of lattice match on the orientation and morphology of the crystals. Additionally, we use FTIR to measure the optical constants of single crystal VO₂ in the M1, M2, and R phases, and investigate the possibility of patterning metallic domains in VO₂ crystals with local laser heating. On c-cut sapphire—previously used to grow epitaxial nanowires—we observe large microcrystals stabilized in the insulating M2 phase at room temperature, which we attribute to strain arising from lattice mismatch and thermal expansion. On yttrium-stabilized zirconia (YSZ) (100)—previously used as a buffer layer to grow epitaxial VO₂ thin films—we observe preferred orientation of large microcrystals, as well as formation of yttrium vanadate and zirconia due to the instability of YSZ at high temperatures. These crystals exhibit unique surface morphologies, such as terrace-like steps. We also report on the growth of single crystals on other cuts of sapphire and YSZ, and on spinel (111) (MgAl₂O₄, known to support epitaxial VO₂ thin films). Finally, we describe the creation of 2D heterostructures using these crystals, and demonstrate their applicability for infrared spectroscopy experiments.

The exotic physical and chemical properties of layered transition metal dichalcogenides (TMDCs) make them prime candidates for use in a variety of applications including electronics, optoelectronics, valleytronics, and catalysis.^[1] The ability to rationally manipulate the electronic structure of atomically thin films of TMDCs is essential to enable their practical implementation. To achieve this goal, there are several examples of surface functionalization of TMDCs using small molecules to control the charge-carrier polarity and doping of TMDCs.^[2] However, our understanding about the nature of the interaction between small molecules and the TMDCs or how the structure of small molecules impacts the properties of the TMDCs, remains limited. Here, we describe studies which rationally tune the electronic structure of MoS₂ monolayers using a library of systematically designed small molecules with different structures and redox properties. Electrical and magneto-transport measurements show a factor of ten improvement in electron mobility and carrier density of functionalized MoS₂ devices with well-defined organic reductants. To obtain a holistic description of the doping effects, a range of characterization techniques including X-ray photoelectron, photoluminescence, and Raman spectroscopies, and scanning tunneling and transmission electron microscopy, as well as secondary ion mass spectrometry are utilized.


[1] Q. H. Wang et. al., Nat. Nanotechnol. 2012, 7, 699.
[2] S. Bertolazzi et. al., Chem, Soc, Rev. 2018, 47, 6845.

Red phosphorus, an allotrope of phosphorus which is usually known to be amorphous, has several types of crystalline phases. The atomic structures and electrical properties of crystalline red phosphorus, however, have not been studied in detail due to the existence of various crystalline phases of complex structure. The red phosphorus polymorphs are categorized into five types, which include the amorphous phase (type-I). The crystal structures of type-IV (fibrous red phosphorus) and type-V (Hittorf 's phosphorus) have been previously revealed but those of type-II and type-III phases are yet to be identified. Here we investigate the crystal structures and electrical properties of crystalline red phosphorus polymorphs. Using chemical vapor transport method, we synthesize crystalline red phosphorus. Using Raman spectroscopy, electron diffraction, and X-ray diffraction measurements, we confirmed that synthesized red phosphorus is mainly type-II with some mixture of type-IV crystalline phases. The electron crystallography can be utilized to identify the crystal structure of type-II red phosphorus. With its semiconducting nature, the crystalline red phosphorus can be a promising material for electronic and optoelectronic applications.

Atomically thin, two-dimensional (2D) indium selenide (InSe) has attracted considerable attention owing to the dependence of its bandgap on sample thickness, making it suitable for small-scale optoelectronic device applications. In this work, by the use of Raman spectroscopy with three different laser wavelengths, including 488 nm, 532 nm, and 633 nm, representing resonant, near-resonant and conventional non-resonant conditions, a conclusive understanding of the thickness dependence of lattice vibrations and electronic band structure of InSe and InSe/graphene heterostructures are presented. Combining our experimental measurements with first-principles quantum mechanical modeling of the InSe systems, we identified the crystal structure as ε-phase InSe and demonstrated that its measured intensity ratio of Raman peaks in the resonant Raman spectrum evolves with the number of layers. Moreover, graphene coating enhances Raman scattering of few-layered InSe and also makes its photoluminescence stable under higher-intensity laser illumination. The optically induced charge transfer between van der Waals graphene/InSe heterostructures is observed under excitation of the E' transition in InSe, where the observed mechanism may potentially be a route for future integrated electronic and optoelectronic devices.

Electronic devices such as personal computers, smartphones, and wearable devices with many highly sensitive precision electronic components can be malfunctioned due to electromagnetic interference (EMI). An EMI shielding is a solution to prevent electromagnetic (EM) radiation pollution but conventional metal-based EMI shielding materials cannot apply for stretchable and flexible devices. The 3D graphene foam/polymer composites have acquired noteworthy interests in flexible EMI shielding materials owing to the excellent electrical conductivity and large specific surface area. However, their shielding performance still needs further improvement to serve in real world applications.
Herein, the Fe₃O₄ nanoparticles intercalated 2D Ti₃C₂T_x (MXene) through ultrasonic method is decorated on the graphene foam that is grown on nickel skeleton by chemical vapor deposition (CVD) process, before embedding in the poly dimethyl siloxane (PDMS) matrix. After etching nickel away, the composite shows flexibility, elasticity, high conductivity, and a brilliant EM wave absorption. We propose the presence of the Fe₃O₄ nanoparticle intercalated MXene induces the internal multi-reflection, magnetic loss, and interfacial polarization in the graphene foam/polymer matrix resulting in the excellent EMI shielding effectiveness as high as 83.6 dB in the X-band frequency range (8.2-12.4 GHz). Consequently, the lightweight, flexible, and highly conductive graphene foam-based composite opens up the possibility to use as the high-performance EMI shielding materials applicable in the wearable electronics, flexible and stretchable electronics, etc.

Acknowledgment
This work was supported by the STEAM Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2017M3C1A9069590) and the Electronics and Telecommunications Research Institute (ETRI) grant (19ZB1100) funded by the Korean government.

The climate change and necessity of monitoring dynamic changes of humidity in various fields starting from food processing industry, semiconductor fabrication facility, automobile and agriculture, to more sophisticated human respiration, non-contact epidermal sensing has increased the demand for highly sensitive humidity sensor material research. Two-dimensional (2D) material like molybdenum disulfides (MoS₂) is an excellent material for humidity sensing due to the presence of inherent defects, and high moisture absorption even at room temperature. However, the humidity absorption of mono or multilayer MoS₂ are limited by its atomically thin layers and their relatively low defect sites for water molecules. Humidity sensor made from pure MoS₂ films suffer from low sensitivity at room temperature has been reported. Recently, research on atmospheric humidity effect on the electrical characteristics of 3D nanostructure, MoS₂ nanoflakes have been studied with the aim to increase moisture absorption and broaden detection range. However, previous studies are mostly focused on top or bottom gated complex devices that have limitations like hysteresis, stability at high humidity levels, durability.
In this study, we demonstrate a simple resistive type anodic aluminum oxide (AAO) assisted MoS₂ honeycomb structure for superior humidity sensing with fast response. The honeycomb structured MoS₂ tubes were fabricated by filtering the (NH₄)₂MoS₄ solution through anodic aluminum oxide (AAO) membrane by vacuum filtration method. To achieve perfectly open surfaced MoS₂ honeycomb for maximum water vapor absorption site, (NH₄)₂MoS₄ solution vacuum filtration was repeated in cycles to eliminate larger particles from the solution. Next, T-CVD annealing process was used to synthesis MoS₂ tube honeycomb. In our finding, this 3D honeycomb structure based on MoS₂ tubes can respond fast to humidity from atmosphere, human skin, and human breath at the range of RH 20% ~ RH 85%. The sensitivity and response speed are heavily influenced by the MoS₂ tubes, due to their wider surface area and open pores for higher moisture absorption sites. From experimental result, our MoS₂ honeycomb structured humidity sensor showed excellent sensitivity of ~700 (ΔI/I₀) at RH 83%, with a fast response and recovery time of 0.47s and 0.81s, respectively. This study paves the way to fast humidity sensing and non-contact sensation for next generation electronic devices.

Acknowledgement
This work was partly supported by Electronics and Telecommunications Research Institute (ETRI) grant (19ZB1100) funded by the Korean government and Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20181510102340, Development of a real-time detection system for unidentified RCS leakage less than 0.5 gpm).

Recently, the concept of dimensionality has emerged as an essential parameter for tuning material properties. Coinciding with the recent rise of two-dimensional (2D) materials, there have been extensive efforts to engineering additional levels of confinement, and thus lower dimensionality, in few- and monolayer materials. The greatest successes have been achieved in the fabrication and synthesis of graphene nanoribbons (GNRs), where rational bottom-up synthesis has been accomplished using the self-assembly of molecular precursors. This has enabled the growth of GNRs with specific edge structures and atomically precise widths, as well as single GNR heterojunctions with engineered band alignment and topology. Similarly, exciting physics are expected when the 2D transition metal dichalcogenides (TMDs) are further constrained to one-dimension (1D). However, the synthesis and/or fabrication of TMD nanoribbons (NRs) has lagged far behind that of GNRs both in terms of quality and width control. In the past, most studies have relied upon top-down fabrication methods that require lithography and etching processes. This results in NRs with widths greater than 50 nm (too large to observe quantum confinement effects) and a high degree of disorder. Here, we report the growth of ultra-narrow TMD nanoribbons as templated by multiwalled carbon nanotubes. The nanoribbons are grown within the hollow cavity of the carbon nanotubes, limiting their lateral dimensions and layer number, while simultaneously stabilizing them against the environment. The result is pristine nanoribbons that reach the monolayer limit and exhibit widths that can reach below 2 nm, while maintaining a defined edge structure. High-resolution transmission electron microscopy and scanning transmission electron microscopy reveal the detailed atomic structure of the material, and density functional theory is used to explore the effects of additional levels of confinement on electronic properties. The NR-in-nanotube structure represents a unique route toward nanoribbon growth and the exploration of materials under extreme dimensional constraint.

Two-dimensional (2D) semiconductor materials, such as transition metal dichalcogenides, monochalcogenides, and phosphorene have attracted extensive research interests for potential applications in optoelectronics, spintronics, photovoltaics, and catalysis. Understanding the effect of impurities, defects, and dopants on the electronic properties is crucial for the selection of materials. Choosing suitable synthesis and processing conditions allows for some control over their concentrations, and hence to tailor the carrier concentration, character, and mobility in 2D materials. Accurate determination of defect formation energies and charge transition levels enables us to predict their effect on the electronic properties and how they respond to changes in synthesis and processing.

Density functional theory (DFT) calculations of point defects in solids is a mature field with a proven record of experimentally validated predictions. However, modeling charged defects in single-layer materials with common plane-wave DFT approaches poses additional challenges which lead to the divergence of the energy with vacuum spacing. Recently, Freysoldt and Neugebauer developed a correction scheme which employs a surrogate model to restore the appropriate electrostatic boundary conditions for charged 2D materials. We apply this correction scheme to compute the formation energies and charge transition levels associated with point defects and defect complexes in technologically relevant materials such as MoS₂, WSe₂, and phosphorene. We find that dopants can bind with intrinsic defects such as vacancies to form defect complexes under certain experimental conditions, acting as compensating defects and modifying the electronic properties of the individual defects. An example of this is the Re dopant – S vacancy complex in MoS₂ which we found to be thermodynamically favorable and which if formed may in fact passivate the n-type Re dopant. By further analyzing the defect electronic structures, we also predict interesting phenomena such as Jahn-Teller distortions which occur when the defects are in specific charge states. Finally, we are developing a Python-based workflow to facilitate efficient high-throughput calculations of charged point defects in 2D materials and interfacing it with existing materials databases to create a new database of defects in 2D and layered bulk materials.

We report results on a new class of bottom-up synthesized functionalized atomically precise graphene nanoribbons (GNRs). These new ribbons are structurally related to the established chevron GNRs but are expected to have a smaller bandgap and thus higher electrical conductivity, and the edge functionalization is expected to improve their processability for device fabrication. The GNRs were synthesized in solution via Suzuki coupling of structurally different molecular precursors and characterized by a variety of spectroscopic techniques. Dry contact transfer (DCT) was used to deposit these GNRs in UHV onto hydrogen passivated Si(100) surfaces for STM imaging and spectroscopy (STS) to elucidate the structural and electronic features of these GNRs. STS results indicate a strong variation in electronic structure along the length of the GNR. Specifically, the bandgap of the functionalized ends of the reduced chevron GNR exceeds 3.3 eV while the center of the GNR exhibits a bandgap of 2.5 eV. First principle simulations using DFT and GW of the electronic structure will also be presented.

Transition metal dichalcogenides (TMDs) have received considerable attention over the past decade, owing to its unique thickness-dependent electrical and optical properties.^[1] Chemical vapor deposition (CVD) is required to realize the scalable fabrication of TMD devices. However, the optical emission and carrier mobility of CVD-grown TMDs remain low due to the broad emission from defect bound excitons. A wide variety of approaches has been made to alter optical and electrical properties of TMD, including the incorporation of transition metals atoms (Mn, Nb, and Re)^[2–4] in monolayer TMDs using substitutional doping or intercalation.^[5,6] Among those transition metal atoms, iron (Fe) has attracted most attention due to its predicated magnetic properties^[7] and non-degenerate ground state with non-zero spin.^[8]

Here, we systematically investigated the temperature-dependent photoluminescence (PL) and magneto-PL properties of in situ Fe doping of polycrystalline MoS₂ monolayers via CVD.^[9] First, scanning transmission electron microscopy was employed to probe the atomic structure of the Fe-doped MoS₂ monolayers. Fe atoms with lower relative intensity were visible in the MoS₂ lattice at Mo sites, suggesting a substitutional Fe-doping of MoS₂ monolayers. The temperature-dependent PL studies of both bare and Fe-doped MoS₂ monolayers revealed significant suppression of the broad defect-bound emission related to chalcogen vacancies. In addition, we observed the emergence of new emission peaks after Fe atoms are introduced. Since the new emission peaks were absent in bare MoS₂ monolayer, we attribute the new emission peaks o the optically active Fe ion sites. The light emissions from these optically active Fe ion sites had very strong thermal stability from 4K to 300K. Magneto-PL studies provided insights into the magnetic behavior of the Fe related emissions. We found ~ 30% circular dichroism in the Fe-related emissions even at room temperature without applying a magnetic field. Furthermore, a ferromagnetic behavior of the Fe related emissions was observed by measuring the PL circular dichroism at different magnetic fields (-6.5 T to 6.5 T) with increasing and decreasing sweeping directions. These findings are pathfinders toward enabling potential applications on minimizing bit storage for future electronics based on TMDs.

[1] Q. H. Wang et al., Nat. Nanotechnol. 2012, 7, 699.
[2] K. Zhang et al., Nano Lett. 2015, 15, 6586.
[3] J. Gao et al., Adv. Mater. 2016, 28, 9735.
[4] K. Zhang et al., Adv. Funct. Mater. 2018, 28, 1.
[5] Q. Arnoux et al., Phys. Rev. B 2019, 99, 144405.
[6] D. Sarkar et al., Nano Lett. 2015, 15, 2852.
[7] L. Y. Antipina et al., Phys. Chem. Chem. Phys. 2016, 18, 26956.
[8] T. Smolenski et al., Nat. Commun. 2016, 7, 10484.
[9] X. Wang et al., 2D Mater. 2017, 4, 025093.

Two-dimensional porous membrane is considered as promising candidate for photocatalyst or electrocatalyst, desalination and rapid DNA sequencing. Delicate preparation, low-efficiency and absence of scalability of the current high energy electron irradiation method for nanopores creation, has pushed the community to develop easy, low-cost and scalable methods to create nanopores onto two-dimensional materials. Here, we developed an oxidization and etching method to create easily multiple nanopores onto monolayer molybdenum disulfide (MoS₂). These nanopores are triangles with straight edges and their size is tunable by changing the oxidization temperature. With much more exposed fresh edge sites which are active sites for hydrogen evolution reaction, basal plane activity of monolayer porous MoS₂ is significantly enhanced compared with pristine MoS₂. Our findings imply great applications potential exist in two-dimensional porous membrane and this simple, effective and scalable method to create multiple nanopores on monolayer MoS₂ paves the way to the applications of two-dimensional porous membrane.

Vertically aligned graphene nanosheet arrays (VAGNAs) are grown on Ge substrate using a PECVD process. VAGNAs have unique tip morphology and large surface area, which make them as promising candidates for different applications, such as field emitting based devices, supercapacitors and batteries. Raman study implies that our PECVD-grown VAGNAs has the features of single-layer graphene termination at the top, which is further identified by our High-Resolution Transmission Electron Microscopy (HRTEM) characterization. In addition, to extract the quality of the grown layers, we also carry out the TEM based electron energy loss spectroscopy (EELS) investigation. EELS provides intricate information such as, the hybridization states and defect types present in the grown graphene sheets. Following the experimental EELS study, we also perform first-principles calculations to evaluate the effect of different types of defects on the electronic structures of VAGNAs. Moreover, the quality of the PECVD-grown graphene sheets on Germanium substrate is further compared with the graphene grown on metal substrate fabricated using the standard CVD process.

MXenes, such as Ti₃C₂T_x and V₂CT_x, are fascinating 2D nanomaterials with an attractive combination of functional properties suitable for applications such as batteries, supercapacitors, and sensors. However, fabrication of devices and functional coatings based on MXenes remains challenging as they are prone to oxidize and degrade quickly in a matter of days from reacting with water and dissolved oxygen. We examine the oxidation of Ti₃C₂T_x MXene nanosheets in various media (air, water, organic solvents, and polymer composite) via multiple types of measurements to assess their shelf stability. The degree of MXene oxidation is found directly indicated by the property and structure changes such as electrical conductivity and the TiO₂ content which is detected by X-ray Photoelectron Spectroscopy. The oxidation rate of MXene nanosheets is observed fastest in liquid media and slowest in solid media and can be accelerated by exposure to UV light. More importantly, we demonstrate an effective method to arrest the oxidation of colloidal Ti₃C₂T_x MXene nanosheets by introducing the antioxidants. The success of the method is evident as the MXene nanosheets maintain their composition, morphology, electrical conductivity, and colloidal stability. Even in the presence of water and oxygen, the electrical conductivity of Ti₃C₂T_x nanosheets treated with antioxidants is orders of magnitude higher as compared to untreated ones after 21 days. The conductivity changes also reveal that the resistance to oxidation persists in the dehydrated MXenes as well. In addition, our study includes a ReaxFF molecular simulation performed to elucidate the mechanism for MXene-antioxidant interactions. Our findings have the potential to be generalized to protect more types of MXenes as well and solve the most pressing challenge in the field of MXene engineering.

Two-dimensional (2D) perovskites or Ruddleson Popper (RP) perovskites have emerged as a class of material inheriting the superior optoelectronic properties of two different materials: perovskites and 2D materials. The large exciton binding energy and natural quantum well structure of 2D perovskite not only make these materials ideal platforms to study light-matter interactions, but also render them suitable for fabrication of various functional optoelectronic devices. Nanoscale structuring and morphology control have led to semiconductors with enhanced functionalities. For example, nanowires of semiconducting materials have been extensively used for important applications like lasing and sensing. Catalyst-assisted Vapor Liquid Solid (VLS) techniques, and template assisted growth, have conventionally been used for nanowire growth. However, catalyst and template-free scalable growth with morphology control of 2D perovskites have remained elusive. In this manuscript, we demonstrate a facile approach for morphology-controlled growth of high-quality nanowires of 2D perovskite, (BA)₂PbI₄. We demonstrate that the photoluminescence (PL) from the nanowires are highly polarized with a polarization ratio as large as ~ 0.73, which is one of the largest reported for perovskites. We further show that the photocurrent from the device based on the nanowire/graphene heterostructure is also sensitive to the polarization of the incident light with the photocurrent anisotropy ratio of ~3.62 (much larger than the previously reported best value of 2.68 for perovskite nanowires), thus demonstrating the potential of these nanowires as highly efficient photodetectors of polarized light.

Heterostructures among two-dimensional (2D) layered materials have attracted great interest due to their unique properties, and potential applications. And wrapping the 2D layered materials into one-dimensional nanotubes brings out opportunities in creating chiral tubular structures and radial heterojunctions, with more interesting functionalities and phenomena. Early efforts have focused on using a tubular template to confine the formation of another nanotube inside by capillary wetting method, such as PbI₂@WS₂ and PbI₂@CNT.^[1,2] But in those cases, the nanotube template ends must be open and a long-term thermal annealing process (up to 1 month) is usually carried out.^[1] For practical applications, it is more desired to have the target nanotube sitting outside the template and accessible to foreign environment.
As most of inorganic crystalline materials have rigid atomic planes, wrapping a conformal inorganic layer around the outer surface of template (where spatial confinement is absent) turns to be rather difficult. Here, we report controlled fabrication of heterogeneous inorganic nanotubes consisting of a CNT core and crystalline MoS₂ multi-shells with long-range continuity. Starting from a three-dimensional porous CNT sponge as template, we adopt a thiourea-assisted solvothermal method to deposit precursors and form the CNT@amorphous MoS₂ structure at first. Then by high-temperature annealing process, the amorphous sheath is converted into highly crystalline multi-walled MoS₂ nanotubes wrapping around individual CNTs throughout the sponge. The resulting core-shell CNT@MoS₂ nanotube have a controllable layer number of outer MoS₂ nanotube from 2 to 40 layers, by adjusting the precursor concentration during thiourea-thermal method. And such MoS₂ nanotube shows increasing surface strain as the nanotube diameter decreasing. In this way, when used as oxygen reduction reaction (ORR) electrocatalyst, the heterogeneous inorganic nanotubes with thinner-walled (2~6 layers) MoS₂ nanotube exhibits superior kinetics and lower overpotential. Besides, the stable hetero-nanotube structure endows CNT@MoS₂ sponge with a specific capacity of 740 mAhg^-1 at 100 mAg^-1 and high cycling stability when used as a freestanding electrode for Li-ion batteries.

[1] R. Kreizman, S. Y. Hong, J. Sloan, R. Popovitz-Biro, A. Albu-Yaron, G. Tobias, B. Ballesteros, B. G. Davis, M. L. Green, R. Tenne, Angew. Chem. Int. Ed. Engl. 2009, 48, 1230.
[2] L. Cabana, B. Ballesteros, E. Batista, C. Magen, R. Arenal, J. Oro-Sole, R. Rurali, G. Tobias, Adv. Mater. 2014, 26, 2016.

Two-dimensional semiconducting transition metal dichalcogenides (TMDCs) provides rich platform to study several interesting many body complexes such as excitons, trions, biexcitons and excitonic states for fundamental understanding and optoelectronic applications.^{1, 2} Remarkably, strong Coulombic interaction with reduced dielectric screening effect in TMDCs makes experimental realization of such complexes even at room temperature which are otherwise observed at low temperatures. Further, similar to Hydrogen atoms, excitons exhibit excited states known as Rydberg states which are under scientific exploration and opens up new avenues for future quantum information processing, optoelectronic and photonics.³ Owing to weak signal from biexciton and excitonic Rydberg states in TMDCs, such complexes are less understood than excitons and trion. Further, the extent of spin orbit splitting of valence band due to d-orbital of transition metal plays an important role in observation of excitonic states. Large spin orbit splitting (> 400 meV) resulted in observation of series of Rydberg states of A exciton in ML of WS₂ and WSe₂ using several linear and non-linear spectroscopic techniques. On the contrary, the small spin orbit splitting ~ 150 meV in ML MoS₂ provides an additional challenge in distinct realization of 2s state of A exciton (A_2s) because of the overlap with large spectral width of B exciton. Recently, A_2s state in ML MoS₂ has been theoretically predicted and experimentally observed in ML MoS₂ encapsulated by hBN ~ 2.1 eV using advance non-linear spectroscopic measurements.⁵ In our study, we performed temperature dependent photoluminescence (PL) studies up to 4 K to minimize spectral width of A and B excitons and accessed excitonic complexes as well Rydberg states. We discuss about the systematic observation of biexciton (~ 1.90 eV), sulfur vacancy mediated bound exciton (1.85 eV) and A_2s state (~ 2.13 eV) distinct from A- trion (~ 1.92 eV), A exciton (~ 1.96 eV) and B exciton (~ 2.07 eV) in ML MoS₂ using laser power dependent and temperature dependent PL spectroscopy from 4 K to 300 K. At low temperatures, the observed weak signal of excited state A_2s ultimately merges with spectral broadening of B exciton with raising temperatures.⁶ From the estimated binding energy of biexciton (~ 60 meV), we suggest that biexcitons can be stable even at room temperature.

References

1. Lee, H. S.; Kim, M. S.; Kim, H.; Lee, Y. H. Physical Review B 2016, 93, (14), 140409.
2. Mak, K. F.; He, K.; Shan, J.; Heinz, T. F. Nature Nanotechnology 2012, 7, 494.
3. Chernikov, A.; Berkelbach, T. C.; Hill, H. M.; Rigosi, A.; Li, Y.; Aslan, O. B.; Reichman, D. R.; Hybertsen, M. S.; Heinz, T. F. Physical Review Letters 2014, 113, (7), 076802.
4. Liu, E.; van Baren, J.; Taniguchi, T.; Watanabe, K.; Chang, Y.-C.; Lui, C. H. Physical Review B 2019, 99, (20), 205420.
5. Robert, C.; Semina, M. A.; Cadiz, F.; Manca, M.; Courtade, E.; Taniguchi, T.; Watanabe, K.; Cai, H.; Tongay, S.; Lassagne, B.; Renucci, P.; Amand, T.; Marie, X.; Glazov, M. M.; Urbaszek, B. Physical Review Materials 2018, 2, (1), 011001.
6. Pandey, J.; Soni, A. Applied Surface Science 2019, 463, 52-57.

Two-dimensional (2D) transition-metal dichalcogenides (TMDs) with MX₂ formula (M: transition metal, X: chalcogen) offer a plethora of opportunities for device engineering as well as fundamental studies. TMDs can exist in various crystalline phases including semiconducting and metallic. More importantly, the electronic bandgap of the semiconducting phase can be controlled via changing M or X elements, offering a digital portfolio of 2D materials for optoelectronic applications over a wide spectral range from visible to near infrared. In addition, this digital portfolio can be turned into an analog one through the alloying of 2D TMDs, [1, 2] a unique feature that enables the design of 2D materials with customized properties. Such a diversity of material properties renders 2D TMDs a family of unmatched qualities for the formation of heterostructures (HSs) via vertical stacking or lateral stitching of dissimilar TMDs. Unlike the diverse range of developed methods for the synthesis of vertical TMD HSs, the formation of lateral TMD HSs has been mostly limited to the edge-epitaxial growth in which a dissimilar TMD crystal (e.g., MX₂) is grown on the unsaturated edge of a pre-grown TMD crystal (e.g., MX’₂) to form a lateral HS (i.e., MX₂-MX’₂). [3] Despite the high-quality junctions, the edge-epitaxial method fails to provide a systematic control over the shape, lateral dimensions, and even the location of HSs, a set of features that are mandatory for the real-world applications.

Here we introduce an alternative method for the synthesis of lateral TMD HSs with highly controllable dimensions and shapes in predefined locations. Our method relies on combining the standard lithography with the post-growth alloying of TMDs [1] for converting a uniform MX₂ monolayer into a MX₂-MX’₂ lateral heterostructure. We will show that the involvement of the lithography facilitates the control over the geometrical aspects of the HSs, and the post-growth alloying enables a full control over the electronic band alignment at the interface of the MX₂-MX’₂ HSs. Furthermore, using various spectroscopic characterizations and electrical measurements, we will demonstrate potentials of lateral TMD HSs for opto-electronic applications.

[1] H. Taghinejad et. al “Defect-Mediated Alloying of Monolayer Transition-Metal Dichalcogenides,” ACS Nano (2018).
[2] H. Taghinejad et. al “Strain Relaxation via Formation of Cracks in Compositionally Modulated Two-Dimensional Semiconductor Alloys,” npj 2D Mat. and App. (2018).
[3] Y. Gong, et. al “Vertical and in-plane heterostructures from WS₂/MoS₂ monolayers,” Nat. Mater. (2014).

Graphene, a two-dimensional (2D) allotrope of carbon, has exceptional surface-area-to-volume ratio (up to 2630 m² g^-1) and highly catalytic edges. To leverage these properties, efforts have been made to arrange graphene flakes in three-dimensional (3D) geometries with an eye towards integration into functional electronic devices. Electron transport mechanisms in 2D single-crystal graphene films have been extensively studied with relation to the material’s structure (e.g. edge termination, defect type and density, crystallinity, layer stacking and orientation). However, these mechanisms cannot be directly extrapolated to 3D polycrystalline nanostructures due to the added dimensionality and intricate morphology. Thus, integrating 3D arrangement of graphene flakes into functional electronic devices and developing fundamental understanding of electron transport in such structures remains an open challenge.

Here we report synthesis of a multi-dimensional graphene nanostructure: nanowire templated-3D fuzzy graphene (NT-3DFG), where each nanowire is composed of controlled 3D arrangement of graphene flakes free-standing on the surface of a one-dimensional (1D) silicon nanowire (SiNW). Such an arrangement exposes both surfaces of each graphene flake maximizing the surface-area-to-volume ratio and exposed edges. Free-standing graphene flakes form a graphitic structure surrounding the SiNW core. Electron transport in NT-3DFG occurs through parallel channels formed by the graphitic shell (metallic transport) and the free-standing graphene (variable range hopping (VRH)). The material exhibits negative magnetoresistance from cryogenic conditions up to room temperature, which we attribute to a VRH-based interference mechanism distinct from weak localization. Our study opens new avenues for synthesizing and characterizing 3D arrangement of 2D materials to understand electron transport in multi-dimensional nanostructures.

Two-dimensional (2D) transition-metal carbides/nitrides (MXenes) have attracted great attention in energy storage and electrochemical applications because of their excellent electrical conductivity and flexible layered structures. 2D transition metal dichalcogenides (TMDs) are widely utilized in gas-sensing devices for the detection of NH₃, NO₂, H₂, and volatile organic compounds (VOCs) owing to their high specific surface area and tunable electronic structures. Therefore, hybridization of MXenes nanomaterials with TMDs, currently the main type of advanced sensing materials, is a promising approach to the manufacturing of low-cost, high-performance gas sensing devices.
A wireless flexible gas sensor prepared by a simple fabrication process is appealing because it suits the needs of the internet of things (IoTs). We have been working on gas-sensing 2D materials by hybridization (and surface functionalization) of a given MXene, Ti₃C₂T_x, with a variety of TMDs (e.g., MoS₂, WSe₂). Herein, we report the results of using liquid phase exfoliation (LPE) and inkjet printing, both being the most promising cost-effective, mass-production methods, to synthesize a typical TMD/MXene nanohybrids, 2D MoS₂/Ti₃C₂T_x. The hybridization of MoS₂ and Ti₃C₂T_x films through surface modification and electrostatic interaction will be reported.
The research results indicated that, even at room temperature, a MoS₂-decorated Ti₃C₂T_x gas sensor exhibited superior ability to detect extremely low concentrations (1 ppm) of VOCs, as compared to a pristine Ti₃C₂T_x (reference) sensor. Furthermore, the sensitivity of the MoS₂/Ti₃C₂T_x sensor is more than 10-fold higher than that of the pristine Ti₃C₂T_x sensor resulting from the synergetic effect of high electrical conductivity of Ti₃C₂T_x and high specific surface area of MoS₂ favoring gas adsorption. A variety of material analyzing methods will be used to elucidate the reasons behind the achievement of sensing performance by optimal decoration of MoS₂. For example, chemical bonding structures of MoS₂/Ti₃C₂T_x nanohybrids will be studied by X-ray photoelectron and Raman spectroscopies. Surface morphologies and microstructures of the hybrid materials will be analyzed by scanning electron microscopy and transmission electron microscopy.

Organic films with uniform, ultra-thin thickness can serve as building blocks to be assembled into more complex structures by sequential transfer processes. The approach provides organic interfaces engineered at the molecular-scale, which form the basis for discovery of novel organic materials and development of functional organic devices. However, transfer of such films often result in structural damages to the films, due to their weak mechanical strength. In addition, programmed organic interfaces easily deform by intermolecular diffusions. Here, we report several molecular thick C₆₀ and pentacene films with high spatial uniformity as prototypical electron acceptor and donor components, and transfer them layer-by-layer while maintaining structural integrity in the final structures. To obtain the structural stability, we integrate organic films with monolayer hexagonal boron films (ML-hBN), which act as both supporting layer for transfer and molecular diffusion blocking layer across organic interfaces. Layer-by-layer assembly of different organic films successfully generates composite films, including C₆₀/pentacene pn diode and C₆₀/pentacene vertical superlattice, where thickness and composition are modulated at the molecular-scale. Ultra-thin organic photodiodes and photoconductors have been demonstrated based on the precisely programmed organic interfaces and shown high performances.

Two dimensional (2D) transition metal dichalcogenides (TMDs) have generated strong interest between a wide scientific audience due to the variety of promising applications for nanoelectronics, photonics, sensing, energy storage, and opto-electronics, to name a few. In particular, tungsten disulfide (WS₂), tungsten diselenide (WSe₂) and molybdenum disulfide (MoS₂), individually or their combination to form vertical and lateral heterostructures, present unique mechanical, electrical and optical properties, being exceptional candidates for the fabrication of opto-electronic devices. The physical properties of these structures are strongly dependent of the quality of the materials and presence of defects, limiting somehow the performance in devices. ¹
We have studied the subsurface defects with diverse Scanning Probe Microscopies (SPM) based in the Atomic Force Microscopy (AFM). The combination of the AFM with ultrasonic excitation of the sample or/and the cantilever allow the mapping of the mechanical properties with nanoscale resolution – namely, Ultrasonic Force Microscopy (UFM), Waveguide UFM (W-UFM) and Heterodyne Force Microscopy (HFM), respectively.² Similarly, by the electrical excitation of the sample, we can probe the dielectric properties and materials work function, using Dielectric Electrostatic Force Microscopy (D-EFM) and Kelvin Probe Force Microscopy (KPFM). The preliminary results of the top surface nanomechanical maps of vertical heterostructures of WS₂ and WSe₂ with pyramidal shape, show clear contrast corresponding to missing planes in the buried layers, subsurface mis-orientation of the crystallographic axis and edge dislocations. To complete the study, the pyramids have been sectioned with the Beam Exit Cross-sectional Polishing (BEXP) permitting the direct scanning of the inner part of the 3D structures, identifying the layers and revealing inhomogeneities in the layer stack via local mechanical and electrical properties mapping. For lateral heterostructures of WS₂ and MoS₂ we use nanomechanical maps to reveal the unique mechanical behaviour creating ripples at the interface of the two materials, likely corresponding with the strain produced by the lattice mismatch.³

1. W. Choi, N. Choudhary, G. H. Han, J. Park, D. Akinwande and Y. H. Lee, Materials Today 20 (3), 116-130 (2017).
2. M. T. Cuberes, H. E. Assender, G. A. D. Briggs and O. V. Kolosov, Journal of Physics D-Applied Physics 33 (19), 2347-2355 (2000).
3. Y. Han, K. Nguyen, M. Cao, P. Cueva, S. Xie, M. W. Tate, P. Purohit, S. M. Gruner, J. Park and D. A. Muller, Nano Letters 18 (6), 3746-3751 (2018).

We have studied theoretically how a generic bilayer kagome lattice behave upon layer rotation. Based on Density Functional theory calculations we developed a Tight Binding model with one orbital per site and found for low rotational angles and at low energies the same flat bands structure like in twisted bilayer graphene, moreover at high energies and due to the superstructure symmetry regions we found a characteristic three band dispersion of a kagome lattice, its width decreases for low angles confining them within a few meV, generating two sets of flat bands in different energy regions and localized in different spatial regions. The value of the magic angle is greater than in the rotated graphene bilayer which could result in a higher superconductivity temperature.
The kagome lattice arises in metal organic frameworks(MOF) where metals and organic ligands are the building blocks [1], a phenomenon as exotic as a quantum spin liquids was reported in spin 1/2 kagome-lattice antifferromagnet herbertsmithite[2].
Our tight binding parameters are based on a recent experimental measurement of a bilayer kagome compound Fe₃Sn₂ [3], the system is a soft ferromagnet with intrinsic anomalous Hall conductivity, the Dirac cones are separated by a gap of 30 meV as a consequence of the spin-orbit coupling.
One way to achieve larger magic angle is by increasing the relative value of the interlayer coupling, in MOF the effective coupling within the layer might be of the same order as the interlayer coupling, we explore also how this ratio might enhance the value of the angle at which the magic angle is obtained and eventually it might results in larger critical temperature.

[1] D. Rodrguez-San-Miguel, P. Amo-Ochoa, and F. Zamora. Chem. Commun. 52, 4113 (2016).
[2] T.H. Jan et al. Nature 492, 406 (2012)
[3] M. Kang, D.C. Bell, R. Comin, et al. Nature 555, 638(2018)

Tailoring 2D semiconductor heterostructures with specific bandgaps is a key aspect of leveraging new quantum materials for electronics and optoelectronics. The basic heterostructure assembly principle is simple: exfoliate, for example, a monolayer of MoS₂, put it on top of another mono- or few- layer crystal, c.a. WS₂, add another 2D crystal, and so on. The resulting heterostructure represents an artificial material assembled in a specified sequence with single layer precision, held together by van der Waals forces.

Raman micro-spectroscopy is used to characterize a variety of properties in 2D materials, including chemical, magnetic, electronic, symmetry, and layer orientation. Many interesting phenomena emerge across the 4K-500K temperature range. Often, new information about a sample can be obtained with temperature dependent measurements. We demonstrate efficient hyperspectral mapping of 2D materials with a micro-Raman spectroscopy and photoluminescence platform that maintains high spatial resolution and collection efficiency across the entire temperature range (4K – 525K). The system overcomes challenges associated with variable temperature sample drift by incorporating a low thermal mass sample stage, in-vacuum high NA objective, and coupling optics to a low astigmatism, broadband spectrograph.

The exciton recombination dynamics of stacked MoS₂-WS₂ heterostructures were previously studied at room temperature. The maximum of the valence band and the minimum of the conduction band are separated in the layers of WS₂ and MoS₂, respectively. The reported energy gap for MoS₂ is 2.39eV and for WS₂ is 2.31eV. The energy difference between the maximum valence bands of MoS₂ and WS₂ is about 350meV. Photoexcited free electron–hole pairs prefer to stay separated at layer interfaces. The excited electrons in WS₂ tend to accumulate in the conduction band of MoS₂ while holes generated in the valence band of MoS₂ tend to transfer to WS₂ at the interface. The interlayer radiative recombination of spatially separated carriers can lead to the extra peaks in the PL spectra. We observed the temperature dependence of the interlayer exciton in the hyperspectral mapping study which is at lower energy compared with pure monolayer MoS₂ and WS₂.

One-dimensional (1D) nanostructures such as nano-wires, -tubes, -rods and -fibers, have garnered great attention due to their excellent electrical, thermal, mechanical and optical properties and their suitability in a myriad of applications and devices. Applications of 1D nanostructures exist in solar cells^1,2, LEDs³ and transparent conductors amongst other applications.

In this study, we demonstrate the previously unrealized ability to grow nanorods and nanotubes of 2D materials by femtosecond laser treatment of the respective flakes in solution. In as short as 20 mins, nanorods of tungsten disulfide, molybdenum disulfide, graphene, and boron nitride grow resulting in an up to 10-time increase in length. Due to the laser treatment, it was found that for some materials, notably the WS₂ particles reassemble to form new phases. Also resulting in the chemical modification of the nanorods with functional groups from the solvent atoms. The WS₂ untreated flakes are originally in the crytallographic 2H phase, the laser treatment reassemble and induces new primary bonds growing the particles into nanorods consisting of the 1T metallic crystallographic phase. Due to this transition, alongside the one-dimensional nature of the fabricated nanorods, the WS₂ nanorods display substantial improvements in electrical conductivity and optical transparency when employed as transparent conductors.

(1) Garnett, E. C.; Brongersma, M. L.; Cui, Y.; McGehee, M. D. Nanowire Solar Cells. Annu. Rev. Mater. Res. 2011, 41 (1), 269–295. https://doi.org/10.1146/annurev-matsci-062910-100434.
(2) Musselman, K. P.; Wisnet, A.; Iza, D. C.; Hesse, H. C.; Scheu, C.; MacManus-Driscoll, J. L.; Schmidt-Mende, L. Strong Efficiency Improvements in Ultra-Low-Cost Inorganic Nanowire Solar Cells. Adv. Mater. 2010, 22 (35), E254–E258. https://doi.org/10.1002/adma.201001455.
(3) Könenkamp, R.; Word, R. C.; Schlegel, C. Vertical Nanowire Light-Emitting Diode. Appl. Phys. Lett. 2004, 85 (24), 6004–6006. https://doi.org/10.1063/1.1836873.

Manual assembly of atomically thin materials into heterostructures with desirable electronic properties is an approach that holds great promise. Despite the rapid expansion of the family of ultra-thin materials, stackable and stable ferro/ferri magnets that are functional at room temperature are still out of reach. We report the growth of air-stable, transferable ultra-thin iron oxide crystals that exhibit magnetic order at room temperature. These crystals require no passivation and can be prepared by scalable and cost-effective chemical vapor deposition. We demonstrate that the bonding between iron oxide and its growth substrate is van der Waals-like, enabling us to remove the crystals from their growth substrate and prepare iron oxide/graphene heterostructures.

Large-scale manipulation and structural modifications of two-dimensional (2D) materials for nanoelectronic and nanofluidic applications remain obstacles to their industrial-scale implementation. Here, I will discuss how focused ion and electron beams can be utilized to engineer defects and tailor the atomic, optoelectronic, and structural properties of monolayer and heterostructure 2D materials such as graphene, phosphorene, hexagonal boron nitride, and various transition metal dichalcogenides (TMDs). A combination of aberration-corrected scanning transmission electron microscopy (AC-STEM), Raman spectroscopy, photoluminescence spectroscopy, and density functional theory (DFT) is used to study the characteristics of 0-dimensional (0D) defects and other low-dimensional nanostructures such as nanoribbons, antidot arrays, and nanoporous membranes. The results shown here lend the way to the scalable fabrication and processing of 2D nanodevices.

Graphene, an atomically thin material, is an excellent candidate for next generation sensing platform thanks to its unique electrical, optical, mechanical and chemical properties. Its all-surface structure gives it high sensitivity to the environment. However, it also makes graphene-based devices very sensitive to fabrication process. Small differences in graphene quality, such as surface contamination and defects, can lead to large variation in device behavior.

In this work, we developed a graphene-based sensor array functionalized with ion selective membranes (ISMs) that enables us to measure ion concentration in an electrolyte. With more than 200 functional devices per array, the system is able to detect ion concentration spanning over several orders of magnitude with a near-ideal Nernstian sensitivity, excellent selectivity, high reversibility and fast response time. The intrinsic variability in 2D material based sensor performance can be greatly mitigated by taking advantage of the large sample size and utilizing the redundancy in device response. Sensor redundancy helps to tight the 95% confidence intervals from ±50% to ±10%, while the reversibility can be increased from less than 80% to almost 95%.

This work uses an ion-selective functionalization based on a polyvinyl chloride based membrane with ionophores targeting specific ions, such as calcium, sodium and potassium. We chose these ions because their physiological importance. A material jetting 3D printer is used to print the ISMs with accurate control of the size and location. Several different ISMs can be deposited onto different regions of the sensing array, which allows for a multiplexed sensor array. By using statistical analysis techniques we are able to measure the concentration of each target ion within a complex background of interfering ions.

In summary, we developed a novel sensor platform that can simultaneously measure over 200 graphene-based multiplexed sensors to characterize the concentration of sodium, calcium and potassium ions in aqueous phase. By collecting data from a statistically significant sample size, this work presents a manufacturable and reliable technology that can be employed for a wide range of physiologic monitoring applications.

Deposition of metal contacts on graphene is a crucial step for the fabrication of a wide array of novel optoelectronic and sensing devices. Magnetron sputtering is a state-of-the-art versatile technique for large-scale thin metal film synthesis. However, its inherent tendency for generating energetic species (both neutrals and ions) that cause ballistic defects and disorder on the surface upon which they form a film, renders sputtering incompatible with processes where graphene is used as substrate. The objective of this study is to establish the effect of sputtering conditions, during growth of thin Ag films, on the structure and electronic properties of graphene. Films are grown on monolayer graphene synthesized by SiC sublimation of C-face 4H-SiC and by CVD deposition on Cu foils. We use top-view SEM from which we establish that the produced Ag films exhibit a pronounced 3D growth morphology that can be attributed to the weak interaction between Ag atoms and graphene. We also employ micro-Raman spectroscopy and define the graphene crystalline quality and the point-defect density as a function of the deposition parameters. The energetic sputtered species cause disorder, evidenced by the appearance of the defect-related D-band and inhomogeneous broadening of the characteristic Raman peaks due to charge and local strain fluctuations. We also find that the effect of the energetic deposition flux on the pristine graphene structure and electronic properties becomes less pronounced with increasing sputtering pressure. These findings suggest that it may be possible to facilitate disorder-free growth of thin metal films on graphene by magnetron sputtering if deposition parameters (e.g., sputtering pressure, target-to-substrate distance, magnetic field configuration) are carefully adjusted in a way that minimizes the energetic input from the sputtering flux to the growing surface.

Black phosphorus is an intriguing material with intrinsic anisotropy induced by its unique puckered honeycomb structure. In bilayer or multi-layer black phosphorus, the system's anisotropy as a whole can be tuned by an interlayer twist angle, and thereby the material's electronic, optical, and mechanical properties. Thus, we here try to investigate the effect of interlayer twist angles or stacking orientations on twisted black phosphorus using first-principles calculations. The effects of twist angle on its band structure, phonon dispersion, mechanical property, etc. are presented. We discuss the mechanism of controlling electronic structures by twisting layers, and also, an effective way to handle the long-range Moire superlattice induced by incommensurate lattices in an arbitrary twist angle.

Triggered by the demands for high-performance computing requires much higher bandwidth density for inter-chip communication pushes the limit of device miniaturization. Hence, 2-dimensional (2D) materials have recently emerged as promising building blocks for photonics due to their number of fascinating features. Tunning of electronic and optical properties by engineering local strain is an exciting avenue for tailoring performance for 2D materials based integrated photonic device. Besides, continuous tuning of the strain, the ability to apply a spatially controllable strain by using a patterned substrate is even more crucial because it enables the realization of a graded bandgap semiconductor for the photonic integrated device. Here, we study the effect of large localized strain in the electronic band structure of a multilayer MoTe₂by wrapping the 2D material around a non-planarized Silicon-on-insulator waveguide etched down to the buried oxide. Interestingly, induced tensile strain (4%) shifts the bandgap of MoTe₂to about 0.8 eV obtained using DFT calculation, thus significantly increasing the absorption as compared to their pristine counterpart of bandgap (~1 eV). Here, the device is realized in a two-terminal in-plane electrode configuration without applying external gating, showing a high responsivity of (~0.5 A/W) and NEP of 90 pW/Hz^0.5at -2 V at 1550 nm. Our device demonstrates a response rate of ~35 MHz which is limited by the low mobility (~1 cm²/Vs) of the MoTe₂. The integration of a few-layer MoTe₂on Si MRR as active photodetector is envisaged to offer a potential pathway toward the realization of integrated on-chip interconnects for Telecommunication band.

Two-dimensional (2D) monolayer molybdenum disulfide (MoS₂) has been demonstrated as an excellent semiconductor material for future nanoelectronics because of the suitable bandgap, good intrinsic mobility, thermal velocity, and mean free path etc. To fully explore the potential of monolayer MoS₂ for practical applications such as field-effect transistors, memories, photo-sensors, and solar cells, there is a critical need of metal-semiconductor contact engineering and optimization which can lead to the maximization of device performance. In this work, we exploited 2D monolayer hexagonal boron nitride (h-BN) as a decorating layer at the metal-MoS₂ interface, and demonstrated a significant improvement of the carrier transport through a quantum tunneling effect. Both monolayer MoS₂ and h-BN were synthesized by using two-zone chemical vapor deposition (CVD) method. The triangular domains of monolayer MoS₂ were synthesized directly on SiO₂/Si substrates and the continuous monolayer h-BN film was grown on copper foils. The field-effect transistor (FET) and transmission line measurement (TLM) devices were fabricated using electron beam lithography (EBL) and evaporation (10 nm Ti and 100 nm Au). Compared to the conventional MoS₂/Ti/Au contact where the thermionic emission is the dominating mechanism, the MoS₂/h-BN/Ti/Au contact shows the outstanding improvement of the carrier transport. For example, the on-current density was increased by two orders of magnitude, and the electron mobility was improved by ten times. These improvements can be attributed to the dominance of the quantum tunneling when the monolayer h-BN is inserted at the metal-semiconductor interface. Based on the temperature-dependent measurement, a lowering of the effective metal-semiconductor barrier height from ~60 to ~30 meV was induced by the monolayer h-BN decoration, and the reduction of the contact resistance was confirmed by the TLM technique. Our work has demonstrated the great potential of this novel contact engineering technique using the 2D monolayer insulator, which can be applied to other 2D materials and devices for broad applications.

For the newly emerging IoT sensors and display applications, low dimensional semiconductors such transition metal dichalchogenides (TMDCs) can be one of appealing candidates beyond graphene because of their excellent electrical-switching-characteristics, together with novel mechanical, and optical properties. Among them, molybdenum disulfide (MoS₂) has been regarded as one of the most promising materials. Until to this date, in-depth knowledge on MoS₂ FETs in terms of electrical performance improvement, transport mechanism, back channel effects, etc have been accumulated through tremendous research activities. However, device reliability studies on MoS₂ FETs have been limitedly reported. Thus, systematic investigation on device instability of MoS₂ FETs under DC and pulsed mode operation and its comparison on instability behaviors, followed by understanding on the behind mechanism, are highly necessary in order to apply TMDC FETs to next-generation display platforms which are required for ultra-high definition, mechanical flexibility, electrical stability, and others.
Furthermore, for a quantitative understanding on gas adsorption effects on electrical performance of MoS₂ FETs, which are regarded as one of main factors for instability of TMDCs, information on subgap density of states (DOS) in MoS₂ layers without (or with) hydrophobic polymer (CYTOP) encapsulation is one of pre-requisite parameters. On the other hand, in a recent past, trap information on MoS₂ FETs has been quantitatively analyzed by several techniques, whereas the previous reports for characterization of sub-gap traps for m-MoS₂ FETs were limitedly addressed; (i) D_it characterization via Terman method, (ii) shallow traps nearby conduction band (E_c) via multi-frequency method, and (iii) deep level traps through hysteretic gate transfer characteristics, thereby entire characterization for sub-gap DOS was fundamentally limited by method itself and its characterization.

In this study, we investigated into device instability of MoS₂ FETs associated with series of key parameters such as bias polarity, temperature, duty cycles, bias amplitude, and others. Asymmetry behaviors on bias stress instability under DC and pulsed mode operation were reproducibly observed and their mechanisms were systematically analyzed by using frame time dependency in pulsed mode operation, rationally supported by atomic and band models. Moreover, quantitative analysis on subgap states for MoS₂ FETs before and after CYTOP encapsulation were performed by using optical charge-pumping capacitance-voltage spectroscopy. Based on extracted subgap DOS and their deconvolution with analytical model of acceptor (or donor) like states, all electrical parameters were systematically analyzed. Two times increase of field effect mobility (m_FE) are strongly related with decrease of shallow-donor states (N_SD). In addition, significant improvement of subthreshold swing (SS) and hysteresis gap (V_HYS) are attributed to the reduction of tail states (N_TA), along with decrease in mid-gap states (N_Mid)_. All systematic results in this work are highly anticipated to contribute to the understanding on device instability issues which are required for fulfilling the envisioned applications.

In this work, we design a graphene-based metasurface for independent active modulation of amplitude and phase of mid-infrared light. This independent control over the phase and amplitude can be obtained by applying the alternating voltage to the two consecutive units of periodic graphene-gold multiscale structures (meta-atoms). In complex reflectivity plane, we are able to cover a large area of unit complex reflectivity square by only controlling the values of graphene Fermi energy in each set of meta-atoms.
The proposed structure can be analyzed with an effective surface admittance model [1]. For general multiscale structure with back-reflector, it is known that the reflectivity r and the effective admittance of the metasurface Y_S are related as r = –(Y_S+Y_sub–Y₀)/(Y_S+Y_sub+Y₀), where Y_sub and Y₀ are the admittance of the substrate and vacuum, respectively [1]. In this structure of alternating meta-atoms, the admittance of the pair (meta-molecule) is a serial connection of the half-admittances of each meta-atom Y_S1 and Y_S2: Y_S = 2Y_S1Y_S2/(Y_S1+Y_S2). The admittance of each meta-atom is predicted by a circuit model. The circuit model represents the effective conductivity of the meta-atom as a function of its structural parameters by taking the elements constituting the metasurface as corresponding circuit elements, and then calculating the admittance of the circuit connecting the elements. The gold bars array is represented by a serial connection of inductances and capacitances with the geometrical parameter of bars and slits. The array of graphene ribbons placed between the gold bars is treated with the effective admittance of the free-standing graphene ribbon array multiplied by the gain factor P/g, corresponding to the electric field amplification in the gap. In this way, we can approach the complex reflectivity of the metasurface from the viewpoint of its structural parameters, and optimize the whole structure. Further, by inducing the appropriate Fermi energy independently to each meta-atom of the metasurface, the reflected wavefront can be controlled in arbitrary manner (e.g. dynamic beam steering, focusing etc.)

[1] Kim, S., Jang, M. S., Brar, V. W., Mauser, K. W., Kim, L., & Atwater, H. A. (2018), “Electronically tunable perfect absorption in graphene.” Nano letters, 18(2), 971-979.

MoS₂, along with other transition metal dichalcogenides (TMDCs) nanomaterials, have been demonstrated to be effective in photovoltaic (PV) devices with various functions and structures. Interestingly, the TMDC-based electronic devices can be either vertical or lateral, showing the potential of fabricating TMDC-based PV devices with different dimensionality. Recently, we have synthesized a new class of 1-dimensional (1D) van der Waals (vdW) heterostructures by chemical vapor deposition, in which MoS₂ nanotube (MoS₂NT) is wrapping around single-walled carbon nanotubes (SWCNT) to form a MoS₂NT-SWCNT interface [1]. The application of this kind of 1D vdW heterostructures can allow us to fabricate PV devices with 1D dimensionality.
In this research, we demonstrate the excellent hole-extraction ability of MoS₂NT by applying MoS₂NT-SWCNT to normal-type perovskite solar cells. The as-synthesized MoS₂NT-SWCNT functions as both the hole transport layer and anode at the same time. Comparing with SWCNT, the wrapping of MoS₂NT around SWCNT improved the power conversion efficiency (PCE) by almost 20%, owing to the substantial enhancement of the fill factor. A PCE of 15.0% has been achieved after the application of spiro-MeOTAD. The 1D vdW heterostructures are beneficial by a closer physical distance between layers for a better hole-transport ability than 2D vdW heterostructures. In addition, the integration of different materials in one substance can also allow easier fabrication than layer-stacking, and thus reduce the size of the device and the amount of used materials. The successful application of MoS₂NT-SWCNT in perovskite solar cells are expected to ignite the application of 1D vdW heterostructures, as well as the application of SWCNT and TMDC based nanomaterials in PV devices.

[1] R. Xiang, T. Inoue, Y. Zheng, A. Kumamoto, Y. Qian, Y. Sato, M. Liu, D. Tang, D. Gokhale, J. Guo, K. Hisama, S. Yotsumoto, T. Ogamoto, H. Arai, Y. Kobayashi, H. Zhang, B. Hou, A. Anisimov, Y. Miyata, S. Okada, S. Chiashi, Y. Li, J. Kong, E. I. Kauppinen, Y. Ikuhara, K. Suenaga, S. Maruyama, "One-dimensional van der Waals heterostructures", (2019). [https://arxiv.org/abs/1807.06154] (submitted)

For real implementation of 2D material-based nanoscale device integration, the construction of a manufacturable and highly stable metal/semiconductor building block is critically required. To develop practical 2D material-based electronic and photonic devices, the formation of 2D lateral metal/semiconductor junction structures can overcome fundamental technical obstacles such as high contact resistance and limitation of channel length scaling via conventional vertical junction structures. However, a formation method of large-scale 2D lateral metal/semiconductor junction structure with sufficient levels of controllability and quality has not been demonstrated. We herein propose the formation of a scalable 2D metal/semiconductor Mo₂C/MoS₂ lateral junction structure fabricated via selective synthetic integration of large-scale metal (Mo₂C)/semiconductor (MoS₂) junction. We demonstrate that centimeter-scale metallic 2D MXene (Mo₂C) can be synthesized from CVD-grown few-layer MoS₂ with chemical conversion using CH₄ and Cu catalyst. Further, a dimension-controlled Mo₂C/MoS₂ junction is fabricated by depositing a lithographically patterned SiO₂ masking layer. Results show that while chemical conversion to Mo₂C occurs in the uncovered MoS₂ part, the MoS₂ layer under the SiO₂ masking layer is protected from chemical conversion so that a scalable Mo₂C/MoS₂ heterostructure is integrated down to the nanometer scale, exhibiting excellent resistance of contact resistance of 2.1 kΩ μm. This proposed manufacturable and highly stable 2D material-integrated metal/semiconductor junction structure will provide a scalable building block for the implementation of 2D material-based nanoscale device integration.

Two-dimensional (2D) nanostructured transition metal dichalcogenides (TMDs) are promising candidates for catalyzing the hydrogen evolution reaction (HER). However, even though the free energy of hydrogen adsorption (△G_H) is close to zero for many TMDs, experimental catalytic efficiencies have consistently fallen short of thermodynamic predictions. In order to investigate this disparity, the effect of the support substrate on the MoS₂-catalyzed HER was investigated. It has been demonstrated that both heterostructured contacts and the support substrate can modulate the MoS₂-catalyzed HER, but it is difficult to isolate and identify the specific interfacial effects that influence the catalytic efficiency. To address this challenge, we utilized an electrochemical microreactor capable of isolating individual interfacial effects to probe the catalytic efficiencies of MoS₂ [1]. We show that heterostructured supports enhance the HER activity of monolayer MoS₂ [2].
This talk will examine two possible mechanisms to explain the modulation of the HER activities. The first mechanism is that highly conductive support substrates are better than dielectric substrates because they can effectively screen mirror charges that may form during HER. Indeed, MoS₂-WTe₂ and MoS₂-graphene heterostructures consistently outperformed MoS₂ supported on SiO₂, in agreement with our expectation. To further probe this mechanism, we placed MoS₂ on substrates with dielectric constants ranging from 4 to 300. The catalytic efficiency of MoS₂ decreased with increasing dielectric constant, supporting the idea that mirror charges can inhibit hydrogen production. However, the effect of substrates was proven minor compared to the effect of contact metals. Devices with Ni-graphite contacts significantly outperformed devices with Cr-graphite contacts regardless of the support dielectric constant. This suggests that the catalytic efficiency is largely determined by the second mechanism: charge injection. We attribute the increased activity of the Ni devices to lower contact resistance for electrons at the Ni-graphite interface than through the Cr-graphite interface. Thus, the enhanced HER from the MoS₂-WTe₂ and MoS₂-graphene heterostructures is attributed to the efficient charge injection into MoS₂ through large-area heterojunctions.
These results demonstrate the importance of interfacial design in TMD HER catalysts. The microreactor platform presents an unambiguous approach to probe interfacial effects in various electrocatalytic reactions.

[1] Y. Zhou, et al., Adv. Mater. 2018, 30, 1706076.
[2] Y. Zhou, J. V. Pondick, et al., Small 2019, 1900078.

Beyond the silicon and compound semiconductor industries, two-dimensional van der Waals (2D vdW) materials have been predicted to be able to revolutionize electronics, photonics, and other industrial sectors since they possess many fascinating physical and electrical properties. However, 2D materials currently face challenges along the long-term road to commercialization, with increasing efforts being made to satisfy industrial needs. At this moment, the most critical issue for final 2D device performance is related to realizing good electrical contacts to 2D vdW semiconductors. In fact, most 3D (bulk) metals form non-Ohmic Schottky junctions to 2D vdW semiconductors, resulting in relatively high and bias-dependent contact resistances. Furthermore, the Fermi level at the metal-semiconductor junction (MSJ) is often pinned owing to several types of chemical interactions, and the Schottky-Mott rule generally provides incorrect prediction for the Schottky barrier height (SBH).
In this study, we not only performed the wafer-scale production of patterned transition metal (TM) ditellurides (e.g., WTe₂ or MoTe₂) on desired surfaces and positions with high electrical performance but also demonstrated the experimental formation of monolayer MoS₂ FETs containing 2D WTe₂/MoS₂ vdW MSJs with almost perfect interfaces. We have designed simple and efficient methods to grow high-quality, stoichiometric TM ditellurides with electrical performances comparable to those of mechanically exfoliated flakes. The fabricated 2D (W, Mo)Te₂/monolayer MoS₂ vdW MSJs with close-to-perfect interfaces are free from significant disorder effects and Fermi level pinning on the interface, and their SBHs largely follow the trend of the Schottky-Mott limit. For example, the measured SBH (≈103.5 meV) of the WTe₂/MoS₂ interface was the lowest value experimentally observed for metal electrodes formed on monolayer MoS₂ and was very close to the theoretical value (≈100 meV) calculated by the Schottky–Mott rule. Compared to the MSJs with conventional 3D metals, the fabricated 2D WTe₂/MoS₂ vdW MSJs exhibited a much improved electrical injection across the MSJs, which was attributed to the low SBH and depinning of the Fermi level at the interface. Employing position-controlled, patterned vdW contacts with ideal Schottky barriers provides the advantage of controlling the SBHs in a predictable manner. This ability to fabricate position-controlled, conformal vdW metals with different shapes affords new possibilities for the fabrication of multiple 2D nanoelectronics devices.

Two-dimensional transition metal dichalcogenides are promising candidates for ultrathin optoelectronic devices due to their high absorption coefficients and intrinsically passivated surfaces¹. To maintain these near-perfect surfaces, recent research has focused on fabricating van der Waals contacts that limit Fermi-level pinning at the metal-semiconductor interface^2,3. Here, we develop a new, simple procedure for transferring metal contacts that does not require aligned lithography. Using this technique, we fabricate vertical Schottky-junction WS₂ solar cells with template-stripped Ag bottom contacts, 15-nm-thick exfoliated WS₂ absorber layers, and 20-nm-thick semitransparent Au transferred top contacts. Under laser illumination, we observe rectifying behavior and open-circuit voltage above 500 mV in devices with transferred contacts, in contrast to resistive behavior and open-circuit voltage below 15 mV in devices with evaporated contacts. Under one-sun illumination, we measure an open-circuit voltage of 256 mV, a short-circuit current density of 4.10 mA/cm², and a fill factor of 0.44. We calculate a power conversion efficiency of 0.46%, comparable to what others have observed in ultrathin transition metal dichalcogenide photovoltaics. Guided by device simulations that predict power conversion efficiencies close to 9%, we fabricate and characterize further-optimized devices with transparent top contacts and metal work functions that are better aligned to the multilayer WS₂ conduction and valence bands. Our one-sun measurements and device simulation results indicate that this metal transfer process could enable high-specific-power vertical Schottky-junction transition metal dichalcogenide photovoltaics, and we anticipate that this technique will lead to advances for two-dimensional devices more broadly.

References:
1. Jariwala, D., Davoyan, A. R., Wong, J. & Atwater, H. A. Van der Waals Materials for Atomically-Thin Photovoltaics: Promise and Outlook. ACS Photonics 4, 2962–2970 (2017).
2. Liu, Y. et al. Approaching the Schottky–Mott limit in van der Waals metal–semiconductor junctions. Nature 557, 696–700 (2018).
3. Wang, Y. et al. Van der Waals contacts between three-dimensional metals and two-dimensional semiconductors. Nature 568, 70–74 (2019).

We recently reported the realization of ultra-clean vdW contacts between three dimensional metals and single layer MoS₂. Using scanning transmission electron microscopy (STEM) imaging, we show that the 3D metal and 2D MoS₂ interface is atomically sharp with no detectable chemical interaction, suggesting van-der-Waals-type bonding between the metal and MoS₂. We show that the contact resistance of indium electrodes is ~ 800 Ω-μm – amongst the lowest observed for metal electrodes on MoS₂ and is translated into high performance FETs with mobility in excess of 160 cm²-V-s^-1at room temperature without encapsulation. We have extended this work to other monolayer 2D semiconductors such as MoSe2, WSe2 and their p-n interfaces. We have also studied alloying of indium contacts with other metals to engineer the work function of the electrodes for facile electron and hole injection. I will describe our efforts on making good contacts on 2D semiconductors.

For years, solution processing has enabled researchers to employ transition metal dichalcogenides (TMDs) in fields not accessible with traditional solid-state routes and rigid substrates. Scalable synthesis and exfoliation, chemical functionalization, and hybridization are particularly promising advantages of solution-based methods for TMDs. Colloidal suspension allows for facile assembly of TMDs, often of a desired size and thickness, onto arbitrary substrates. Exploratory biomedical applications, from diagnostics to therapy, have benefited from this unprecedented access to 2D semiconductors in solution and resultant functionalization possibilities. Solution-based hybridization between TMDs and complementary materials has enabled developments in electrocatalysis, sensing, and battery technologies, demonstrating the true breadth of diverse fields these techniques can impact. However, some of the most interesting TMD structures, including heterostructures, are barred from these modification and assembly opportunities because existing solution-based synthesis and exfoliation approaches are not compatible with complex geometries. Here we introduce a platform to successfully transfer artificial two-dimensional heterostructures synthesized with solid-state deposition to an aqueous solution. This essential transfer generates a stable solution where heterostructures can be functionalized for a variety of new applications or transferred to other substrates to create hierarchical assemblies.

Recently, we reported unique nanoparticle@TMD core-shell architectures (e.g. Au@MoS₂) and demonstrated their promising properties for electrocatalysis and optoelectronic devices. Subsequently, we have developed an approach to synthesize TMD core-shell heterostructures and utilize Au nanoparticles as functional vehicles to transfer these unusual structures into solution. We use a metal sulfurization process to synthesize Au@MoS₂@WS₂ structures on rigid substrates. High-resolution electron microscopy and Raman spectroscopy confirm the complete conversion to MoS₂ and WS₂ and high crystallinity of the TMDs. We then suspend these particles in water and demonstrate MoS₂/WS₂ heterostructures in solution. These core-shell heterostructures exhibit exceptional natural stability and monodispersity in solution, making them viable in many applications. Additionally, we leverage this colloidal suspension to chemically functionalize the core-shell particles and further expand application opportunities.

This presentation will also cover two of the broad future avenues enabled by this research approach. On one hand, the solution of core-shell particles itself can be employed via chemical functionalization or hybridization with other functional materials. For instance, biomedicine is one particularly ripe area for this application. These particles could be functionalized to carry chemotherapy drugs, or the TMD/plasmonic Au core hybrid could be leveraged for photothermal cancer treatment. On the other hand, this new platform allows one to conveniently transfer the particles to any substrate, including flexible substrates incompatible with high temperature synthesis. Directed assembly into artificial lattices or unique patterns will be achievable to fully exploit the interplay between the heterostructure shell and plasmonic core. Thus, by introducing a broad platform to create colloidal suspensions of artificial structures, this work opens the door for advanced engineering and utilization of TMD heterostructures.

Miniaturization over three-dimensions is very attractive for future on-chip technologies where device efficiencies need to be optimized over small footprints. This is a new challenge, as device miniaturization has been focused to achieve planar-geometries primarily.
Direct Ink Writing (Robocasting), is an additive manufacturing technique that brings the possibility of fabricating architectures with programmable design in the three-dimensions (3D) at different length scales.
Here, we demonstrate 3D printed electrodes for microsupercapacitors and energy conversion systems from highly concentrated, water-based 2D atomically thin material inks using robocasting. The inks are composed by highly concentrated atomically thin sheets of transition metal chalcogenides either exfoliated from bulk powders or obtained via direct synthesis in solution. By tailoring the rheology of our formulated inks, printability has been achieved along with mechanical robustness of the printed architectures.
The printed architectures, from woodpile to interdigitated electrodes, are extended over a few mm in the three-dimensions and present electrodes widths as small as 100 μm. The 3D printed microsupercapacitors show leading areal capacitance and energy density as compared to planar microsupercapacitors, and stability in different electrolytes.

Solution processing of 2D materials [1] allows simple and low-cost techniques such as inkjet printing [2,3] to be used for fabrication of heterostructure of arbitrary complexity. However, the success of this technology is determined by the nature and quality of the inks used. Furthermore, these formulations must be suitable for all-inkjet printed heterostructure fabrication - the remixing of different 2D crystal gives rise to uncontrolled interfaces, resulting in poor performance and lack of reproducibility of the devices. 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 multi-stack fabrication [4]. Examples of all-inkjet printed heterostructures, such as arrays of photosensors [4], logic memory devices [4], capacitors [5] and transistors [5, 6] will be discussed.

[1] Coleman et al., Science 331, 568 (2011)
[2] Torrisi et al, ACS Nano 6, 2992 (2012)
[3] Finn et al. J. Mat. Chem. C 2,925 (2014)
[4] McManus et al, Nature Nanotechnology, doi:10.1038/nnano.2016.281 (2017)
[5] Worsley et al, ACS Nano, DOI: 10.1021/acsnano.8b06464
[6] Lu et al, ACS Nano, submitted

2D materials hold the promise for many interesting applications in electronics. In this talk, we will evaluate their potential towards continued transistor scaling and provide materials targets which enable device performance comparable or surpassing that of scaled Si transistors. While transistor scaling for digital applications is the most coveted for application, MX2 materials have lower mobility than Si when deposited on wafer. Other applications in electronic circuits which do not require very high mobility have to be considered. We will approach this from the perspective of system and device co-optimization and discuss possible applications which require transits with very low leakage.
At imec, we are building a 300mm-wafer platform 2D materials which is being used to assess experimentally the manufacturability of devices with MX2 channel. In this context, we have demonstrated full-wafer (300mm) growth of MoS2 and WS2 by MOCVD and worked on transferring these materials from the growth wafer to a device wafer. We will describe the integration flow we have developed in the imec CMOS fab.
While progress in many aspects of device fabrication with MX2 materials is encouraging, several steps still lack fundamental understanding and methods to solve them. We will outline these outstanding problems and detail current status of where potential showstoppers appear.

The 2D van der Waals crystals have shown great promise as potential future electronic materials due to their atomically thin and smooth nature, highly tailorable electronic structure, and mass production compatibility through chemical synthesis. Electronic
devices, such as field effect transistors (FETs), from these materials require patterning and fabrication into desired structures.
Specifically, the scale up and future development of “2D-based electronics will inevitably require large numbers of fabrication steps in the patterning of 2D semiconductors, such as transition metal dichalcogenides (TMDs). This is currently carried out via multiple steps of lithography, etching, and transfer. As 2D devices become more complex (e.g., numerous 2D materials, more layers, specific shapes, etc.), the patterning steps can become economically costly and time consuming. Here, we developed a method to directly synthesize a 2D semiconductor, monolayer molybdenum disulfide (MoS2), in arbitrary patterns on insulating SiO2/Si via seed-promoted chemical vapor deposition (CVD) and substrate engineering. This method shows the potential of using the prepatterned substrates as a master template for the repeated growth of monolayer MoS2 patterns. Our technique currently produces arbitrary monolayer MoS2 patterns at a spatial resolution of 2 μm with excellent homogeneity and transistor performance (room temperature electron mobility of 30 cm2 V−1 s−1 and on–off current ratio of 107). Extending this patterning method to other 2D materials can provide a facile method for the repeatable direct synthesis of 2D materials for future electronics and optoelectronics.

Studying the intrinsic behavior 2D materials requires attention to both external and internal sources of disorder. This talk will first review the techniques used to create clean heterostructures with hBN to reduce environmental disorder. In graphene, ten years of progress has led to device performance now rivaling he highest-quality GaAs-based heterostructures. On the other hand, semiconducting transition metal dichalcogenides (TMDs) are also limited by atomic defects within the crystalline layers, which requires efforts in synthesis and characterization of high purity crystals. This talk will present recent progress in synthesis of TMD crystals with dramatically lower defect density using a self-flux techniuqe. Combining higher crystal quality and clean encapsulation allows observation of greatly enhanced optical properties, including near-unity photoluminescence quantum yield, and long excited-state lifetime in TMD heterostructures. In addition, electronic transport measurements show improved carrier mobility and reveal many new details in magnetotransport measurements, including observation of fractional quantum Hall states in monolayer TMDs. These high-quality crystals also allow studies of twisted bilayer TMDs, which show the emergence of many-body correlated states.

Sample synthesis is supported under NSF DMR-1420634
Studies of magnetotransport in monolayer WSe2 are supported under DOE SC-0016703
Studies of twisted bilayer materials are supported under DOE DE-SC0019443

One of the major challenges that limits the development of 2D semiconductors is the low mobility of electrons/holes. Here using Boltzmann transport theory with the scattering rates determined from first principles, which allows us to accurately calculate the intrinsic (phonon-limited) mobility, we will present the intrinsic mobility for a variety of 2D semiconductors, including MX2 and monoelement ones. Moreover, we will discuss how the mobilties are related with underlying physical properties, causing the mobility difference across different materials, and (if time allows) how the strain and thickness would alter the mobility.
Ref: L. Cheng, Y. Liu, JACS, DOI: 10.1021/jacs.8b07871

Band tails are ubiquitously observed at the band-edge of a semiconductor, often characterized by an exponential decay in absorption below the band gap. Here, we characterize for the first time these band tail states in various ultrathin (<20 nm) transition metal dichalcogenide (TMDC) photovoltaic devices using photocurrent spectroscopy. We attribute the origin of these tail states to both intrinsic disorder within the TMDC (e.g. MoS₂, WS₂, MoSe₂, WSe₂) layer as well as the surrounding electrostatic environment in the form of a built-in field. Using our experimental measurements of the band tail states, we further analyze their implications on the fundamental limits of open circuit voltage in ultrathin TMDC photovoltaics using optoelectronic reciprocity relations. We find that in addition to the external radiative efficiency of TMDCs, band tail states may reduce the achievable open circuit voltage by over 100 mV due to increased recombination and therefore sets the maximum efficiency of ultrathin TMDC photovoltaics.

Charge and thermal transport in a crystal are carried by free electrons and phonons (quantized lattice vibration), the two most fundamental quasi-particles. Above the Debye temperature of the crystal, phonons mediated thermal conductivity (k_L) is typically limited by mutual scattering of phonons, which results in k_L decreasing with inverse temperature, whereas free electrons play negligible role in k_L. We report an unusual case in a charge-density-wave single crystal, where k_L is limited instead by phonon scattering with free electrons, resulting in a temperature-independent k_L. In this system, the conventional phonon-phonon scattering is alleviated by its uniquely structured phonon dispersions, while unusually strong electron-phonon (e-ph) coupling arises from its Fermi surface strongly nested at wavevectors where phonons exhibit Kohn anomalies. The finding reveals new physics of thermal conduction, offers a unique platform to probe e-ph interactions, and provides potential ways to control heat flow in materials with free charge carriers.

Heat management becomes more and more critical, especially in miniaturized modern devices, so the exploration of highly thermally conductive materials with electrical insulation and favorable mechanical properties is of great importance. Here, we report that high-quality monolayer boron nitride (BN) has a thermal conductivity (κ) of 751 W/mK at room temperature. Although smaller than that of graphene, this value is larger than that of cubic boron nitride (cBN) and only second to those of diamond and lately discovered cubic boron arsenide (BAs). Monolayer BN has the second largest κ per unit weight among all semiconductors and insulators, just behind diamond, if density is considered. The κ of atomically thin BN decreases with increased thickness. Our large-scale molecular dynamic simulations using Green-Kubo formalism accurately reproduce this trend, and the density functional theory (DFT) calculations reveal the main scattering mechanism. The thermal expansion coefficients (TECs) of monolayer to trilayer BN at 300 to 400 K are also experimentally measured, and the results are comparable to atomistic ab initio DFT calculations in a wider range of temperatures. Owing to its wide bandgap, high thermal conductivity, outstanding strength, good flexibility, and excellent thermal and chemical stability, atomically thin BN is a strong candidate for heat dissipation applications, especially in the next generation of flexible electronic devices.

Stacking two-dimensional (2D) materials, like graphene (GR) and single layer transition metal dichalcogenides (1L-TMDs), has opened up unlimited possibilities to design and engineer new functional 2D van der Waals heterostructures (HSs).
Here we exploit ultrafast transient-reflection spectroscopy to optically investigate the charge transfer (CT) dynamics in a large area 1L-WS₂/GR HS, with an unprecedented temporal resolution, i.e. sub-20fs, and across a broad energy range spectrally covering both A and B excitons of WS₂ (1.8-2.5eV).
When the HS is excited with a photon energy of 2eV, above the optical gap of 1L-WS₂ (1.98eV), the differential reflectivity signal, i.e. ΔR/R, is dominated A and B excitonic features of WS₂. On the other hand, if carriers are photo-injected in the HS well below the WS₂ bandgap, e.g. with 0.8eV photon energy, only graphene is actually excited. Nevertheless, the signal across A and B peaks appears immediately after excitation. This surprising result suggests an extremely fast CT from GR to WS₂. Indeed, when WS₂ alone is excited below bandgap with the same photon energy and fluence, no transient signal is detected.
The observed ultrafast charge transfer cannot be exclusively attributed to the direct excitation of new intermediate charge transfer states arising from the overlap of the graphene and the TMD bandstructures[1]. This is indicated by the excitation fluence dependence of the transient signal, which becomes strongly nonlinear at infrared excitation energies, revealing that the mechanism behind the observed ultrafast CT from graphene to the semiconductor is hot electron/hole transfer[2].
The timescale for the hot electron transfer is expected to be extremely short, below 50 femtoseconds[2]. The rise time of the A exciton dynamics is a direct estimation of the timescale for this process, and it is related both to the hot GR Fermi-Dirac distribution thermalization and the tunneling process, i.e. the time needed for hot electrons, exclusively excited in GR, to overcome the Schottky barrier and reach the semiconductor[3]. The extracted build-up of the A exciton transient signal, measured on the HS following a sub-20fs IR excitation at 1.3eV, is faster than 20 fs. This is the first direct measurement of the hot electron/hole transfer in time-domain.
Hot electron transfer is extremely promising for charge extraction at the HS interface, making this novel 2D HS suitable for the development of broadband and efficient low-dimensional photodetectors.

[1] Yuan, et al., Sci. Adv., 4 (2018)
[2] A. Tomadin, et al., Phys. Rev. B 88, 035430 (2013)
[3] Massicotte et al., Nat. Commun., 7, 12174 (2016)

Organic/2D heterostructures have emerged as promising materials for ultrathin, large-area optoelectronic devices. Both organic and 2D semiconductors lack dangling bonds at their surfaces and interact via van der Waals forces, enabling their heterostructures to form ideal interfaces without the need for energy-intensive growth processes, such as molecular beam epitaxy. As such, ultrathin optoelectronic devices have been fabricated from organic/2D heterostructures including photovoltaics, photodetectors, and field-effect transistors, demonstrating large on-off ratios and anti-ambipolar behavior in the latter. Recently, organic conjugated polymer/2D heterostructures fabricated from the polymer, PTB7, and the 2D material, MoS₂, have demonstrated the photovoltaic effect with record values of the photovoltaic figures of merit normalized to the device active layer thickness. Despite this, devices fabricated from organic/2D heterostructures have shown relatively low internal quantum efficiencies, suggesting non-ideal charge transport throughout their devices. Investigating charge transfer dynamics across organic/2D semiconductor interfaces at fundamental timescales is an important part of understanding how to improve the low internal quantum efficiency of these devices.

Here, we employ photoluminescence and femtosecond transient absorption spectroscopies to study the charge transfer dynamics in large-area, organic conjugated polymer/MoS₂ heterostructures. The heterostructures were formed between chemical vapor deposited monolayer MoS₂ thin-films and 3 different conjugated polymer films: P3HT, PCDTBT, and PTB7. These 3 polymers have been used extensively as electron donors in high-efficiency polymer:fullerene organic photovoltaics, and each form Type-II heterojunctions with MoS₂. Although electron transfer was expected to occur from the polymers to MoS₂, we show that electron transfer occurs in the opposite direction: from MoS₂ to each of the conjugated polymers via polaron pair states. We show that electrons are transferred from MoS₂ to P3HT within 9 ps, and from MoS₂ to PCDTBT and PTB7 in under 120 fs. Despite this, we demonstrate that the P3HT/MoS₂ heterostructure is the most efficient because the transferred charges have an order-of-magnitude increase in their lifetimes, leading to enhancement in the photoluminescence from P3HT. We propose that the differences in the lifetime of transferred charges arises from the crystallinity of the conjugated polymer films.

Mixed-dimensional heterojunctions (MDHJs) combine the characteristics of component materials such as the discrete orbital energies of zero-dimensional (0D) molecules and the extended band structure of two-dimensional (2D) semiconductors. While MDHJs have shown promise for optoelectronic applications – such as photodetectors because their performance exceeds that of the individual components – increased understanding of fundamental mechanisms and time scales of charge separation at the interface would drive improved design of these systems. Here, comparison between two different type-II organic/2D MDHJs comprised of copper (Cu) and free base (H₂) phthalocyanine (Pc) and monolayer MoS₂ elucidates the influence of interfacial morphology on charge separation lifetimes in these systems. In particular, time-resolved optical spectroscopy reveals two ultrafast charge transfer processes from selective excitation of each component material: sub-picosecond photoinduced hole-transfer and sub-320 fs photoinduced electron-transfer processes at the interfaces of CuPc/MoS₂ and H₂Pc/MoS₂ MDHJs. In CuPc/MoS₂ heterojunctions, charge separation lasts as long as 70 ns, which is a factor of 17 longer than that in H₂Pc/MoS₂ heterojunctions and a factor of 40 longer than that in previously reported transition-metal dichalcogenide-based heterojunctions. Polarized Raman spectroscopy shows that preservation of the charge-separated state is attributed to the face-on orientation of CuPc on the MoS₂ surface, which templates stacking of CuPc molecules and facilitates hole migration away from the interface, whereas H₂Pc molecules adopt a mixed edge-on and face-on orientation. This work highlights the role of molecular structure in determining the interfacial geometry and, ultimately, charge-transfer dynamics in 0D/2D heterojunctions and suggests principles for the rational design of other organic/2D MDHJs.

Atomically-thin layered semiconductors as exemplified by few-layer MoS₂ and WSe₂ are an emerging class of materials with their potential use for semiconductor devices [1, 2]. In this context, dominant carrier distribution on these materials and devices is one of the key electronic properties for understanding their electrical characteristics and improving device performance. In fact, recently, an unintentional electron doping effect in MoS₂ has been reported, which suggests the difficulty in fabricating intrinsic and p-type atomically-thin MoS₂ [3]. This effect is attributed to surface electron accumulation due to the formation of sulfur vacancies on the surface of MoS₂ and therefore becomes dominant in atomically-thin layers due to their high surface-to-volume ratio [3]. In order to investigate the transition of dominant carrier polarity on atomically-thin MoS₂ layers, here we perform the nanoscale carrier distribution imaging on atomically-thin natural and Nb-doped MoS₂ including single layer areas. By using a microwave-based scanning probe microscopy method called scanning nonlinear dielectric microscopy (SNDM) [4], we actually visualize unintentional n-type transition confined on single layer Nb-doped MoS₂, even though Nb-doped MoS₂ is nominally expected to be p-type in contrast to natural MoS₂ with n-type dominant carriers.
SNDM allows nanoscale dominant carrier distribution imaging through the detection of differential capacitance induced by small ac-bias voltage (so-called dC/dV imaging). As the polarity of dC/dV is inverted depending on the polarity of dominant carriers, we can identify n- or p-type on the local area below the tip. We have already succeeded in the detection of single layer MoS₂ [5], because SNDM is exceptionally sensitive to tiny dC/dV. In this study, we used novel intermittent contact SNDM with boxcar integration, which will be described elsewhere [6], to obtain higher S/N ratio than that in the conventional intermittent contact SNDM. The samples were natural and Nb-doped MoS₂ on SiO₂/Si substrates prepared by the mechanical exfoliation with so-called Scotch tape method. The measurement was performed in air at room temperature using a commercial scanning probe microscopy system (Bruker, Icon) integrated with a lab-made SNDM setup. As expected, for natural MoS₂, n-type contrast were seen even on few-layer MoS₂ including a single layer area. In contrast, we found that single layer Nb-doped MoS₂ showed unintentional n-type contrast, while multilayer areas were seen p-type. This unintentional electron doping effect was reproducibly observed on the single layer areas of other Nb-doped MoS₂ samples. In addition, unlike the multilayer areas, those with a few to several layer number often showed negligible dC/dV signal level. These results are reasonable if considering the counter doping effect depending on the surface-to-volume (S/V) ratio discussed in Ref. [5]. The effect does not manifest for multilayers with a low S/V ratio, but, for few-layers, it is balanced with Nb-doping, and then overcompensates it in single layer areas with the highest S/V ratio. This implies the difficulty in p-type conduction on our as-prepared MoS₂ samples. The results here indicate that SNDM gives direct information on nanoscale carrier distribution on the areas with different stacking layers, which will provide a clue to controlling carrier types and doping levels in atomically-thin layered semiconductors.
References:
[1] B. Radisavljevic et al., Nature Nanotechno. 6, 147 (2011).
[2] K. F. Mak, C. Lee, J. Hone, J. Shan, and T. F. Heinz, Phys. Rev. Lett. 105, 136805 (2010).
[3] M. D. Siao et al., Nature Commun. 9, 1442 (2018).
[4] Y. Cho, A. Kirihara, and T. Saeki, Rev. Sci. Instrum. 6, 2297 (1996).
[5] K. Yamasue and Y. Cho, Appl. Phys. Lett. 112, 243102 (2018).
[6] K. Yamasue and Y. Cho, (in preparation).

Van der Waals layered materials, such as transition metal dichalcogenides (TMDs), are an exciting class of materials with weak interlayer bonding which enables one to create so-called van der Waals heterostructures (vdWH). One promising attribute of vdWH is control over the twist angle between layers, which leads to the formation of moiré patterns that enable tuning of vdWH properties. In the case of TMD vdWH, moiré patterns have been used to explain unique optoelectronic behavior of these structures. However, the moiré theory used to explain the behavior of TMD vdWH have only considered the constituent layers as rigid lattices and have not allowed for atomic-level reconstruction, i.e. rearrangement of the atoms with the layers. The existence of atomic-level reconstruction would have a significant impact on the electronic and optoelectronic properties and would fundamentally change our theoretical understanding of these systems. Here, we present experimental proof of atomic-level reconstruction in twisted TMD vdWH, which is a significant departure from the rigid-lattice moiré theory. Using conductive atomic force microscopy (CAFM), we probe the nanometer-scale electronic behavior of twisted MoSe₂/WSe₂ and MoS₂/WS₂ heterostructures and show that, for twist angles around 1° or less, large domains of constant electronic behavior form rather than a continuously varying pattern of electronic behavior, as expected from conventional moiré theory. We find that the stacking arrangement (H-type or R-type) has a profound impact on the reconstruction behavior. We use transmission electron microscopy to corroborate the presence of atomic reconstruction, to demonstrate commensurate stacking within the reconstructed domains, and to demonstrate a transition from atomic reconstruction at small twist angles (≤ 1°) to a rigid-lattice moiré pattern for larger angles (≥ 3°). We use density functional theory (DFT) calculations to predict the low-energy stacking configurations of the heterostructures and to calculate the band structures of the low-energy stacking configurations. Agreement between CAFM and DFT enables us to assign stacking arrangements to the different reconstructed domains. Finally, we show that the difference in calculated bandgaps between heterostructures with H-type and R-type stacking agree with photoluminescence measurements of reconstructed heterostructures of both types. These results provide fundamental insight into the behavior of this exciting class of semiconductor heterostructure.

Following the isolation of individual graphene layers, several other two-dimensional (2D) materials have been discovered, including boron-nitride (BN), phosphorene, and transition metal dichalcogenides (TMDs) [1]. These 2D materials lack dangling bonds allowing for their seamless stacking to create multilayered heterostructures preserving their sharp interfaces [2]. The physical properties of these multilayer structures, including out-of-plane tunneling rates, may be tailored via composition and stacking order. As thousands of 2D materials have been isolated or predicted to be stable [3], the number of heterostructures that can be formed grows very rapidly. Here we develop methodology to characterize carrier transport in heterostructures formed with 2D materials. Our approach efficiently combines high-throughput first principles calculations and ballistic quantum transport. Our workflow generates vertical heterostructures with low-strain epitaxial mismatches between layers. Their geometries and electronic properties are obtained via first principles calculations within the density functional theory. Ballistic transport methodology based on the Landauer formalism employs representations of the electronic structure formed by projections of pseudo-atomic orbitals for efficient computations [4]. In particular, we analyze the modulation of the tunneling current as a function of strain perpendicular to the basal plane [5,6] for a variety of heterostructures based on graphene electrodes, and boron-nitride or transition metal dichalcogenides as tunneling barriers. Results are rationalized in terms of composition, stacking order and orientations. For the same number of layers, BN junctions exhibit larger tunneling currents than TMDs as its smaller thickness counterweights its larger band gap. We find, however, that the interlayer tunneling current of TMDs are more susceptible to mechanical strain than BN ones. We also compare results for the case of bulk electrodes and discuss virtues and limitations of approximations employed in this description. The results of this work may prove useful to the study of novel physical phenomena such as charge and spin transport in tunneling heterostructures based on 2D materials. This work has been partially supported by the NSF DMR-1848344.

[1] Novoselov et al., PNAS 102, 10451 (2005)
[2] Geim and Grigorieva, Nature 499, 419 (2013)
[3] Mounet et al., Nature Nanotechnology 13, 246 (2018)
[4] M. Buongiorno-Nardelli et al, Comp. Mat. Sci. 136, 76 (2017)
[5] Nayak et al., Nature Communications 5, 3731 (2014)
[6] Fu et al., Appl. Phys. Lett. 103, 183105 (2013)

Ti₃C₂ MXene, a new class of two-dimensional transition metal carbides has attracted much attention as a sensing material due to its high surface-to-volume ratios and excellent electrical conductivity. Furthermore, the presence of surface terminated functional group such as oxide (-O) and hydroxyl (-OH) onto the Ti₃C₂ MXene provides reaction sites for water absorption. Recently, many studies have been investigated to understand the charge transfer mechanism when water molecules were sorption in MXene surface or intercalation between the MXene nanosheets for improving response speed and sensitivity. However, the electrical response to water vapor absorption that determines the performance of the humidity sensor can still be limited due to the high electrical conductivity of MXene. Therefore, it is required to study the novel MXene structure which can improve the sensor performance by maximizing the electrical change while the water vapor or moisture is adsorbed on the film surface.
In this work, we synthesized TiO₂ nanowires/Ti₃C₂ MXene composites by facile solution process manipulating of concentration of NaOH solution. As a result, the density of TiO₂ nanowires increased in proportion to the concentration of NaOH solution and formed a network on the surface and interlayers of Ti₃C₂ MXene. The sensing performance of TiO₂ nanowires (NWs)/Ti₃C₂ MXene composites based humidity sensor was evaluated under relative humidity condition from 20 to 80 %. As a result, it showed a high sensitivity and fast response compared with pristine Ti₃C₂ MXene-based sensor. In addition, piezoresistive pressure sensor was fabricated by coating TiO₂ NWs/ Ti₃C₂ MXene on cotton fabric. Thereafter, the performance of the device evaluated over a wide pressure range of 0.33 to 63.1 kPa. As a result, our sensor showed high sensitivity of 2.5 kPa^-1 with excellent linearity. In addition, it exhibited an ultrafast response speed with a rise time of ~13 ms and decay time of ~17 ms. This study opens up a new avenue of two-dimensional layered hybrid materials for applications of advanced contact and non-contact electronic devices.

Acknowledgement
This work was partly supported by Electronics and Telecommunications Research Institute (ETRI) grant (19ZB1100) funded by the Korean government and Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20181510102340, Development of a real-time detection system for unidentified RCS leakage less than 0.5 gpm).

Recently, molybdenum disulphide (MoS₂) has attracted great attention for diverse applications due to its material properties. It is thought that these material properties soon will make MoS₂ attractive for the development of radiation-hard electronics. Therefore, for the first time, we studied the effect of gamma-ray (γ-ray) irradiation on the crystal quality of a few-layered MoS₂ by using Raman and X-ray photoelectron spectroscopy techniques. The γ-ray irradiation dose of 120Mrad was applied to 3.2, 4.5, and 5.2 nm thick MoS2 films which were grown on Si by using a two-step synthesis method (sputtering of Mo, followed by sulphurisation). Before γ-ray irradiation, three active Raman modes (E_1g A_1g) was detected. After γ-ray irradiation, the Raman peaks attributed to the E_1g and A_1g modes almost disappeared. The dramatic decrease in the S2p peak accompanied by an increase in the O1s peak and the change in the Mo3d3/2 binding energy from what is known for stoichiometric MoS₂ to that for native Mo oxide was observed in the XPS spectra. In conclusion, dramatic chemical changes in the MoS2 films after irradiation is believed to be related to the fact that S vacancies are formed due to γ-ray irradiation, which subsequently transforms MoS₂ to a native MoO_x.

The discovery of exotic electronic properties of graphene has led to extensive research efforts to find new functional two-dimensional (2D) materials. For example, transition metal dichalcogenide (TMD) compounds are an excellent candidate for various electronic devices due to their thickness-tunable band gap. However, their relative low carrier mobility limits their applications. Recently, honeycomb-like monolayer materials composed of main group elements such as GeP₃ and antimonene were reported to exhibit high carrier mobility. However, this property is only observable in epitaxially grown samples on specific substrates. Here, we present a new structure-type, metastable 2D ZnSb which exhibits remarkably high carrier mobility. The structure features unique two atom-thick Zn-Sb honeycomb layers stacked along the out-of-plane direction. This compound can be only achievable by topotactic chemical reaction we developed. Temperature-dependent electrical resistivity and Hall effect measurements on single crystalline 2D ZnSb show p-type metallic conduction with hole mobility of ~400 cm²V^-1s^-1 at room temperature. A new 2D ZnSb is readily cleavable to form a few layer-thick nanosheets by various irradiation techniques. This new phase shows completely different crystal structure and physical properties from a well-known three-dimensional semiconducting polymorph.

We present a new, highly sensitive biosensing approach using a graphene-based barristor [1]. The sensor is formed by interfacing graphene with semiconductor. The resulting Schottky barrier (between graphene and semiconductor) is highly susceptible to the device surface condition; binding of target biological molecules can turn on the device with very large signal gain. This mechanism makes the barristor far superior to conventional graphene FET sensors that are based on the charge-density modulation[2]. To prove the concept, we fabricated n-type barristor sensors and used them to detect DNA molecules. With target DNA binding, the height of the Schottky barrier increases, leading to the decrease in device currents. Even at very low DNA concentrations (~1 aM), we observed more than 40% changes in the device conductance. The technology may open up a new material and device concept for ultrasensitive biosensing.

Two-dimensional (2D) ZnO nanostructures are becoming more popular due to their properties such as nanometer-scale thickness, high surface to volume ratio, and good mechanical stability ¹. However, current strategies are usually limited to low production rate, complicated method, and high cost. Here in this study, we report a high-yield method to synthesize large-area ZnO 2D nanosheets by utilizing recyclable anodic aluminum oxide (AAO) template. ZnO nanosheets were synthesized by microwave-assisted hydrothermal synthesis method without using seeding layer, with a large lateral size and nanometer thickness. ZnO nanosheets could be obtained by a sonication process, with a yield of 3mg/cm² in 3 hours. XRD analysis indicated the formation of ZnO nanosheets exposed with (100) facets and the successful doping of Al. During electron microscopy imaging, an ultrathin morphology was observed with good wrinkling flexibility. Besides, due to the intactness of the template after nanosheet transferring using polymer, the AAO template can be cyclically utilized as the substrate for repetitive growth of ZnO nanosheets, which further reduces the cost and waste. Various application demonstrations are conducted based on this ZnO nanosheets, including gas sensor, thermochemical CO₂ hydrogenation, and photocatalysis. Owing to its ultrathin 2D morphology with abundant amounts of active sites and defects, a higher sensitivity and a short response time toward ppb–level gas detection may be obtained, and a higher conversion efficiency of CO₂ hydrogenation is achieved ^2-3. High binding energies between gas molecules and oxygen vacancies existing in the exposed ZnO (100) facets lead to large surface reconstructions, which make a contribution to the gas sensing properties and CO₂ hydrogenation. Furthermore, the highly exposed surface of nanosheets also facilitates gas sensing and catalysis performance. The demonstrated method here enables a new path to the high-yield synthesis of 2D ZnO for applications in energy-related field and beyond.
Reference
1. Gaddam, Venkateswarlu, et al. "Morphology controlled synthesis of Al doped ZnO nanosheets on Al alloy substrate by low-temperature solution growth method." RSC Advances5.18 (2015): 13519-13524.
2. Yuan, Hongye, et al. "ZnO Nanosheets Abundant in Oxygen Vacancies Derived from Metal–Organic Frameworks for ppb–Level Gas Sensing." Advanced Materials 31.11 (2019): 1807161.
3. Geng, Zhigang, et al. "Oxygen vacancies in ZnO nanosheets enhance CO2 electrochemical reduction to CO." Angewandte Chemie International Edition 57.21 (2018): 6054-6059.

Since the discovery of graphene in 2004, two dimensional (2D) materials have led an intensive attention because of their unique physics and new device architectures. Nevertheless, there was limitation over having large-scale, monolayer 2D materials because it is extremely difficult to control the kinetic of 2D materials during the growth. Especially, it is mandatory to have monolayer of transition dichalcogenides (TMDCs) for optoelectronic applications as multilayer 2D materials have indirect band gap which involving phonon loss. In addition, since 2D materials themselves do not have sufficient external quantum efficiency (EQE), we cannot expect highly efficient luminescence from 2D material-based optoelectronics. Thus, it has been required to develop an alternative approach for 2D materials-based optoelectronics.
Here, we propose an interesting approach for highly efficient luminescence from large scale TMDCs by effectively engineering optical loss. Last year, we reported a new approach to make large scale, monolayer of the TMDC, called layer-resolved splitting. This approach enables us to have large-scale monolayer-by-monolayer stacking with pristine interface as it does not accompany any polymer residue and chemical contamination at the interface. Therefore, it minimizes the optical loss through interface interference. In addition, we successfully engineer additional optical loss by incorporating distributed bragg reflector (DBR) structures with TMDCs since DBR has broad reflection band and phase shift of DBR is close to zero. Thus, we can observe the improvement of photoluminescence in TMDC/DBR structure using layer-resolved splitting process. We strongly believe that this process will lead a new opportunity for highly efficient 2D material-based LED optoelectronic applications.

Crystals with low symmetry and low dimensionality have drawn tremendous fundamental and practical interest, since such a strongly quantum confined anisotropic system gives rise to many novel and complicated many-body interactions. Black phosphorus (BP), a two-dimensional elemental semiconductor, is a representative member of such low-symmetric nanomaterials. With the highly anisotropic electronic structure and phonon dispersions, black phosphorus has been an excellent platform for understanding the rule of symmetry related multi-particle interactions. Here, we investigated the electron-phonon coupling in few-layer and bulk BP using polarization-dependent resonance Raman spectroscopy with 16 laser lines. The in-plane A_g² mode, which shows larger intrinsic resonance effect than the out-of-plane A_g¹ mode when excitation polarization is along the armchair direction, demonstrating symmetry-dependent electron-phonon coupling effect in BP. This understanding is further supported by the quantum perturbation theory and first-principles calculations on the anisotropic electron distributions. Moreover, the comparison of intrinsic resonance effects along two crystalline orientations allow us to resolve the existing controversies on the physical origin of Raman anomaly in BP and its dependence on excitation energy, sample thickness, phonon modes, and crystalline orientation.

Graphene has been demonstrated to be promising for spintronic devices due to its intrinsically low spin orbit interaction¹. Although long spin relaxation lengths (> 2µm) have been experimentally achieved with graphene, the degree of polarized carrier injection still low (< 10%) due to scattering at the contact/graphene interface². Ferromagnetic electrodes such as cobalt are most commonly used for injecting spin-polarized electrons into non-ferromagnetic graphene. Recently, atomic resolution scanning transmission electron microscope studies on metal/2D material interfaces have revealed that defects are created during metal deposition^3,4. This has led the studies of inserting tunnel barriers such as h-BN⁵, MgO⁶, and TiO₂⁷ between Co and graphene layers for improving the spin injection efficiency. Here, we report results from our recent work on realizing ultra-clean van der Waals contacts using Indium electrodes for spintronics. Specifically, we use 3 – 10 nm indium to achieve defect-free contact, then deposit ferromagnetic Co layers on top. Annealing of the contacts leads to ferromagnetic In/Co alloy with defect free interface with graphene that can be used to inject spin polarized carriers with high efficiency.

References:
1. Gmitra, M., Konschuh, S., Ertler, C., Ambrosch-Draxl, C. & Fabian, J. Band-structure topologies of graphene: Spin-orbit coupling effects from first principles. Phys. Rev. B 80, 235431 (2009).
2. Tombros, N., Jozsa, C., Popinciuc, M., Jonkman, H. T. & van Wees, B. J. Electronic spin transport and spin precession in single graphene layers at room temperature. Nature 448, 571–574 (2007).
3. Wang, Y. et al. Van der Waals contacts between three-dimensional metals and two-dimensional semiconductors. Nature 568, 70 (2019).
4. Liu, Y. et al. Approaching the Schottky–Mott limit in van der Waals metal–semiconductor junctions. Nature 557, 696 (2018).
5. Gurram, M. et al. Spin transport in fully hexagonal boron nitride encapsulated graphene. Phys. Rev. B 93, 115441 (2016).
6. Han, W. et al. Tunneling Spin Injection into Single Layer Graphene. Phys. Rev. Lett. 105, 167202 (2010).
7. Dankert, A. & Dash, S. P. Electrical gate control of spin current in van der Waals heterostructures at room temperature. Nat. Commun. 8, 16093 (2017).

Memristors is are a great promising candidate for a next-generation nonvolatile memory due to its their small footprint, low power consumption, thus, suitable for handling tremendous data set such as learning neural-networks. At the early stage, memristor devices was were fabricated with titanium dioxide (TiO₂) controlling movement of oxygen vacancies, nowadays, for improved performance and various applications, several materials such as organic polymer and ferroelectric thin film haven been widely studied s been suggested as a switching medium. Recently, memristors utilizing two-dimensional (2D) layered material based memristors, which are atomically thin, flexible and easily heterogeneously integrated, for example, MoS₂, MoTe₂, h-BN have also shown their feasibility as they are atomically thin, flexible and easily heterogenous integrated, successfully making 2D layered material them as a promising switching medium for atomically thin, flexible, and vertically stackable memristor.
In this study, we propose a memristor device architecture consisting of report WSe₂/h-BN heterostructure showing low leakage current, pristine interface, and, pristine interface, memristor devices. stable filament formation in switching mediums, which are key factors for stable memristor operation. Our main approach for them is engineering material combination with pristine interface. As h-BN itself has low leakage current properties and WSe₂ itself has stable filament formation, we have found out that WSe₂/h-BN heterostructures have much large memory window (on/off ratio of ~10⁴) and stable switching operation (over 100 times of repeatable cycling). Heterogenous material combination often generates interfacial issues, however, heterostructure on large area was successfully fabricated exploiting advantages of each materials; low leakage current of h-BN and repeatable switching of WSe₂. These improved characteristics were also estimated with several devices fabricated in same process bawe could achieve high-quality interface by using our quasi-dry transfer process. We recently developed a quasi-dry transfer, allowing us large-scale 2D materials with pristine interface because the quasi-dry transfer does not accompany any chemical contamination at the interface during the process. Thus, we successfully engineer material combination with maintaining the properties of each material for stable switching behavior. We strongly believe that WSe₂/h-BN heterostructures by the quasi-dry transfer has a strong potential for stable memristors.

Neuromorphic computing supported by synaptic devices, which mimics the structure and functions of neural networks at the hardware-level, is one of emerging candidates for the next generation cognitive computing. Although synaptic devices in artificial neural networks (ANN) feature small device footprints and reduced energy consumption, it has been challenging to find an optimal synaptic device for large-scale neuromorphic computing due to stochastic and unpredictable nature in conductance update. Also, for the synaptic device to be used as analog synaptic weight facilitating a simple multiply-accumulate (MAC) operation, it is required to have two synaptic devices to represent one weight since synaptic devices merely possesses positive conductance value, which can be expressed as weight = G⁺ - G^-.
Here, we propose a 2D/Organic heterostructure synaptic device coupled with an optical source, which enables to solely represent each weight in neural networks without having two synaptic devices. Our key approach for the function is controlling reverse current by engineering i) symmetry of Metal-WSe₂ contact and ii) input pulse characteristics and postsynaptic voltages. We have figured out that amount of photo-generated carrier diffusion is proportional to the symmetry of Metal-WSe₂ contact. Less symmetric contact causes more asymmetric photo-generated carrier diffusion, making the reverse current and negative conductance possible. Also, direction of the diffusion from WSe₂ to Metal is opposite to the direction of carrier injection from metal to WSe₂, thus, balancing of them offers a way to control the polarity of conductance. By changing input pulses and postsynaptic voltage, we successfully balanced two types of carrier flows and gradually changed conductance states from negative to positive, which can be directly used as a weight without constituting conventional weight = G⁺ - G^- cell. We believe that optically coupled 2D/organic synaptic device will further revolutionize the current neuromorphic computing system in the way that it cuts the size of crossbar circuit in half and thus provides an energy-efficient platform for post von Neumann Computing.

MoTe₂ belongs to transition metal dichalcogenides - materials, which have been intensively studied for the last few years. They provide a rare occasion to test relativistic physics with low energy excitations in condensed matter experiments. The unusual properties of the charge carriers, which are chiral, helical and have a linear E(k) dependence lead to many interesting physical phenomena such as chiral anomaly [1], extremely large positive magnetoresistance [2, 3, 4], or the planar Hall effect [1, 5]. The most commonly used methods to obtain MoTe₂ layers, beside exfoliation from natural crystals [6] are the flux method [3, 4, 7] and chemical vapour transport [8]. Currently, Molecular Beam Epitaxy (MBE) technique comes into play [9, 10, 11, 12] giving the perspectives to combine different phases of MoTe₂ and to grow heterostructures and hybrid structures.
In this paper we present the results of structural, optical and transport investigations of MoTe₂ thin layers. The samples were prepared on Al₂O₃ and SI-GaAs [111B] substrates which both, at the appropriate growth temperature, allow us to obtain 2H and 1T’ polytypes of MoTe₂ (as evidenced with Raman scattering) as well as other phases, e.g. Mo₆Te_6. Controlling the growth conditions allows to modify sample morphologies from regular 2D planes, through 3D precipitations to nanowires as studied in-situ with RHEED images, and ex-situ by TEM and SEM microscopy. Transport measurement were performed on large, millimeter-size Hall bars with metal (Ti/Au or In) contacts deposited via shadow masks. The resistance of the samples was measured in the temperature range 300 K – 1.5 K revealing either an increase of the resistivity for semiconducting samples or a very weak temperature dependence, showing significant disorder for the samples with the metallic phases. In the latter case, for high concentation (n ~ 1●10¹⁶ cm^-2) several monolayer sample, the low value of carrier mobility (2 cm²/Vs) and a linear magnetoresistance up to 12 T revealed sample inhomogeneity. Furthermore, we have observed that the poor stability of the MoTe₂ layers can be improved by different capping layers.

[1] S. Nandy et al., Phys. Rev. Lett. 119, 176804 (2017)
[2] P. S. Alekseev et al. Phys.Rev. B97, 8, 085109 (2018)
[3] F. C. Chen et al, Phys. Rev. B 94, 235154 (2016)
[4] X. Luo et al., Appl. Phys. Lett. 109, 102601 (2016)
[5] A. A. Burkov et al., Phys. Rev. B 96, 041110(R) (2017)
[6] I. G. Lezama et al., Crystals. 2D Mater. 1 (2), 021002 (2014)
[7] S. Thirupathaiah et al., Phys. Rev. B 95, 241105(R) (2017)
[8] K. Deng et al., Nature Physics 12, 1105 (2016)
[9] S. Tang et al., APL Materials 6, 026601 (2018)
[10] A. Roy et al., ACS Appl. Mater. Interfaces 8 (11) (2016)
[11] Y. Yu et al., Carbon 115 (2017)
[12] S. Vishwanath et al., Journal of Crystal Growth 482 (2017)

* corresponding author: mgryglas@fuw.edu.pl

This research was supported by NCN Opus 2017/27/B/ST5/02284