Symposium F.EL04—Beyond Graphene 2D Materials—Synthesis, Properties and Device Applications

Inspired by the nano/microscale hierarchical structure, topological network and precise inorganic/organic interfaces, we fabricated artificial ‘nacre-like’ papers and aerogels based on noncovalent functionalized boron nitride nanosheets (NF-BNNSs) and varying poly(vinyl alcohol) (PVA) concentration via vacuum-assisted self-assembly and freeze-drying (or ice-templating) techniques. The electron microscopy showed an ordered ‘brick-and-mortar’ arrangement of NF-BNNSs and PVA paper and aerogels, in which the long-chain PVA molecules act as the bridge to link NF-BNNSs via hydrogen bonds. Likewise, due to the need for heat removal in modern electronic devices, polymer composites with high thermal conductivity have drawn much attention. However, the traditional method to enhance polymers' thermal conductivity by random addition of nanofillers usually creates composites with not only limited thermal conductivity but also other detrimental effects due to a large amount of fillers required. Here, novel polymer composites are prepared by first constructing 3D boron nitride nanosheets (3D-BNNS) network using an ice-templated approach with varying BNNS sheets concentration and then infiltrating them with epoxy matrix. The thermal and mechanical properties of resulting free-standing nacre-like papers, aerogels and composites are investigated. They elucidated high glass transition temperature, low interfacial thermal resistance, or high thermal conductivity with relatively low BNNS loading thus revealing their superiority as flexible substrates. These results demonstrate that this approach opens a new avenue for the design and preparation of polymer composites with high thermal conductivity useful in advanced electronic packaging techniques, namely, thermal interface materials, underfill materials, and molding compounds.

Nanofiltraton has gained much attention in recent years due to its possible applications as a highly efficient and low-energy usage separation and purification technology[1,2]. Moreover, nanofiltration has shown to be potentially superior and more cost-effective than other filtration techniques. Nanofiltration was initially adapted for wastewater treatment, but has been implemented by other industries, such as the pharmaceutical and the food industries, where the standard for high quality products and stringent safety requirements necessitate further and more specialised separation approaches[1,3].
The main aim of this project was to use inorganic 2D nanomaterials in order to develop new membranes for nanofiltration applications, in particular, boron nitride (BN). In this project, BN was exfoliated by liquid assisted exfoliation and the membranes were prepared by vacuum filtration using hydrophilic polytretrafluoroethylene as template. Boron nitride was also partially oxidised before exfoliation proving new chemically functionalised BN materials. All membranes were tested using different dyes and calculating the rejection of each one. Almost 100% rejection rates have been achieved for optimised BN membranes demonstrating the superb nanofiltration performance for selected dyes. The new membranes are to be used for the separation of D- and L- lactic acid and various mono- and disaccharides from agricultural residues.

References:
[1] B. Van der Bruggen, M. Manttari, M. Nystrom, Drawbacks of applying nanofiltration and how to avoid them: A review, Sep. Purif. Technol. 63 (2008) 251–263. doi:10.1016/j.seppur.2008.05.010.
[2] P. Eriksson, Nanofiltration extends the range of membrane filtration, Environ. Prog. 7 (1988) 58–62. doi:10.1002/ep.3300070116.
[3] N. García Doménech, F. Purcell-Milton, Y.K. Gun’ko, Recent progress and future prospects in development of advanced materials for nanofiltration, Mater. Today Commun. 23 (2020) 100888. doi:https://doi.org/10.1016/j.mtcomm.2019.100888.

Van der Waals heterostructures comprised of two-dimensional (2D) materials offer a platform to obtain materials by design with unique electronic properties. Until recently, these van der Waals heterostructures (vdWH) materials were synthesized in the laboratory by vertically stacking different atomically thin materials, which often leads to stacks with the presence of fabrication defects and air contaminants between layers. Franckeite is a naturally occurring vdWH comprised of two distinct alternately stacked semiconducting layers. Because it is naturally heterostructured, it enables the study of a complex layered system without the presence of fabrication defects, and where the crystal orientation between layers has been preserved. Unlike other layered sulfide-based 2D materials, Franckeite is a grained-textured layered solid composed of few-millimeters flakes in random orientation. For this reason, mechanical exfoliation of thin flakes with uniform and large surface area for optical characterization and transfer techniques is especially challenging, leading to a knowledge gap of the fundamental properties of Franckeite. Raman spectroscopy is a widely and powerful method to characterize and study the fundamental properties of molecules in 2D crystals. All of the power that Raman spectroscopy provides in characterizing 2D materials stems from a large and well-established literature on the Raman signature of these crystals, which is lacking for Franckeite. In this work, we performed an extensive micro-Raman spectroscopy study of Franckeite. At room temperature, Raman spectroscopy data were acquired for a large sample size of flakes with thickness ranging from 6 to 100 nm prepared by mechanical exfoliation on SiO₂ substrates. The Raman signature of Franckeite is detected for flakes thicker than 10 nm. We observed that the position of all Raman peaks is independent of sample thickness. As expected, signal intensity is enhanced for flakes prepared on ultra-flat gold (Au) substrates (root-mean-square (rms) surface roughness < 0.2 nm), but different Raman modes undergo different rates of enhancements when comparing flakes of the same thickness on SiO₂ and Au. Raman data was acquired for excitation sources of both 532 and 633 nm of wavelength. Comparison data plots indicate resonance-enhanced Raman scattering with 633 nm of excitation, confirmed by additional absorption measurements. Temperature-dependent data was acquired from room temperature to 225 °C. All Raman modes present redshift with increasing temperature, suggesting that the chemical bond length of Franckeite increases with increasing temperature, and therefore the bond’s vibrational force constant decreases. Moreover, we observed new low-frequency shearing and breathing modes in the Raman spectra. These low-frequency vibrational modes of Franckeite can provide a broader understanding of interlayer vibrational modes in 2D crystals, and especially in van der Waals heterostructures.
References
1. Molina-Mendoza, A. J. et al. Franckeite as a naturally occurring van der Waals heterostructure. Nat. Commun. 8, (2017).
2. Velický, M. et al. Exfoliation of natural van der Waals heterostructures to a single unit cell thickness. Nat. Commun. 8, 1–11 (2017).
3. Ray, K. et al. Photoresponse of Natural van der Waals Heterostructures. ACS Nano 11, 6024–6030 (2017).
4. Liang, L. et al. Low-Frequency Shear and Layer-Breathing Modes in Raman Scattering of Two-Dimensional Materials. ACS Nano 11, 11777–11802 (2017).

Atomically thin two-dimensional (2D) molybdenum disulfide (MoS₂) is highly attractive due to its unique properties and promising applications in nanoscale electronics. Characterization of thermal expansion coefficient (TEC) of 2D MoS₂ is ubiquitous for reducing the thermal mismatch and improving the thermal management of MoS₂-based devices. However, measuring the TEC of 2D MoS₂ is particularly challenging owing to its nanoscale thickness, microscale area and optical transparency properties. As a result, the thermal expansion on 2D MoS₂ still extensively relies on simulation. In this work, we report a pure experimental approach to characterize the TEC of monolayer MoS₂ using micro-Raman spectroscopy. We analyzed the stress and temperature dependent phonon frequency of monolayer MoS₂ using the perturbation theory and symmetry analysis. To decouple the effect of thermal stress on phonon frequency shift, monolayer MoS₂ flakes were transferred on fused silica, pure copper and micro-hole patterned thermal oxide substrates. We characterized the temperature coefficients on three substrates through the temperature-dependent Raman measurement. The in-plane TEC of monolayer MoS₂ was then extracted by comparing these coefficients. The resulting in-plane TECs were (7.6±0.9)×10^-6 1/K and (7.4±0.5)×10^-6 1/K, which were independently obtained from two different phonon modes of MoS₂ and in reasonable agreement with previous first-principle reported values. This work provides accurate measurement on the TEC of monolayer MoS₂. Additionally, the general experimental approach can be widely applied to study many other 2D materials or thin films in a simple way.

Hexagonal boron nitride (h-BN) is an ultrawide (~ 6.0 eV) semiconductor that has potential applications for developing electronic, optoelectronic and nanophotonic devices including source of ultraviolet emitters. A key issue in these applications is understanding the effect of impurities, especially carbon and oxygen on its properties. Deep UV photoluminescence spectroscopy was employed for optical characterization of irregular-shaped h-BN crystal powders. Fourth harmonic laser (195 nm) generated from the Ti: sapphire laser was used for optical excitation in the experiments. The luminescence properties are investigated by annealing the sample at different temperatures under different conditions. The deep UV photoluminescence spectra from h-BN samples annealed at 700 ^oC in ambient air reveals a strong phonon-assisted band-edge emissions along with a sharp atomic-like emission line at 4.09 eV, and its phonon replicas at 3.89 and 3.69 eV. In addition to these peaks, sharp multi-phonon peaks are observed adjacent to them. The PL spectra of the sample shows the broadening of the main peaks as the annealing temperature exceeds 900 ^oC. We will present our findings of atomic-like features that could have potential applications as quantum sources in UV regions for quantum information technologies.

We synthesized single-crystal BiOI via vapor transport, subjected the resulting crystals to a series of surface treatments, and quantified the resulting chemical states and electronic properties. Vapor transport methods included both physical vapor transport from single-source BiOI, as well as chemical vapor transport from Bi₂O₃ + BiI₃ and from Bi + I₂ + Bi₂O₃. Surface treatments included tape cleaving, rinsing in water, sonication in acetone, an aqueous HF etch, and a sequential HF etch with subsequent sonication in acetone. X-ray diffraction, XRD, and X-ray photoelectron spectroscopy, XPS, probed the resulting bulk crystalline species and interfacial chemical states, respectively. In comparison with overlayer models of idealized oxide-terminated or iodide-terminated BiOI, angle-resolved XPS elucidated surface terminations as a function of each treatment. Ultraviolet photoelectron spectroscopy, UPS, established work-function and Fermi-level energies for each treatment. Data reveal that HF etching yields interfacial BiI₃ at BiOI steps that is subsequently removed with acetone sonication. UPS establishes n-type behavior for the vapor-transport-synthesized BiOI, and surface work function and Fermi level shifts for each chemical treatment under study. We discuss the implications for processing BiOI nanofilms for PV and photocatalysis applications.

A monolayer electrolyte, developed by our group, has been used to demonstrate a solid-state nonvolatile two-dimensional (2D) crystal memory based on an electric-double-layer (EDL) gated field effect transistor. (Nano Lett., 19, 8911, 2019) The monolayer electrolyte consists of cobalt crown ether phthalocyanine (CoCrPc), with four crown ethers that can solvate one lithium ion each. By simply drop-casting and annealing, CoCrPc forms a flat and ordered array on 2D crystals including graphene and WSe₂. The key mechanism of the nonvolatile memory is the toggling of lithium ions up and down through the cavity of crown ether (i.e., away from or nearby the channel). This toggling is accomplished by an applied electric field to overcome the diffusion barrier for ion movement. Our previous experimental work focused on one combination of materials: CoCrPc with 15-crown 5-ether (15C5), LiClO₄, and ethanol as the solvent for the salt. However, additional material combinations should result in varying diffusion barriers, thereby offering a path to tune the voltage required to toggle the ions. In this study, we investigate CoCrPc with 15C5 combined with two kinds of salts, LiCl and LiI, and three solvents: anhydrous acetone, acetonitrile, and nitromethane. LiCl and LiI were chosen to isolate the impact of anion identity (Cl vs. I vs. ClO₄) on the quality of the monolayer, and the three solvents were chosen for their reported tendency to promote ion-crown complexation. Monolayer quality is evaluated by measuring the thickness and surface roughness with atomic force microscopy (AFM) after deposition on 2D crystals in a water and oxygen-free glovebox. This study establishes the design rules for monolayer electrolyte deposition on 2D crystals, which can be used to engineer devices with multiple diffusion barriers in monolayer electrolyte random-access memory (MERAM). This work is supported by the NSF under Grant # NSF-DMR-EPM CAREER: 1847808.

Single element atomic monolayer honeycomb lattice films (also called “Xene” films) are a quickly-growing topic in condensed matter physics and materials science. In contrast to graphene, Xene films from higher Z elements such as bismuthene (from bismuth) and stanene (from tin) are predicted to exhibit quantum spin Hall effects at temperatures closer to room temperature due to their larger topological band gaps, making them more useful for potential applications in next generation electronics. Although these two Xene films have strong theoretical and computational evidence supporting their candidacy for higher temperature quantum spin Hall effects and topologically protected edge states, they are both very difficult to produce because hexagonal unit cells are unstable configurations for the crystallization of these elements. Here, we present our work on using molecular-beam epitaxy (MBE) to synthesize monolayers and thin layers of bismuthene.

The chromium-containing alkaline earth metal tetrasilicates, ACrSi₄O₁₀ (A= Ca, Sr, and Ba), are noteworthy as examples of rare, high spin Cr(II) in square-planar coordination complexes in chromium oxide-crystal chemistry. Though the synthesis of ACrSi₄O₁₀ has been known in the literature for several decades, much of the chemistry surrounding ACrSi₄O₁₀ formation remains unknown. In this contribution, we describe magnetic properties of all ACrSi₄O₁₀ phases prepared through improved synthetic routes as well as the subsequent nanostructuring of them into nanosheets by exfoliation. For flux-based routes, we utilize a metal envelope made from folded Cr foil to maintain a low oxygen fugacity in the reaction environment to avoid the formation of highly oxidized Cr species. We further extend the chemistry of BaCrSi₄O₁₀ to include mixed metal centers, specifically Cr(II) and Fe(II). The incorporation of mixed metal centers in ABSi₄O₁₀ phase is new territory in metal tetrasilicate chemistry, and this is the first example of transition metal substitution in any of the known ABSi₄O₁₀ systems (B = Cu, Fe, Cr). We find that the targeted stoichiometry of Ba(Cr,Fe)Si₄O₁₀ crystals yields an experimentally-determined composition with up to a ~40% incorporation of Fe(II), which appears to constitute the maximum incorporation of Fe(II) into BaCrSi₄O₁₀. Interestingly, Ba(Cr,Fe)Si₄O₁₀ crystals show a different magnetic infrastructure due to the interaction of antiferromagnetic Cr(II) and ferromagnetic Fe(II) centers in a single lattice. In this paper, we characterize the magnetic properties of Ba(Cr,Fe)Si₄O₁₀ crystals and nanosheets using Mössbauer and EPR spectroscopies.

Salt-assisted mechanical milling and chemical exfoliation in cosolvent were adopted to prepare the large-area MoS₂ nanosheets with a few layers in thickness. First, the MoS₂ powder was blend with salt, e.g., potassium chloride (KCl), calcium chloride (CaCl₂) or sodium carbonate (Na₂CO₃), and sent to a ball mill containing zirconia beads with 1 cm in diameter. The milling process was carried out at 300 rpm for 2 hrs so as to generate numerous dangling bonds at the edges of MoS₂. The dangling bonds would react with the salt to form the functionalized groups which may induce the electrical repulsion for the dispersion of MoS₂ in subsequent chemical exfoliation process. The functionalized MoS₂ powder was added in a centrifuge tube containing the cosolvent comprised of water and alcohol at the weight ratio of 1:1. After the ultrasonic vibration for 30 mins at room temperature, the MoS₂ nanosheets with thickness of 1-2 nm could be obtained via the centrifugation of solution. Most studies of chemical exfoliation reported the MoS₂ nanosheets with the lateral size smaller than 1 μm. However, our study observed the MoS₂ nanosheets which average area of 12×5.6 μm² using Na₂CO₃ and average area of 14×6 μm² using KCl. Presently, the MoS₂ FET devices containing the large-area MoS₂ nanosheets are prepared and their electrical properties are measured.

Since, the isolation of single layer graphene, a great attention has been paid to integrate distinct emerging beyond graphene 2D materials into van der Waals (vdW) heterostructures. The emerging 2D transition metal dichalcogenides (TMDCs), such as semiconducting (2H-phase) tungsten sulfide (WS₂) and their mixed dimensional heterojunctions have drawn most vibrant areas of nanoscience/photonic research because of fascinating optical and electrical properties in their low dimensional interface. [1-3] However, most of the preparation (chemical) methods do not preserve the semiconductive properties of WS₂. Here, we have prepared a novel mixed dimensional 2D semiconductor p-n heterojunction of few layer WS₂, which is exfoliated in the presence of an organic semiconductor (OS) and characterized by various experimental techniques. Our results revealed that the few hundred nanometer size, few layer (4-5 layers) thick WS₂ sheets are grafted with OS. Hexagonal pattern in HR-TEM image intimate the preservation of 2H-phase WS₂. Significant changed in the electrical behavior in the p-n nanoheterojunctions are presented. The insight spectroscopic analysis indicated the possible hybrid optoelectronic applications. Furthermore, the heterojunction dispersion WS₂ sheets are quite stable with prior little aggregation after 180 days.
Reference
[1] D. Jariwala, T.J. Marks and M. C. Hersam, Nat. Mater., 2017, 16, 170-181.
[2] A. S. Sarkar, S. K Pal, J. Phys. Chem. C, 2017, 121, 21945-21954.
[3] A. S. Sarkar, I. Konidakis, I. Demeridou, E. Serpetzoglou, G. Kioseoglou and E. Stratakis, 2020, https://arxiv.org/abs/2005.02852

Powder vaporization is a common method for the generation of large-area, single-crystal, two-dimensional molybdenum disulfide. While commonly employed as a growth method, the fundamental molecular mechanisms are not well understood. Recent ab initio analyses have shown that molybdenum oxysulfide rings play a key role in the sulfurization of molybdenum trioxide from elemental sulfur. Here, we present the utilization of molecular dynamics simulations with a reactive force field and ab initio calculations to elucidate the reaction pathway of sulfur with molybdenum trioxide. The molecular dynamics simulations demonstrated that for all sulfur allotropes the reaction pathway could be reduced to that of disulfur, trisulfur, or a combination of the two, and that molybdenum trioxide can catalyze the decomposition of larger sulfur allotropes. Ab initio calculations were used to illuminate the intermediates and transition states in the reaction pathways for disulfur and trisulfur. Analysis of the temperature dependence of the transition state energies show that the maximum reaction rates occur between 1000 and 1100 K, which corresponds with commonly reported experimental growth temperatures for molybdenum disulfide.

Two-dimensional (2D) nanomaterials, especially those with controlled size and shape, have attracted significant attention over the last decade, due to their enhanced performance in comparison to their bulk counterparts.^1–3 One such chemical compound is tin monoxide (SnO), whose layered crystal structure renders it amenable to the fabrication of 2D architectures.
SnO has demonstrated promising performance in many relevant applications, including thin-film transistors,^4–6photocatalysis,⁷ gas sensing⁸ and, primarily, energy storage - as next-generation battery anode materials.^9–11 However, the synthesis of SnO nanomaterials with controlled morphology still poses a considerable challenge in the field.^9,12–17 In this work, we provide a comprehensive study of the complex relationship between wet chemistry synthesis conditions and resulting nanoparticle morphology. The solvent nature is observed to strongly influence the kinetics of nucleation and crystal growth in solution, and thus the final morphological and electrochemical properties.
Extensive characterization of the precipitate nanomaterial has been performed, including scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HRTEM), selected area electron diffraction (SAED), x-ray diffraction (XRD) and thermogravimetric analysis (TGA), in order to comprehensively describe the solvent-morphology dependence. Furthermore, high-level electronic structure theory, including dispersion corrections to account for Van der Waal’s effects, were employed to augment our understanding of the underlying chemical mechanisms. Using supercell calculations, electronic vacuum alignment and surface energies were determined, allowing the prediction of the thermodynamically-favored crystal shape (Wulff construction) and surface-weighted work function, important parameters for battery performance. Finally, the synthesized nanomaterials were tested as battery materials, illustrating the impact of particle morphology on electrochemical performance.
Our results reveal promising pathways to the controlled synthesis of nanomaterials with tailored morphologies for enhanced performance in specific applications.



¹ S. Suresh, Nanoscience and Nanotechnology 3, 62 (2013).
² Y.W. Jun, J.S. Choi, and J. Cheon, ANGEW CHEM (INT ED ENGL) 45, 3414 (2006).
³ Z. Wu, S. Yang, and W. Wu, Nanoscale 8, 1237 (2016).
⁴ K.J. Saji, K. Tian, M. Snure, and A. Tiwari, Advanced Electronic Materials 2, 1 (2016).
⁵ Y. Ogo, H. Hiramatsu, K. Nomura, H. Yanagi, T. Kamiya, M. Kimura, M. Hirano, and H. Hosono, Physica Status Solidi (A) Applications and Materials Science 206, 2187 (2009).
⁶ S.S. Lin, Y.S. Tsai, and K.R. Bai, Applied Surface Science 380, 203 (2016).
⁷ L. Liang, Y. Sun, F. Lei, S. Gao, and Y. Xie, Journal of Materials Chemistry A 2, 10647 (2014).
⁸ I. Sm, N. St, N. Yh, B. Dk, S. Fj, M. Rs, R. Ns, G. Sk, A. Dk, and P. Vb, J Colloid Interface Sci 493, 162 (2017).
⁹ F. Zhang, J. Zhu, D. Zhang, U. Schwingenschlögl, and H.N. Alshareef, Nano Lett. 17, 1302 (2017).
¹⁰ C.S. Lim, Z. Sofer, O. Jankovský, H. Wang, and M. Pumera, RSC Advances 5, 101949 (2015).
¹¹ Y.C. Lu, C. Ma, J. Alvarado, T. Kidera, N. Dimov, Y.S. Meng, and S. Okada, Journal of Power Sources 284, 287 (2015).
¹² Y.-H. Sun, P.-P. Dong, X. Lang, and J.-M. Nan, Chinese Chemical Letters 25, 915 (2014).
¹³ H. Uchiyama, H. Ohgi, and H. Imai, Crystal Growth and Design 6, 2186 (2006).
¹⁴ S.C. Wang, R.K. Chiang, and P.J. Hu, Journal of the European Ceramic Society 31, 2447 (2011).
¹⁵ W. Liu, L. Yin, R. Zhang, H. Yang, J. Ma, and W. Cao, Nanotechnology 29, 284002 (2018).
¹⁶ J. Ning, T. Jiang, K. Men, Q. Dai, D. Li, Y. Wei, B. Liu, G. Chen, B. Zou, and G. Zou, Journal of Physical Chemistry C 113, 14140 (2009).
¹⁷ S. Wang, S. Xie, H. Li, S. Yan, K. Fan, and M. Qiao, Chemical Communications 507 (2005).

Ruddlesden-Popper phase lanthanum nickelates (La_n+1Ni_nO_3n+1) with two-dimensional perovskite structure have attracted interests for their electronic properties related to the structure and mixed valence Ni-ions ^[1]. These two-dimensional perovskites are also expected as precursors to develop infinite-layer nickelate superconductors such as Sr-doped NdNiO₃ ^[2]. Development of epitaxial thin films of lanthanum nickelate with multilayered structure and modifying their structure, electronic state, and properties according to interfacial strain and substitutional doping would contribute to develop further understanding and application of their solid-state physics. There are some reports of triple-layered La₄Ni₃O₁₀ for its metal-insulator-transition due to oxygen deficiency and possible application to such as SOFCs and gas sensors ^[1,3]. However, aliovalent doping in La₄Ni₃O₁₀ epitaxial thin films similar to La₂NiO₄ and La₃Ni₂O₇ still requires further researches to explain its effect on structure and properties ^[4,5]. In this study, effect of aliovalent dopants substituting La³⁺ in triple-layered La₄Ni₃O₁₀ on its epitaxial growth as well as on the structure and electric properties of the thin films was investigated.
The lanthanum nickelate thin films were prepared by pulsed laser deposition technique equipped with a KrF excimer laser (λ=248 nm, d=20 ns, and E~1.0 J/cm²). The films were deposited on atomically stepped SrTiO₃ and NdGaO₃ single crystal substrates at 700–750°C in O₂ atmosphere (10 Pa) using sintered targets of La₄Ni₃O₁₀ doped with Hf and Sn. The thin films were then thermally treated at ~950°C for 2 hours in high-purity O₂ atmosphere (1 atm, 200 sccm) to modify the crystal phases and the valence of Ni-ions. X-ray diffraction results indicated epitaxy of La₄Ni₃O₁₀ (001) thin films after annealing in O₂, while epitaxial growth of La₂NiO₄ (100) was observed for the as-grown films. The formation of La₂O₃ buffer layer at the interface was suggested to have contribution on the epitaxy of La₄Ni₃O₁₀ phase. The Hf and Sn-doped La₄Ni₃O₁₀ (001) thin films with thickness of ~100 nm demonstrated low resistivity of ~5 mΩcm at room-temperature measured by a 4-probe DC method. Further detailed structural analyses of the lanthanum nickelate thin films, and the effect of the aliovalent doping on the electronic property related to the structure would also be discussed.
[1] K.-T. Wu et al., J. Mater. Chem. A, 5 (2017) 9003–9013.
[2] D. Li et al., Nature, 572 (2019) 624–627.
[3] V. Pardo et al., Phys. Rev. B, 83 (2011) 245128.
[4] R. J. CaVa et al., Phys. Rev. B, 43 (1991) 1229–1232.
[5] S. A. Nedilko et al., J. Alloys Compd., 367 (2004) 251–254.

Thermoelectric effects in conductive polymers (CPs) have received significant attention owing to its high electrical conductivity, high Seebeck coefficient, and low thermal conductivity. Among the CPs, poly(3,4-ethylenedioxythiophene) (PEDOT) is a notable material by taking advantage of solution processability, high degree of crystallinity, and bipolaronic network. We prepared a PEDOT grafted graphene film by solution casting polymerization using surface functionalized graphenes, to provide thermoelectric PEDOT on graphene (PEOG) film. The PEOG showed a high degree of crystallinity, which enhanced both electrical conductivity and Seebeck coefficient. Taking advantage of graphene functionality, TE properties were modulated via several external stimuli. This result reinforced the application potential of CP-based TE materials in a multifunctional energy conversion.

Semiconducting 2D materials are becoming more and more attractive for various optoelectronics and sensing applications; however, synthesis of high-quality 2D materials for lab-scale research is not often practical for many research laboratories, since the synthesis process often requires reactors, carrier gases, or furnaces of various levels of complexities. A simple method of growing various types of 2D materials or their combined structures (i.e. heterostructure, alloys etc.) that is easy to set up and run can further accelerate 2D research. We report on a new synthesis technique, finding that single and multilayered 2D materials of various types can be easily fabricated by direct thermal evaporation of bulk sources under an inert atmosphere. This is in stark contrast with conventional vapor-phase techniques which require chemical reactions of precursors, and a flow of one or more inert carrier gases, thereby substantially reducing the cost and complexities involved. In this study, sources of target nanomaterials are taken in their commercial bulk form, reduced to a powder for convenience and placed on a clean container in close proximity of the target substrate inside an inert environment, then heated to a range of high temperature values and held at those temperatures for material-specific durations. The nanomaterials evaporate from the source and deposit directly on the substrate in the form of well-formed crystals that can be evaluated for quality or modified for basic research or proof-of-concept applications. This synthesis methodology is much simpler than the other synthesis techniques such as CVD, MBE, ME, etc. We present detailed analysis (AFM, optical images, Raman, photoluminescence) of the various types of 2D materials and structures possible through direct, mixed, and sequential evaporation of bulk powders.
By studying the photoluminescence (PL) spectra of the 2D materials fabricated with this method and comparing them with the PL spectra of the 2D materials synthesized by other methods, found in the literature, we show that the quality of these samples, observable in FWHM of the A-exciton, is comparable with the samples grown with the best synthesis techniques. Furthermore, by detailed analysis of the PL spectra of these samples, we observe metrology-dependent variations in quantum properties of these crystals, where the relative distance of the A-exciton and A^--trion becomes smaller for the specific crystals with high PL intensity, which appear to have wrinkles that may have originated from spacial confinement, induced by seed in the nucleation cite.

Modulating the physical properties of two-dimensional materials by stacking different layers to form vertical heterostructures gives rise to the new exciting phenomenon. In this work, we developed a method to improve charge separation in the type-II band aligned heterostructures (InSe/GeS) by using thin dielectric spacers (h-BN). The optical transitions at the junctions exhibit quenched emission without h-BN layers indicating indirect recombination due to the strong interlayer coupling. However, the InSe/GeS heterojunction with h-BN layers in-between exhibits strongly enhanced emission granting an additional degree of freedom to modulate the optical property. The photoresponse of the heterostructures showing enhanced photocurrent density for InSe/h-BN/GeS device by refraining interlayer charge recombination backs up the argument. Thus, the improved charge separation with h-BN mediation demonstrates a higher photoresponsivity and detectivity in the order of 10² A W^-1 and 10¹⁴ Jones, respectively. Furthermore, the high electron detection is evident from the photogain estimated in the order of 10³. In addition, the negative short-circuit current reveals the photovoltaic effect of the device. Therefore, our semiconductor-insulator-semiconductor (InSe/h-BN/GeS) device structure will be a promising candidate for future optoelectronics.

Transition metal dichalcogenides (TMDC), such as MoTe₂, are a promising new class of materials with wide ranging applications in the computing, optics, and photovoltaic industries. The deposition quality of these films has a drastic effect on overall film properties, and therefore device performance. In particular, achieving uniform and reproducible nucleation is important for creation of single monolayer films. However, islanding often occurs during film growth in which isolated regions of the film form initially, creating a discontinuous film and/or non-uniform film thickness, both of which are undesirable. We have investigated the uniformity and thickness control of molybdenum oxide films that are deposited via atomic layer deposition (ALD) and are precursors to MoTe₂ TMDCs. High-sensitivity low energy ion spectroscopy (HS-LEIS) was used to assess surface coverage and islanding of a variety of film MoO_x films ranging in thickness from 0.2 nm to over 7 nm. HS-LEIS is used for this purpose due to its ability to selectively detect and differentiate between atoms in the outermost atomic layer and subsurface atoms. Absence of a signal from the substrate indicated that uniform nucleation and coverage of the surface with MoO_x occurred at film thicknesses of approximately 0.5 nm. Ongoing research is focused on using Monte-Carlo simulations to predict LEIS spectra, which would allow for more quantitative analysis of nucleation and film growth. Quantitative analysis of nucleation and film growth is of great importance for next generation electronics that rely on layers that are only one or several atomic layers in thickness.

Low-dimensional materials such as monolayer transition metal dichalcogenides (TMD) and two-dimensional (2D) perovskites have demonstrated enormous potential for optoelectronic applications.^[1,2] TMDs have been explored for fast, high-responsivity photodetection owing to their sizeable bandgaps and high optical absorption coefficients, but their photovoltaic efficiency is quite low.^[3] On the other hand, 2D Ruddlesden-Popper phase hybrid organic-inorganic perovskites have exhibited encouraging solar cell power conversion efficiencies and better ambient stability compared to the typical 3D perovskites.^[4] However, the major challenges in realizing 2D perovskite based photovoltaic devices are its vulnerability to oxygen/humidity, large exciton binding energy, as well as poor optical absorption in the visible wavelengths.^[2] In this work, using first-principles calculations, we propose a new strategy of employing TMD/2D perovskite heterostructures for optoelectronic applications to mitigate the intrinsic drawbacks of using TMD or 2D perovskite independently. These heterostructures formed via van der Waals (vdW) interaction between the constituents could be custom-designed for specific applications.
We show that the stacking of air stable TMDs like MoS₂ on top of the 2D perovskite (e.g., BA₂PbBr₄, BA is Butylammonium) could lead to improved ambient stability in addition to novel properties of the heterojunction. The calculated electronic band structure of the MoS₂/ BA₂PbBr₄ heterostructure shows a type-II band alignment with good charge separation at the heterojunction. The effective bandgap at the heterointerface is ~ 0.8 eV, much lower than the intrinsic bandgaps of BA₂PbBr₄ (~ 3 eV) and monolayer MoS₂ (1.9 eV).^[1,4] Also, the calculated fermi level alignments of MoS₂ and BA₂PbBr₄ show that the TMD can function as a selective electron transport layer resulting in improved solar cell performance with enhanced air stability. On the other hand, selenide based WSe₂/BA₂PbBr₄ heterointerface demonstrates a type-I alignment with a comparatively higher hetero-bandgap at ~ 1.5 eV, useful for light emitting applications at visible wavelengths.
In summary, this study explores the properties of TMD/2D perovskite heterostructures and provides new insights towards realizing high-performance next-generation optoelectronic devices. The high tensile strength of TMDs and molecular softness provided by the large organic chains could further enable the use of these heterostructures for flexible solar cells or light emitting devices.
References
[1] K. F. Mak, J. Shan, Nat. Photonics 2016, 10, 216.
[2] G. Grancini, M. K. Nazeeruddin, Nat. Rev. Mater. 2019, 4, 4.
[3] D. Jariwala, A. R. Davoyan, J. Wong, H. A. Atwater, ACS Photonics 2017, 4, 2962.
[4] L. Dou, A. B. Wong, Y. Yu, M. Lai, N. Kornienko, S. W. Eaton, A. Fu, C. G. Bischak, J. Ma, T. Ding, N. S. Ginsberg, L. W. Wang, A. P. Alivisatos, P. Yang, Science (80-. ). 2015, 349, 1518.

Two-dimensional (2D) van der Waals (vdW) materials offer a versatile platform to explore their potential applications in electronic, photonic, and spintronic devices. Among them, the family of vdW layered metal phosphosulfides (MPS₃, M = Ni, Fe, Co, Mn, etc.) show intriguing electronic physics, potential magnetic properties, and electrocatalysis applications. Recently, it has been reported that a strong long-range coupling between the electronic and magnetic structures results from super-exchange mechanism in 2D vdW NiPS₃ antiferromagnet (AFM). In this work, we demonstrate a new chemical route for suppression of AFM ordering in NiPS₃ system. Composition-tunable Ni_1-xCo_xPS₃ (0≤x<0.50) few-layered nanosheets (NSs) are successfully synthesized. Significantly, obvious suppression of AFM ordering below Néel temperature (T_N ~ 143-155 K) in magnetic susceptibility measurements were observed in Ni_1-xCo_xPS₃ NSs, which contrast with the magnetic properties observed in bulk single crystals of Ni_1-xCo_xPS₃. These findings provide a new direction for disruption of AFM long range ordering and opens a new window of opportunities for investigation of correlated physics and magnetism in 2D materials systems.

As silicon-based field effect transistors (FET) face miniaturization limitations, various two-dimensional (2D) material-based FETs are promising for next-generation electronic and optoelectronic devices. These days, not only study to produce high performance 2D-FETs, but also to implement integrated circuits with 2D materials has been reported. In terms of integrated circuits, PN junction is essential for complementary circuit. To fabricate 2D PN junction, usually transferring and stacking with exfoliated flakes has been used. Even if the flakes electronic quality is good, it needs trained skills and is time consuming. In addition, it is hard to ensure interface quality and has residue problems which deteriorate device performances.

In this paper, we report the direct growth of PN heterostrucutures with transition metal dichalcogenide (Nb doped WSe₂/MoSe₂) on SiO₂/Si substrates with chemical vapor deposition (CVD) method with one mixed solution source. To confirm the composition and structural properties of our as-grown heterojunction sample Raman, PL, XPS and AFM analysis is done. At inner and outer domain, obviously different center Raman and PL peaks are observed. The niobium (Nb) doping signal also confimed by Raman peaks and XPS peaks. In AFM height profile data, inner and outer height show no difference which represents lateral heterostructure. To obtain electronic properties at inner, outer and junction region, multi electrodes are evaporated. At junction, current rectification behavior is shown as a PN diode with 1.3 ildeality factor which is close to ideal ideality factor of 1. Also the current level between P (Nb doped WSe₂) and N (MoSe₂) domain is simillar. Under lasers illumination, optoelectronic characteristic is shown. Several properties such as built-in potential across the junciton, type-II band alignment of the materials (WSe₂ and MoSe₂), low lattice mismatch and atomically sharp interface from directly growth mechanism, allow high advantages in optoelectronic devices. Additionally, the niobium acceptor dopants allow wide range of absorption wavelength from ultraviolet (UV) to near infrared (NIR) which overcomes the limitation of the TMD photodetectors. Additionally, back gated PN photodiode can effectively suppress dark current through gate voltage modulation, thus realizing very high detectivity.

Thus, our direct grown Nb doped WSe₂/MoSe₂ PN heterostructure photodiode shows outstanding optoelectronic performances, such as high detectivities (10¹⁵ Jones), I_light/I_dark (10⁵) under V_DS = 0 V. Our approach of direct grown PN heterostructure photodiode allows fabricate high quality and easy integrated optoelectronic 2D atomic layer devices.

Two-dimensional (2-D) nanomaterials have recently received considerable attention with their unique characteristics, which were not seen in their bulk and 1-D nanomaterials. Zinc oxide (ZnO) is chemically stable, non-toxic, and available at low cost, which makes it into one of the most intriguing materials. Since the surface is surrounded largely by basal c-planes, 2-D ZnO nanomaterials have the good potential for tailoring the functional properties through adjusting their surface morphologies. To achieve the growth of such 2-D ZnO nanoplatelets, the growth direction must be along the a-axis, while suppressing the growth along the c-axis. This growth is challenging because the growth along the c-axis is energetically favorable. To control the growth direction, a seed layer plays a significant role. r-plane (1-102) aluminum oxide (Al₂O₃) can serve as a good seed layer for 2-D ZnO growth as the lattice mismatch between r-plane Al₂O₃ and a-plane ZnO is only 1.53% along ZnO [0001] and Al₂O₃ [2-201] directions. In this study, we prepared r-plane oriented Al₂O₃ on Si (111) substrate as a seed layer by spin coating followed by annealing at high temperatures. r-plane Al₂O₃ also has an epitaxial relationship with Si (111). 2-D ZnO nanoplates were then grown on the seed layer by the hydrothermal method. The c-plane of ZnO nanoplates contains a high concentration of the surface defects and is off-stoichiometric; therefore, the thinner the nanoplate is, the more it is influenced by the defects on the c-plane. It makes these 2-D nanoplates very attractive as an effective photocatalytic surface for the degradation of organic contaminants. As a result, the templated growth of the 2-D ZnO nanoplates shows the great potential to form functional surfaces that can be very sensitive to external stimuli such as light and environment by controlling the concentration and type of defects. Photoluminescence (PL) and Raman spectra are used to characterize defect characteristics of various nanoplate morphologies. This presentation highlights current research progresses made for the control of the defect population, size, aspect ratio, and packing density of the ZnO nanoplates at various conditions and the effect of the resulting morphologies of the nanoplates on their photocatalytic properties.

Black phosphorus (BP) is an elemental two-dimensional semiconductor with quantized electronic structure, high mobility and tightly bonded exciton. Especially, due to the quantum confinement in the out-of-plane direction and significant interlayer interaction, BP has direct band gap for all thicknesses that is highly tunable from visible region to mid-infrared region. Besides, the conduction band and valence band in phosphorene split into quantized subbands, which constitute a series of thickness-dependent interband electronic transitions. These pronounced subband transitions, evidenced by sharp thickness-dependent absorption peaks in the visible region and infrared region, give rise to strong coupling with photons and phonons in few-layer BP. Here we investigated the electron-photon and electron-phonon coupling in few-layer BP samples by resonance Raman spectroscopy. Firstly, we observed Raman splitting effect in atomically thin-BP, and the origin of this phenomenon will be revealed. Secondly, based on the collective interlayer vibration modes, the accurate interlayer interaction of BP was measured in experiment. Thirdly, A_g¹ mode shows much smaller resonance effect than A_g² mode when the incident polarization is along armchair direction of BP, which can be explained by the symmetry-dependent electron-phonon coupling. These investigations give direct evidence of the significant interlayer interaction and deep insights into fundamental electron-photon and electron-phonon coupling in atomically-thin BP.

Molybdenum disulfide (MoS₂) is one of the most promising hydrogen-evolution-reaction (HER) electrocatalysts to replace noble metals in water splitting. Raman scattering is one non-destructive, efficient, and mature technique to characterize renewable energy materials. Here, the 2D MoS₂ nanosheets with various crystallinity are prepared and their electrocatalytic behaviors are in-situ investigated by Raman scattering under varied electric fields and temperatures. The results would enlighten the electrocatalytic mechanism of MoS₂ nanosheets and help us to improve its HER activities over the Pt catalyst.

Hexagonal boron nitride (hBN), a layered van-der-Waals material, is employed as a template and an insulator to improve the performance of devices made from graphene and other two-dimensional (2D) materials. Moreover, due to its oxidation resistance, it is used to protect 2D materials devices from the environment through encapsulation. This study aims to understand the ultimate protective qualities of hBN layers at high temperatures in dry vs. ambient air. Single crystals of hBN were thermally oxidized at different temperatures ranging from 800 °C to 1100 °C for different durations, from 20 min to 100 min. Atomic force microscopy analysis revealed surface pits of about 50-100 nm in width, and 1-2 nm in depth, for the layers exposed to 900 °C for 20 minutes in dry air. Increasing the temperature and duration did not significantly change the surface pit sizes, in contrast to the thermal oxidation in ambient air, where longer oxidation times and higher temperatures, produced larger etch pits. Having no effect of thermal oxidation as a function of temperature and duration in dry air could be due to formation of the surface oxide layer, which slows down the etching process. Complimentary transmission electron microscopy will be employed to compare the extent of surface oxidation in ambient versus dry air and to study the origin and morphology of the surface pit formation.

Atomic layers of transition metal dichalcogenides (TMDCs) have attracted much attention because of their unique two-dimensional structure and high electrical performance. For electronics applications, it is essential to develop a sophisticated way to control their carrier density of materials. So far, carrier control of TMDCs is achieved via surface charge transfer by using chemical doping. To further improve doping concentration, it is highly desired to find any stable and efficient dopants for TMDCs. Recently, Nonoguchi et al. have reported the electron doping of carbon nanotubes using a series of ordinary salts with crown ethers. In this study, we have applied this technique to monolayer MoS₂ and investigated the transport and optical properties of such chemically-doped MoS₂.
Monolayer MoS₂ films were grown on SiO₂/Si substrates by using chemical vapor deposition method. Electron doping was conducted by spin coating of butanol solution of KOH and benzo-18-crown-6. The doping process changes the transfer curves of MoS₂ field-effect transistors (FETs) from a typical n-type semiconducting state to metallic one in air. We also found photoluminescence quenching and peak shift in the A’₁ Raman mode for doped MoS2. These results indicate that the salt-based doping provides an effective way for the electron doping of MoS₂. In the presentation, the details of transport and optical properties will be discussed.

Atomically thin two-dimensional (2D) Van der Waals (vdw) magnets have received extensive attention due to their interesting magnetic phenomena inherent to low-dimensional magnetism such as Berezinskii-Kosterlitz-Thouless transition in 2D XY models and Mermin-Wagner Theorem. However, it is not until 2017 that the first 2D ferromagnetic magnets such as Cr₂Ge₂Te₆ and CrI₃ were experimentally realized via mechanical exfoliation method using Scotch tape. Since then, although a number of potential 2D vdw magnets have been theoretically predicted a very limited number of atomically thin or few-layered 2D vdw magnets still have been experimentally available. Specifically, since 1960’s, 2D Ruddlesden-Popper organic-inorganic halide perovskites (RH-OIHP) have been considered as the ideal model closest to 2D Heisenberg model. Recent theoretical calculation using first principle density functional theory shows that the cleavage energy of (R-NH₃)₂CuCl₄ (R = a monovalent organic moiety such as C_nH_2n+1 or C₆H₅C₂H₄) is smaller by a factor of 2-2.5 than that of graphite (0.36 J/m²). It means that 2D RP-OIHP monolayer by mechanical exfoliation from bulk crystals would be experimentally feasible. However, until now, there has not been successful report on fabrication of atomically thin or few-layered 2D RP-OIHP thin films with intrinsic magnetism at all. We will demonstrate the few-layered 2D OIHP magnets via solvent engineering based on spin coating technique.

Silicene, an analogous silicon counterpart of graphene with unique physical properties, has attracted a great deal of attention. Its interaction with environmental molecules, e.g ozone needs to be investigated thoroughly as it has a great role to play for silicon-based technology at the nanoscale. In this study, we study the interaction of O, O_2, and O₃ with silicene using density functional theory. We find that O, being more electronegative than Si, draws electrons from the Si-Si bond. Consequently, it is chemisorbed at the bridge site forming an epoxide like Si-O-Si configuration with a binding energy of 5.76 eV. On the other hand, O₂ and O₃ interacting with silicene dissociate at the surface, in contrast to what has been reported for graphene. Moreover, the dissociation process is spontaneous and exothermic without any energy barrier leading to the formation of Si-O bonds at the surface.

Nanosized materials often possess unique characteristics compared to their bulk counterpart and are useful in various applications. Bismuth vanadate (BiVO4), a type of mixed-metal oxide, gains popularity in the field of catalysis reaction, especially the light-induced related process. They are active towards visible light irradiation and able to stimulate a catalytic reaction, for example, the water oxidation process. However, the main drawback of BiVO4 is the photogenerated electron-hole pairs tend to recombine in a very short period, which negatively impacts their catalytic performance. Coupling metal oxide with other materials is often regarded as one of the promising methods to alter their properties and improve catalytic performance. In the field of catalysis, polyoxometalates (POMs) have a distinctive property to act as electron and hole scavengers, which can retard the recombination of photogenerated electron-hole pairs effectively. They belong to a large class of nanosized metal-oxygen anion complexes of early transition elements with tunable structure, their properties greatly influence by the type of metals present in their structure. In this experiment, all the materials were synthesized in mild condition, without the use of strong base and acid. By avoiding a harsh environment, the complexity in the synthesis process can be reduced and are safer. A nanocomposite of BiVO4 and cobalt-based POMs is produced to integrate the excellent properties of both materials for the application in photochemical water oxidation. Characterization tests are performed, such as field emission scanning electron microscopy (FESEM) to study the shape and size of composites materials, X-ray diffraction (XRD) for confirming the crystal lattice structure and Fourier-transform infrared spectroscopy to identify the types of chemical bonds. A sheet-like behavior of BiVO4 can be noticed from the FESEM analysis. The synthesized BiVO4 is in a monoclinic phase as confirmed by the XRD result, in which intense peaks are noticed for (130) and (040) planes, while the POMs produced are in the triclinic phase. It is believed that the BiVO4/POMs composite is able to improve the photocatalytic water oxidation reaction.
Keywords: Bismuth Vanadate, Nanomaterials, Polyoxometalate, Mild condition

Antimonene, one of the group V elemental monolayers, has attracted intensive interest due to their intriguing optical properties. Considering the case of atomically-flat antimonene for which directional bonding between Sb atoms appears to be analogous to C-C bonds in graphene, we present its optical absorption properties from near IR to visible spectral regime. The results based on density functional theory predict that absorbance in multilayer antimonene is equivalent to or higher than that calculated for multilayer graphene. Specifically, the IR absorption in antimonene is significantly higher with a prominent peak at about 4 mm associated with the dipole-allowed interband transitions. Moreover, a strong dependence of absorbance on topology is predicted for both antimonene and graphene which can be attributed to the subtle variations in their stacking-dependent band structures. Our results suggest the possibility of multilayer antimonene coating be realized as effective optical limiting material in the IR region for optoelectronic materials.

Graphene-based composite materials and nanostructures have been intensively studied in terms of unique electrical properties and a wide spectrum of applications. Among them, reduced graphene oxide(rGO) has attractive advantages of flexibility, electrical and thermal conductivity, and low-cost production. We have already reported memristive properties in rGO films partially reduced by the laser irradiation to graphene oxide(GO). However, the resistance switching mechanism has not been fully understood. In this study, by electrically characterizing the transition region from GO to the laser-irradiated rGO, a possible mechanism for the memristive properties is discussed.
Reduced GO film preparation: The GO films prepared by drop-casting commercially available GO colloid (4mg/mL) on PET film were placed on a 3D-shaker at room temperature until the water was fully evaporated. A PC-driven 405nm laser (scan speed 3 min/cm2) was used for the photothermal reduction of GO films with a rectangular shape(~1.5mm width).
Electrical characterization: GO films were initially non-conductive, while it became conductive with the laser-power increase. Interestingly, when GO films were reduced at the threshold laser-power for GO reduction, memristive properties were clearly observed. Around this region, on/off ratio was rather sharp (digital change) and large (maximum ratio was nearly 12). In addition, hysteric I-V characteristics (continuous change) were also detected. In this study, we mainly discuss the continuous hysteresis observed in the transition region between laser-irradiated and non-irradiated GO films.
The electrical measurements were performed by two-probe measurement. The probe-head was made of tungsten carbide with r = 25μm. The probe distance was 1 mm and the probe weight was fixed at 50g. The current was monitored by a source-measure unit. Two probes were set in parallel to the laser-irradiated side region.
Two-probe resistance systematically changes from 6M ohm to 3kohm, as the probes were approached from off region to laser-irradiated one. Furthermore, the continuous hysteresis was observed, which means the resistance changed from the high(Rup) to low states(Rdown) by increasing and decreasing the applied voltage. At the distance of 1mm from the laser-reduced parts, Rup/Rdown, i.e., (resistance when applied voltage increases)/(resistance when applied voltage decreases after increasing) is quite large(very large hysteresis). Around or on the laser reduced GO area, Rup even becomes smaller than Rdown.
As the mechanism of electrical conduction in rGO films, the hopping conduction between highly conductive sp2 domains has been proposed. The sp2 domain created by the laser irradiation are electrically connected by the hopping conduction in fully reduced area. When the current increases by applying voltage, the carrier trapping that can be the potential barrier for the transport could occur. It can cause the increase in the resistance, resulting in Rup/Rdown <1. On the other hand, at the area with the short distance (0.5~1mm) from the laser-irradiated GO, as the remained sp2 domains due to the imperfect reduced graphene oxide could be critically connected, small current still flows and large continuous hysteresis is observed. As conduction paths cannot respond to the applied voltage instantaneously, Rup could be larger than Rdown by the voltage sweeping.
Summary: We systematically investigated the electrical transport in partially-reduced GO films by changing the distance from the laser-reduced area. We characterized electrical properties in the partially reduced area between GO to rGO and found the continuous hysteresis. Our results show that GO film can be a resistor with memory function by laser assisted reduction; an inexpensive and simple method. The memory effect is basically originated from graphene and oxygen-containing functional groups bonded to carbon. Therefore, this type of resistor can be scaled down to the atomic level.

Gadolinium-containing nanocrystals as T₁ MRI contrast agents (CA) offer the potential for intracellular accumulation, active targeting, and clearance pathways that avoid the kidneys. However, the most efficient T₁ relaxation mechanism requires that water protons come within a few angstroms of surface Gd³⁺, a difficult condition to meet as these surfaces are necessarily covered with polymers that prevent aggregation in biological environments. This problem is resolved with two-dimensional Gd₂O₃ nanoplates (GONP) which, due to their shape, possess accessible edge Gd³⁺ for inner-sphere water interactions. A novel sulfonated surface coating further contributes to water-surface exchange and second-sphere interactions while also providing colloidal stability over a wide range of conditions. At clinically relevant field strengths (≥ 1.4 T), these nanoplates exhibit ionic T₁ relaxivities an order of magnitude greater (r₁ = 63.0 mM-Gd^-1s^-1 at 1.4 T) than commercial gadolinium-chelates and significant performance even at 9.4 T (r₁ = 32.6 mM-Gd^-1s^-1). Systematic size and magnetic field dependence studies suggest while inner-sphere relaxation processes at nanoplate edges contribute to r₁, second-sphere interactions are also important and become more significant at higher fields. Nanoplates are readily taken into cells where they maintain contrast performance and exhibit minimal cytotoxicity. In vivo, nanoplates preferentially accumulate in the liver and their circulation half-life is 1.5 times faster than commercial gadolinium-chelates. These features allow for the novel demonstration that nanoscale T₁ MRI contrast agents can be applied to detect non-alcoholic fatty liver disease (NAFLD) in mice.

Visualizing biological activity in real time poses a novel way for biosensing. Single-layer Molybdenum disulfide (MoS₂) shows strong photoluminescence (PL) under photo-excitation due to its direct bandgap property [1]. The PL intensity of MoS₂ transistors has been demonstrated to change with the internal electron density by applying a back-gate voltage [2]. It has also been reported that the PL intensity of mechanically exfoliated MoS₂ was modulated by pH changes in aqueous solution [3]. In this work, we demonstrate pH-sensitive photoluminescence of single-layer MoS₂ grown by chemical vapor deposition (CVD). While CVD MoS₂ grown from MoS₂ powder did not have a strong pH sensitivity, MoS₂ grown from MoO₂ powder exhibited a sensitivity to the solution pH in its PL intensity [4]. Next, we utilized the single-layer MoS₂ as the active layer of an imaging sensor to monitor the activity of individual bacteria by the MoS₂ PL. For the demonstration, we employed Lactobacillus. It is one of the most studied probiotic bacteria which consumes glucose and produces lactic acid by fermentation. The bacteria were placed on a single-layer MoS₂ and its PL intensity was monitored in real time.
Experiments were performed in the following ways. For device fabrication, CVD MoS₂ grown on a Si substrate was transferred onto a cover glass with patterned gold micro-electrodes. An electrolyte solution was place on the MoS₂ with electrode. A reference electrode was also inserted into the solution to apply a voltage between the MoS₂ and the reference electrode. Using these three electrodes, the MoS₂ acted as an electrochemical transistor. The PL image was observed through the cover glass by an inverse microscope with an oil immersion lens under a controlled voltage of the reference electrode. The PL intensity of the MoS₂ transistor was modulated by the applied voltage from the reference electrode in the buffer solution. As decreasing the voltage at the reference electrode, the PL intensity was appeared at a threshold voltage and increased monotonically. We observed that the threshold voltage changed depending on the concentration of lactic acid in the bulk solution. Next, we placed a solution of Lactobacillus rhamnosus GG (LGG) onto MoS₂. We found that LGG specifically adhered onto MoS₂ rather than glass substrate. PL images of MoS₂ with LGG showed a pattern that correlates with LGG adhesion. The contrast of the PL images was tuned via the applied voltage. Furthermore, PL intensity at the vicinity of the adhered LGG changed over time, which is probably due to a change of lactic acid concentration. We assume that the local pH change caused by the lactic acid possibly modulated the carrier density in MoS₂. In this work, we demonstrated that PL imaging of MoS₂ has the potential to visualize activity of living individual bacteria label-freely [5].

Mak KF, He K, Lee C, et al. Nat Mater. 12(3):207-211 (2012).
Mouri S, Miyauchi Y, Matsuda K. Nano Lett. 13(12):5944-5948 (2013).
Zhang W, Matsuda K, Miyauchi Y. J Phys Chem C. 122(24):13175-13181.(2018).
Seki T, Ihara T, Kanemitsu Y, Hayamizu Y. 2D Mater. 7 034001 (2020).
Tezuka S, Seki T, et.al. 2D Mater. 7 024002 (2020).

Cellular layers play critical role as barriers, filters, and membranes in our bodies. The performance of cellular layers is found to be determined by the properties of individual cellular gaps. Yet these gaps often fall below the diffraction limit, hampering their detection. Here, we develop a novel sensing modality, where the diffusion of a harmless chemical tracer is used to reveal cellular gaps in human umbilical vein endothelial cellular (HUVEC) layers. For the readout, we employ atomically-thin and optically-active MoS2, whose two-dimensional form-factor enables it to serve as an array of 20,164 wireless sensors over 420×420 µm² area. A rigorous screening of 30 analytes identifies several tracer candidates with MoS2 dynamic range spanning over five orders of magnitude in tracer concentration. In a single layer of HUVEC cells, we find local spots with diffusivities ranging from 1×10^-9-3×10^-8 cm²/s that are spatially correlated with immunofluorescence of vascular endothelial (VE)-cadherin proteins found on cell membrane. These spots correspond to breaks in nanoscale cellular junctions with <40 nm in length. This platform enables future studies on properties and dynamics of cellular gaps of various tissues.

Nanopore biosensing technology has been developed over the past two decades as a label-free, single-molecule sensing platform of biopolymers (DNA, RNA, and proteins). In this technology, a nanometer-sized pore, either a naturally-occurring biological channel self-assembled in an organic membrane (biological nanopore) or a synthetically-fabricated channel in a solid-state membrane (solid-state nanopore), is the only liquid junction between two buffer bath chambers. A charged biopolymer, such as DNA, is then threaded through the nanopore by applying a voltage across the nanopore, resulting in partial physical obstruction of the pore detected as a transient drop in the ionic current. This event is used to electrically and/or optically probe structural features of the DNA or determining its nucleic acid sequence. A key feature in this technology is the geometry of the nanopore (diameter and thickness in solid-state nanopores) that defines the total sensing resolution. For instance, two geometric requirements for DNA sequencing are a pore with comparable geometry to the width of a single strand of DNA (~1 nm in diameter), and a pore thickness of 0.5 nm (distance between bases on a DNA strand) to provide single base resolution. The one-atom-thick monolayer graphene with a thickness of 0.3 nm has been a candidate for this application since it was first introduced in 2010¹. However, despite the atomically-thin pore, a phenomenon known as “access resistance” still dominates, producing a sensing region longer than the physical pore thickness and reducing the resolution². Moreover, other challenges associated with graphene nanopores such as surface hydrophobicity and poor mechanical stability have phased out its application in nanopore technology.
Recently, we have demonstrated that “MXene nanopores” with high mechanical robustness, long-time stability, and low-noise ion current recordings could serve as an alternative to graphene³. MXenes are an emerging family of two-dimensional transition metal carbides and nitrides with a general formula of M_n+1X_nT_x (i.e. Ti₃C₂T_x), where M is a transition metal, X represents carbon or nitrogen (n=1, 2, and 3), and T_x indicates different functional groups (O, OH, F) present on MXene surface. An intriguing property of MXenes is the ability to accommodate various kinds of organic molecules and cations into their interlayer spacing which has led to the high performance of MXenes electrodes and films in energy storage and sensing applications. In this work, we have harnessed this property of Ti₃C₂T_x MXene along with its in-plane electrical conductivity to engineer a novel nanopore sensing device with increased sensing resolution by eliminating the access resistance. In this design, bilayer Ti₃C₂T_x films serve as an embedded electrode and reservoir of cations, intercalated in the interlayer space, and support membrane of the nanopore, where the cations enter the pore from within the plane of the membrane rather than the bulk solution. By elimination of the access resistance contribution, this technology could reach the resolution of detecting three DNA bases at the pore sensing region, surpassing the commercially available nanopore technology with sensing resolution of four bases at the pore constriction.

1. Garaj, S., Hubbard, W., Reina, A. et al. Nature 467, 190–193 (2010). DOI:10.1038/nature09379
2. Comer, J., Aksimentiev, A. Nanoscale 8, 9600-9613 (2016). DOI:10.1039/c6nr01061j
3. Mojtabavi, M., VahidMohammadi, A. et al. ACS Nano 13, 3042-3053 (2019). DOI:10.1021/acsnano.8b08017

This talk will present our latest progress on materials, transistor and memory devices based on 2D atomically-thin films. In particular, we explore a new substrate, namely, solid lithium-ion electrolytic substrates, for the direct growth and realization of high-performance TMD devices and amplifiers. Electrostatic gating with ionic liquids (ILs), leading to the accumulation of high surface charge carrier densities, has been often exploited in thin-film transistors. However, the intrinsic liquid nature of ILs inhibit it from constituting an ideal platform. In this regard, a lithium-ion solid electrolyte, often used for battery applications, offers similar advantages as ionic-liquids, with the added benefit of solid-state compatibility. With the solid electrolyte, we demonstrate its application in high-performance n-type MoS2 and p-type WSe2 transistors with sub-threshold values approaching the ideal limit of 60 mV/dec and complementary inverter amplifier gain of 34, the highest among comparable amplifiers with a 1 V power supply. These results establish lithium-ion substrates as a promising alternative to ionic liquids for the study of thin-film devices.

Field-effect transistors (FETs) based on atomically thin transition metal dichalcogenides (TMDs) such as molybdenum disulfide (MoS₂) and tungsten disulfide (WS₂) have attracted immense attention in the scientific community due to their superior electronic properties.[1] The immense promise provided by the TMD FETs needs to be complemented by a thorough statistical and benchmarking study of relevant device parameters. In this regard, we report a comprehensive study of the carrier transport in monolayer MoS₂ and WS₂, which were grown over a large area using metal-organic chemical vapour deposition (MOCVD) process. 160 WS₂ FETs and 230 MoS₂ FETs with channel length ranging from 100 nm to 5 μm were investigated for benchmarking field-effect carrier mobility, maximum current, threshold voltage, saturation velocity, device linearity, contact resistance, subthreshold slope (SS), current, ON/OFF ratio, and drain induced barrier lowering (DIBL). Our study shows low device-to-device variation owing to high growth quality. We are also able to demonstrate record high carrier mobility of 33 cm²V^-1s^-1 in WS₂ FETs, with over 1.5X improvement compared to the best reported in the literature. [2-4] This work also demonstrates the first comprehensive statistical study on WS₂ films, showing comparable performance to MoS₂ films. Our experimental demonstrations confirm the technological viability of two dimensional (2D) semiconductors for use in commercial electronics.

References
[1] D. S. Schulman, A. J. Arnold, and S. Das, "Contact engineering for 2D materials and devices," Chem Soc Rev, vol. 47, pp. 3037-3058, May 8 2018.
[2] Y. Gong, V. Carozo, H. Li, M. Terrones, and T. N. Jackson, "High flex cycle testing of CVD monolayer WS₂ TFTs on thin flexible polyimide," 2D Materials, vol. 3, p. 021008, 2016.
[3] K. Kang, S. Xie, L. Huang, Y. Han, P. Y. Huang, K. F. Mak, et al., "High-mobility three-atom-thick semiconducting films with wafer-scale homogeneity," Nature, vol. 520, pp. 656-60, Apr 30 2015.
[4] S. J. Yun, S. H. Chae, H. Kim, J. C. Park, J. H. Park, G. H. Han, et al., "Synthesis of centimeter-scale monolayer tungsten disulfide film on gold foils," ACS Nano, vol. 9, pp. 5510-9, May 26 2015.

Two-dimensional (2D) materials offer the freedom to create novel condensed matter systems, with unique properties, by mechanically assembling different (or same) 2D materials layer-by-layer to form atomically sharp vertical or lateral heterostructures. The van der Waals (vdW) heterostructures with small lattice mismatch and a relatively small twist angle between the constituent layers, have shown to exhibit coexisting complex phases of matter including Mott insulating state, superconductivity, bound quasiparticles, and topological states. In addition, the extreme surface sensitivity of two-dimensional (2D) materials provides an unprecedented opportunity to engineer the physical properties of these materials using external perturbations. In this talk, I will discuss the utilization of angle-resolved photoemission spectroscopy (ARPES) with high spatial resolution to investigate the electronic band structure of 2D heterostructures and their devices. This can shed light on the intricate relationship between controlled external perturbations, substrate, and electronic properties of 2D materials [1, 2]. In particular, I will discuss our in-operando nanoARPES results that exhibit highly tunable many-body effects in graphene devices [3] and tunable van Hove singularities in twisted bilayer graphene [4].
References:
[1] Katoch et. al., Nature Physics 14, 355-359 (2018).
[2] Ulstrup, et. al., Science Advances, Vol. 6, no. 14, eaay6104, (2020).
[3] Muzzio, et. al., Physical review B Rapid Communications 101, 201409(R) (2020).
[4] Jones, et. al., arXiv:2006.00791 (2020). Accepted for publication in Advanced Materials.

Two dimensional (2D) van der Waals semiconductors are exciting candidates for next generation optoelectronics. The layer-by-layer assembly of 2D monolayers enables the creation of heterostructures with unique and controllable properties. It has been shown previously that photoexcitation of 2D heterostructures induces ultrafast charge transfer across the van der Waals junction on timescales faster than 100 fs, resulting in the formation of interlayer excitons. However, the coupling between this charge transfer process and thermal excitations of the lattice is so far unexplored. Here, using mega electron volt (MeV) ultrafast electron diffraction (UED), we make the first direct measurements of atomic vibrations within the individual layers of a type II van der Waals heterostructure following femtosecond photoexcitation. The UED technique employs ~4 MeV electron pulses to probe a large range of reciprocal space up to ~12 A^-1, with a time resolution of ~100 fs. We track the propagation of energy from one layer to the next by measuring the changes in intensities of Bragg peaks. First principles calculations link the measured differential intensities with thermally-activated atomic displacements, providing a sub-nanometer temperature scale. Our results provide insights into the role of ultrafast charge transfer on lattice dynamics, and decouple the vibrational and electronic pathways for energy transfer across a van der Waals interface.

MXenes, derived from metallic MAX phases, are a class of two-dimensional materials with emerging applications in energy storage, electronics, and catalysis due to their high surface areas and attractive electronic properties. MXene properties are heavily influenced by their surface chemistry but a detailed understanding of the surface functionalization for the niobium MXenes Nb₂CT_x and Nb₄C₃T_x is still lacking. The parent MAX phases, Nb₂AlC and Nb₄AlC₃, exhibit interesting structural and electrical properties but can be sensitive to oxidation. Solid-state NMR allows us to probe the interfacial chemistry, as well as the connectivity using correlation experiments; the phase purity including the presence of amorphous/nanocrystalline phases; and the electronic properties of the MXene and MAX phases with quantitative chemical sensitivity. In this work, we systematically study of the chemistry of Nb MAX and MXenes using solid-state NMR spectroscopy and examine a variety of nuclei (¹H, ¹³C, ¹⁹F, ²⁷Al and ⁹³Nb) with a range of techniques (T₁-filtering, QCPMG, MATPASS, HETCOR, TRAPDOR, robot-automated VOCS). Hydroxide and fluoride terminations were identified, found to be intimately mixed, and their chemical shifts compared with other MXenes. In addition to the identification of surface species and their spatial distribution, this multinuclear NMR study demonstrates that diffraction alone is insufficient to characterize the phase composition of MAX and MXene samples as numerous amorphous or nanocrystalline phases were identified including NbC, AlO₆ species, aluminum nitride or oxycarbide, AlF₃×nH₂O, Nb metal, and unreacted MAX phase. To the best of our knowledge, this is the first work to examine the metal resonances directly in MXene samples, and the first ⁹³Nb NMR of any MAX phase. Insights from the ⁹³Nb even enabled the previously-elusive assignment of the complex overlapping ^47/49Ti spectrum of Ti₃AlC₂. The results and methodology presented here provide fundamental insights on MAX and MXene phases and can be used to obtain a more complete picture of MAX and MXene chemistry, to prepare realistic structure models for computational screening, and to guide the analysis of property measurements.

Following the success of moiré engineering in modulating (opto-)electronic properties of twisted bilayer graphene and graphene/hBN heterostructures, studies have extended the moiré toolbox; from double bilayer, to trilayer graphene, transition metal dichalcogenide (TMDC/TMDC), and hexagonal boron nitride (hBN)-graphene-hBN vdW heterostructures [1]. Moiré engineering has yet to be explored in ‘quasi-vdW’ systems, such as the interface of a two-dimensional (2D) material and three-dimensional (3D) metal, despite its influence on contact resistance, photo-response, and high-frequency performance [2]. Herein, we report moiré modulation at a ‘quasi-vdW’ interface between a 3D metal and 2D material, in the model system of 0 ± 0.1^o epitaxially aligned Au{111}/MoS₂.

By applying a predictive geometric convolution technique and four-dimensional (4D) scanning transmission electron microscopy (STEM) to epitaxially aligned Au{111}/MoS₂, we demonstrate double moiré periodicity (18 Å and 32 Å) at the 2D/3D interface. 4D STEM allows us to decouple these multiple moirés and image weaker moiré superlattices, otherwise hidden in conventional (S)TEM. Using ab initio electronic structure calculations, we find that charge density is modulated with a period of 32 Å, illustrating electronic conductivity and (opto-)electronic modulation via moiré engineering at the 2D/3D interface. With growing potential for tunability in quasi-vdW layers, these results highlight the potential for moiré engineering at the 2D/3D interface, as well as showcasing the growing opportunities for advanced (S)TEM techniques in multiple moiré direct imaging.

[1] Balents, L., Dean, C.R., Efetov, D.K. et al. Superconductivity and strong correlations in moiré flat bands. Nat. Phys. (2020).
[2] Wang, Y. et al. Van der Waals contacts between three-dimensional metals and two- dimensional semiconductors. Nature 568, 70–74 (2019).

The quality of the interfaces is critical for two-dimensional (2D) materials, as atomically thick 2D materials do not have any bulk to screen impurities.¹ For electronic devices based on 2D heterostructures, it is thus vital to have clean and smooth interfaces. Mechanical assembly is the most common and versatile way to assemble 2D heterostructures,^2,3 which often traps contaminants at the interfaces during assembly and does not yield high quality devices. Contact mode atomic force microscopy (AFM) cleaning is a reliable and controllable technique to remove interface contaminants of 2D heterostructures after assembly.^4–6 Reduction of nanometer scale corrugations was surmised to be responsible for improved electrical performance of 2D heterostructures after contact mode AFM cleaning.⁵ Direct observation of the flattening of nanometer-scale interfacial roughness would help elucidate how contact mode AFM cleaning improves the electrical performance of 2D heterostructures.
This work reports the reduction of nanometer-scale roughness and improvement of interface qualities of 2D heterostructures, by scanning the heterostructure with an AFM tip in contact mode. The prototypical heterostructure system consists of monolayer MoS₂ encapsulated by top monolayer hBN and bottom multilayer hBN. We detected spatial height fluctuations ~ 1 nm with a lateral dimension of tens of nanometer in bubble-free areas of the heterostructures with AFM, which were minimized by contact mode AFM cleaning. The AFM cleaning process also reduced the photoluminescence (PL) linewidth of monolayer MoS₂, indicating a reduction of interface impurities. The optimal cleaning force for the heterostructure was found to be 70 nN. One MoS₂ field-effect transistor (FET) fabricated on AFM-cleaned heterostructure achieved a high mobility of 73 cm²/Vs, which can be further improved by passivation with thick hBN or high-k dielectrics for better electrostatic environment. The mobility of four hBN encapsulated monolayer MoS₂ FETs fabricated on as-assembled heterostructure improved from 21±2 cm²/Vs to 37±6 cm²/Vs after AFM cleaning. Combining the results from AFM, PL and electrical measurements, we infer that contact mode AFM cleaning is effective in reducing interface disorders and enhancing the mobility of hBN encapsulated monolayer MoS₂. Finally, we anticipate that contact mode AFM cleaning will significantly improve the quality of most mechanically assembled van der Waals heterostructures by cleaning and smoothing their interfaces.

References:
(1) Dean, C. R.; Young, A. F.; Meric, I.; Lee, C.; Wang, L.; Sorgenfrei, S.; Watanabe, K.; Taniguchi, T.; Kim, P.; Shepard, K. L.; et al. Boron Nitride Substrates for High-Quality Graphene Electronics. Nat. Nanotechnol. 2010, 5 (10), 722–726.
(2) Novoselov, K. S.; Mishchenko, A.; Carvalho, A.; Castro Neto, A. H. 2D Materials and van Der Waals Heterostructures. Science 2016, 353 (6298), aac9439.
(3) Weston, A.; Zou, Y.; Enaldiev, V.; Summerfield, A.; Clark, N.; Zólyomi, V.; Graham, A.; Yelgel, C.; Magorrian, S.; Zhou, M.; et al. Atomic Reconstruction in Twisted Bilayers of Transition Metal Dichalcogenides. Nat. Nanotechnol. 2020, 1–6.
(4) Schwartz, J. J.; Chuang, H. J.; Rosenberger, M. R.; Sivaram, S. V.; McCreary, K. M.; Jonker, B. T.; Centrone, A. Chemical Identification of Interlayer Contaminants within van Der Waals Heterostructures. ACS Appl. Mater. Interfaces 2019, 11 (28), 25578–25585.
(5) Kim, Y.; Herlinger, P.; Taniguchi, T.; Watanabe, K.; Smet, J. H. Reliable Postprocessing Improvement of van Der Waals Heterostructures. ACS Nano 2019, 13 (12), 14182–14190.
(6) Rosenberger, M. R.; Chuang, H. J.; McCreary, K. M.; Hanbicki, A. T.; Sivaram, S. V.; Jonker, B. T. Nano-"Squeegee" for the Creation of Clean 2D Material Interfaces. ACS Appl. Mater. Interfaces 2018, 10 (12), 10379–10387.

Since the excitement about graphene, a monolayer of graphite, with its 2010 Nobel Prize, there has been extensive research in the synthesis of other non-carbon few/mono-layers exhibiting a variety of bandgaps and semiconducting properties (e.g., n or p type). The main approaches to deposit few/monolayers on a substrate are: (a) bottom-up synthesis from precursors using chemical vapor deposition (CVD) or (b) top-down exfoliation (liquid or mechanical) of bulk layered material. Using a Lego approach of superposing monolayers, we can envisage the fabrication of heterojunctions with original electronic behavior.
Here we show a combined bottom-up and top-down approach where (a) we synthesize in one step high yields of bulk layered materials by annealing a metal in the presence of a gas precursor (sulfur, phosphorous, or selenium) using chemical vapor deposition (CVD) and (b) we exfoliate and deposited (dropcast or Langmuir Blodgett) few/mono-layers on a substrate from a sonicated mixture of our material in a specific solvent. It is interesting to note that, besides the structure being 2D layered, the properties of the nanomaterials synthesized slightly differ from the materials with the same stoichiometry synthesized using conventional chemical methods (e.g., solvothermal).
In this talk, we will discuss the chemical synthesis, the very extensive characterizations, and the lessons we learned in making multiple metal sulfides (Cu-S, Ag-S, Ni-S), metal phosphides (Ni-P, Cu-P), and metal selenides (Ag-Se, Cu-Se, W-Se). We will see how we integrated these new materials into electrochemical devices. We will also explore how to deposit large mono/few layers of these materials using techniques such as dropcasting or Langmuir-Blodgett and to connect them to circuits using focused ion beam (FIB) to test their electrical properties with the goal build heterojunctions.

Molybdenum disulfide (MoS₂) is known as promising 2D material with a tunable band-gap which is an ideal for optoelectronic devices. Uniform, crystalline and atomic layers of MoS₂ can be produced by the chemical vapor deposition method. The commonly used MoS₂ CVD system is a system with the substrate facing down toward the molybdenum source. A new kind of CVD system which loads and protects the molybdenum source with a separate quartz tube in order to avoid poisoning of molybdenum trioxide has been designed in recent years. Here we report and compare the strategies of precursor vapor pressure control under this two typical MoS₂ preparation systems. In order to obtain high crystalline MoS₂, we studied the heating temperature of the precursor, the flow rate of the carrier gas, the horizontal distance and the vertical distance between the molybdenum source and the substrate during the growth of MoS₂ in this two different preparation systems. The results indicate the heating temperature of the precursor has the greatest influence on the growth of MoS₂. The lower heating temperature of the precursor contributes to lower growth rate, which makes the thickness of MoS₂ is easily controlled. When the heating temperature is more than 650 °C, the thickness and morphology of MoS₂ differs under the two preparation systems. Nevertheless, the flow rate of carrier gas has little influence on growth of MoS₂, but proper carrier gas flow rate can prevent precursor poisoning. The horizontal distance and the vertical distance between the molybdenum source and the substrate affect the vapor concentration of the molybdenum source and the nucleation density of MoS₂ during the growth of MoS₂.

Modeling of physico-chemical processes such as adsorption-desorption equilibrium, adatom diffusion and attachment are required to correlate the change in supersaturation to the growth kinetics on the surface. We demonstrate an approach to predictively control the grain size and nucleation density of monolayer MoS₂ by employing such correlation via modeling the reaction kinetics in a true CVD system. The adatom concentration, on which this supersaturation depends, on the growth surface, is governed by the reaction kinetics of the precursor and the concentration gradients across the boundary layer at the growth surface. The varied dependencies of the rate of nucleation and edge growth rate on supersaturation and surface kinetics have been exploited to reduce nucleation density from 10⁷ to 10³ cm^-2 and simultaneously increasing the growth rates of the edges to a significant value of 3.3 µm/s. We show that to obtain large grain sizes in small time frames, we identify the physico-chemical windows in which the nucleation kinetics are suppressed but those which govern the growth of edges are enhanced. As in the case of graphene, we observe from the analysis of the rate of nucleations that with an increase in temperature, the availability of adatoms on the growth surface is reduced which in turn results in control of nucleation rate and hence a decrease in nucleation density. We thus simulate the dependence of pressure and temperature on the degree of such pre-reactions to determine the concentration gradients for three substrates placed at different lengths across the reactor tube so as to determine the effect of reaction kinetics on nucleation density. Mass transfer calculations were performed by using the combined convection-diffusion solver. Owing to the vapor phase reactions close to the inlet, the initial precursor concentration was found to reduce from 6x10^-3 to about 2.5x10^-3 for all the cases within about 2 cm from the inlet. The simulations indicate that despite the vapor phase reactions, in the considered pressure and temperature range, the Mo(CO)₆ precursor concentration at substrates placed across the length of the tube varies only by a factor of 2. This is not expected to alter the nucleation density significantly. This is in accordance with the experimental results. The simulation results thus rule out the impact of vapor phase reactions on the observed variations in nucleation density (and grain size) with temperature, pressure, and flow rates. Thus we show that the nucleation density of MoS₂ domains in a CVD process can be controlled in a predictive manner by tuning the supersaturation and surface kinetics.

Doping is fundamental to tune the properties of 2D transition metal dichalcogenides for their future integration in the CMOS architecture, hence achieving substitutional doping during growth is of paramount importance. N-type monolayer WSe₂ is often exfoliated from bulk crystals and a procedure to directly dope the material is yet to be established. Here, we demonstrate the synthesis of monolayer n-WSe₂ via CVD starting from tungstic acid (H₂WO₄) and zinc selenide (ZnSe). We achieved large WSe₂ triangular monolayer flakes, which exhibit n-type of transport which is attribute to donor states due to the incorporation of Zn ions in the lattice. The doping is very stable and the material exhibits good electron mobilities reaching the highest value at 50 cm²V^-1s^-1. The substitutional doping has been confirmed via atomic probe tomography (APT), showing Zn inclusions into the individual WSe₂ layers. Our material is of high crystal quality characterized by a sharp room temperature photoluminescence (PL) assessed by the PL spectra evolution from 10 K to room temperature that shows the contribution of neutral and defect-bound excitons, a signal so far reported only in mechanically exfoliated material.

Layered materials beyond omnipresent graphene such as chalcogenide-based materials that possess tunable and defined crystallographic structures and elemental compositions offer a broad portfolio of potential applications. The lacking control of large scale and homogeneous formation of 2D layered materials MX₂ (X = S, Se) in commercial formation processes motivated us to develop a unique synthetic approach to layered 2D materials. Complete characterization of synthesized complexes enabled their controlled decomposition to desired materials.
Herein we report a uniform synthesis of molecular building blocks to form stable precursor classes [M(SEtN(Me)EtS)_x] (M = Mo^IV, W^IV, Ti^IV, Zr^IV, Hf^IV, Sn^IV, x = 2; M = Ge^II, Sn^II, Pb^II, x = 1), which can be decomposed to layered 2D van der Waal heterostructures.
Following a simple synthetic protocol, the reaction of tridentate (HSEt)₂NMe donor ligand with suitable metal compounds resulted in (air)stable molecular precursors, which were isolated and characterized by NMR, XRPD and single crystal analysis. Thermal decomposition experiments of synthesized metal precursors where analyzed by TG-DTA measurements and the formation of desired MS₂ and MS (W, Mo, Ti, Sn, Pb) materials was confirmed by XRPD analysis. SEM, XRD and TEM analysis of CVD deposited molecular precursors confirmed the formation of homogeneous crystalline layered sulfur based heterostructures. The number of deposited MX₂ layers was calculated from a combination of TEM and AFM measurements. Moreover, prepared precursors were suitable for wet chemical syntheses as demonstrated via microwave assisted decomposition which resulted in homogeneous spherical SnS₂ particles.
These molecular building blocks provide an extraordinary synthetic approach to a unique molecular precursor class, which reliably delivered chalcogenide-based van der Waals heterostructures of metal chalcogenide 2D materials.

Nanoribbons of atomically thin transition metal dichalcogenides (TMDs) provide an additional space confinement in the two-dimensional plane and a more prominent edge effect compared to two-dimensional (2D) sheets, which have a potential to show even more unprecedented properties according to theoretical predictions. Etching from 2D sheets through lithography has realized TMD nanoribbons with a width down to ~30 nm. However, contamination introduced by lithography process can alter or even diminish their properties. Therefore, direct, lithography-free synthesis of atomically thin TMD nanoribbons are highly desired. In this work, we demonstrate for the first time a direct growth of MoS₂ bilayer nanoribbons with a width down to less than 10 nm and a length up to 300 µm through a chemical vapor deposition method using nickel clusters as nucleation seeds. These sub-10 nm bilayer nanoribbons allow us to study not only the quantum confinement and edge effect but also the effect of a second layer on the electronic structures as well as electrical and optical properties of the nanoribbons. Nickel nanoparticle that terminates the tip of each bilayer nanoribbon, serves as a nucleation seed, defines the width, and drives the tip growth mechanism. More insights of the growth mechanism are provided based on XPS, XRD, AES (Auger electron spectroscopy), SEM, and STEM studies. The synthesis method has been successfully applied on a variety of substrates including SiO₂/Si, sapphire, and mica. Our synthesis method introduces a new family of TMD nanoribbons i.e. bilayer nanoribbons, which opens an avenue for discovery of new phenomena and properties.

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 and large bandgap (E_g ∼ 5.9 eV), hBN appears to be a promising material for emerging applications, including deep UV (DUV) optoelectronics, single photon 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 on sapphire using chemical vapor deposition (CVD). However, high quality hBN film growth typically requires high growth temperatures (1100°C-1500°C). It has been reported that this temperature range alters the sapphire surface under ultra-high vacuum (UHV) conditions. In case of chemical vapor deposition (CVD), the presence of hydrogen in the growth ambient is expected to also modify the sapphire surface. However, its impact on hBN growth has not yet been explored.
In this work, we studied the growth of crystalline and epitaxial hBN in a vertical cold wall MOCVD reactor using diborane (B₂H₆) and ammonia (NH₃) as precursors for B and N, respectively using H₂ as a carrier gas. The growths were performed on substrates with different plane directions (C-plane v/s A-plane) at a growth temperature of 1350°C under a pressure of 50 Torr. Sequential pulsing of the B₂H₆ (2 sec) and NH₃ (1 sec) precursors using N/B ratio of ~4000 resulted in a growth rate of ~300 nm/hr due to reduction in the extent of parasitic gas phase reactions. Continuous thin films of hBN were obtained on C-plane as well as A-plane sapphire along with wrinkles due to thermal expansion coefficient (CTE) difference between sapphire and hBN. The E_2g Raman active mode of hBN was observed at 1368.2 cm^-1 with a FWHM of 24.6 cm^-1 on C-plane sapphire. Residual compressive stress due to CTE mismatch between film and the substrate (wrinkles) along with other inhomogeneities in the film (such as substrate’s surface roughness) could be one of the possible reasons for the red shift in the Raman peak position (from bulk crystal Raman peak position at 1365cm^-1) and a broader FWHM. Therefore, the film was transferred on to Si/SiO₂ substrate using PMMA based transfer technique to release residual stress. After film transfer, Raman peak position was measured to be 1365 cm^-1 with a FWHM of 18.5 cm^-1 (FWHM of the best quality bulk crystal obtained is ~7cm^-1). Also, pre-growth H₂ annealing of C-plane sapphire at the growth temperature alters the surface lattice parameter as confirmed by RHEED and is believed to induce a chemical change as well. RHEED pattern of the obtained hBN film suggests rotationally oriented, crystalline hBN film with []_h-BN|| []_sapphire epitaxial relation. However, for hBN on A-plane sapphire, wrinkle density was higher compared to C-plane sapphire. Also, E_2g mode was observed ta 1366.4 cm^-1 (corresponding well with AFM results showing higher wrinkle density indicating lesser residual stress in the film) with a FWHM of 24.8 cm^-1. The cause is still unknown; however, further studies are underway to provide additional insights into the role of crystallographic symmetry and surface preparation to form highly crystalline hBN films on sapphire.

Recently, interests in a family of noble metals-based 2D crystals are on the rise exploiting their superior electrical properties beyond conventional 2D transition metal dichalcogenides (TMDs). 2D platinum (Pt) ditelluride (PtTe₂) layers are particularly promising due to their unique band structure and topological nature as well as remarkably high conductivity (>10⁶ S/m), surpassing most of previously explored 2D TMD layers. Such intrinsic superiority can be leveraged to practical applications by integrating them with matured 3D semiconductors, which has remained largely unexplored. In this report, we explored PtTe₂/silicon hetero-junction devices and unveiled their electrical and opto-electrical applications. We directly grew a large-area (> cm²) highly metallic 2D PtTe₂ layers on doped silicon wafers via a low-temperature chemical vapor deposition (CVD) process, achieving high-quality 2D/3D hetero-interfaces. The integrated devices incorporating the PtTe₂/silicon vertical junctions exhibit excellent Schottky transport characteristics such as high rectification ratio (∼ 10⁵), small ideality factor (∼ 1.9), and Schottky barrier height of 0.67 eV. These devices present strong photovoltaic properties and photo-responsiveness despite their large (0.5 cm²) lateral junction area, i.e., a well-balanced attribute of high photo-detectivity (>10¹³ Jones) and short photo-response time (∼ 1 µs), significantly outperforming most of the previously explored 2D/3D or 2D layers-based devices. Furthermore, we identified that the CVD 2D PtTe₂ layers exhibit unusually strong hydrophobicity which has been utilized to further improve the device photovoltaic performances; i.e., ~ 300 % and ~ 200 % increase of short circuit current density and power conversion efficiency have been achieved respectively upon an integration of water droplet. Underlying principles for the Schottky junction characteristics, as well as the water-assisted performance enhancement will be discussed.

A thorough understanding of the properties of native oxides and the role of oxygen during processing is essential for designing semiconductor devices. This is no less true for nanomaterials than it is for legacy semiconductors such as silicon, for which control of the native oxide was a seminal achievement of 20th century materials science. Being a very hard anion, oxygen bonds very differently to transition metals than do the chalcogens. Trace oxygen in a transition metal dichalcogenide (TMD) has a substantial impact on material processing and properties, much more so than for instance trace selenium in a sulfide. Since oxygen is all around us, even in our high-vacuum chambers, it is essential to understand and control the effects of oxygen on processing 2D materials. Here we report three studies of the effect of oxygen on processing TMDs.
We find that lowering trace oxygen concentration in the reactor makes it possible to lower the processing temperature for large-area TiS₂ films, made by reacting Ti thin films with H₂S gas. We quantify how lowering oxygen concentration enables faster metal sulfurization at lower temperatures (down to 500 °C), leading to thin films that are smooth and homogeneous.
In contrast, we find the opposite trend for MoS₂: adding trace oxygen enables lower processing temperatures (down to 375 °C) for large-area MoS₂ films, made by sulfurizing Mo thin films. We understand these contrasting effects in the light of particulars of Ti-O and Mo-O bonds, including molecular dynamics (MD) simulations that suggest that oxygen is a catalyst for Mo-S bond formation.
We also report a quantitative study of the rate of native oxidation of pristine TMD surfaces. Layered transition metal dichalcogenides nominally have inert, fully-passivated surfaces, but it is well-known that this is an oversimplification and that many TMDs oxidize readily. MoS₂ surfaces remain pristine for over a year in laboratory ambient conditions, without a trace of oxide formation. In contrast, Zr(S,Se)₂ alloys oxidize rapidly, with the native oxide growing at a rate up to 0.5 Å/min. MD simulations reveal the kinetic mechanisms that limit native oxide growth for MoS₂ and promote it for Zr(S,Se)₂, despite oxide formation in ambient conditions being thermodynamically-favorable in all cases.
Our results provide quantitative detail and atomistic understanding of the role of oxygen during TMD synthesis and processing, and may be broadly useful to ongoing efforts at film synthesis and device integration.

A chemical vapor deposition method is developed for uniform and continuous gallium(II) sulfide (GaS) films on SiO₂ featuring both thickness-controlled (one to four layers), as well as defect-controlled (defective GaS_0.87 and stoichiometric GaS). We identify stoichiometry and concentration as two key elements in the growth mechanism. The 2D GaS films show sensitive layer-dependent Raman and optoelectronic properties. Furthermore, an enhanced stability of the stoichiometric GaS is demonstrated under laser and strong UV light, and by controlling defects in GaS, the photoresponse range can be changed from vis-to-UV to UV-discriminating. The stoichiometric GaS is suitable for large-scale, UV-sensitive, high-performance photodetector arrays for information encoding under large vis-light noise, with short response time (<66 ms), excellent UV photoresponsivity (4.7 A W–1 for trilayer GaS), and 26-times increase of signal-to-noise ratio compared with small-bandgap 2D semiconductors. A defect-induced band further appears within the bandgap of GaS_0.87, which can tune the band alignment of bilayer GaS-WS₂ from type-I heterojunction to type-II heterojunction.
We further synthesized GaS crystals on various substrates including sapphire, Si and other 2D materials such as graphene and MoS₂. The morphology of various GaS crystals is influenced by different surface properties. By comprehensive characterizations from atomic-scale structures to large-scale device performances in 2D semiconductors, the study shows that both layer-controlled defective GaS_0.87 and stoichiometric GaS prove to be promising platforms for study of novel phenomena and new applications.

Reference:
[1] Lu, Y., Chen, J., Chen, T., Shu, Y., Chang, R.-J., Sheng, Y., Shautsova, V., Mkhize, N., Holdway, P., Bhaskaran, H., Warner, J. H., Controlling Defects in Continuous 2D GaS Films for High-Performance Wavelength-Tunable UV-Discriminating Photodetectors. Adv. Mater. 2020, 32, 1906958.
[2] Lu, Y., Warner, J. H., et al., Controlled Growth of Monolayer GaS Crystals on Versatile Substrates. In preparation.

Engineered point defects in two-dimensional (2D) materials offer an attractive platform for solid-state devices that exploit tailored opto-electronic, quantum emission, and resistive properties. Naturally occurring defects are also unavoidably important contributors to material properties and performance. The immense variety and complexity of possible defects makes it challenging to experimentally control, probe, or understand atomic-scale defect-property relationships. Here, we develop an approach based on deep transfer learning, machine learning, and first-principles calculations to rapidly predict key properties of point defects in 2D materials. Properties including band gap and bulk formation energy are predicted for over 4,000 2D materials using deep transfer learning. 10,000 dopant, vacancy, divacancy, and antisite defect structures are generated in 150 wide band-gap materials and more than 1,000 defect band structures are computed via first-principles methods. We use physics-informed featurization to generate a minimal description of defect structures and present a general picture of defects across materials systems. We identify over 100 promising, unexplored dopant defect structures in layered metal chalcogenides, hexagonal nitrides, and metal halides including GeS, h-AlN, and MgI₂. These defects have low formation energies and host material band-gaps calculated to be greater than 2 eV. We also find ten optimal substitutional defects for nonvolatile resistive switching in atomically thin memristor devices, where metal dopants substitute for chalcogen atoms in materials including WSe₂ and MoTe₂. Au and Ag substitutions in WTe₂- and MoTe₂-based devices are predicted to have switching voltages as low as 110 meV. These defect structures are prime candidates for quantum emission, resistive switching, and neuromorphic computing.

Two-dimensional materials offer a wide range of perspectives for hosting functional and highly localized zero-dimensional states, e.g. vacancy defects, that offer great potential for nano-photonic applications [1]. The site-selective generation of such defects and the fundamental understanding of their optical and electronic properties are of key importance for integrated quantum photonic devices.
In this talk, we will demonstrate the highly local formation of individual single-photon sources in a monolayer MoS₂ van der Waals heterostructure [2,3]. Individual defects that act as our single-photon emitters are created by He-ion irradiation. Irradiation after encapsulation of the MoS₂ with hBN reveals spectrally narrow and spatially localized spectral lines that exhibit anti-bunched emission at the defect characteristic energy of ~1.75 eV (194 meV redshifted from 2D exciton). The emission is highly homogeneous and background-free due to the encapsulation [4]. We investigate these individual single-photon emitters spectroscopically using photoluminescence excitation spectroscopy and temperature-dependent measurements. The line shape reveals a strong asymmetry resembling the interaction with LA/TA phonons. Applying the independent boson model to our emission lines, we find that the emitters are spatially localized to a length scale of 2 nm. We attribute the emission to atomistic defects induced by the He-ion irradiation and discuss their microscopic origin in the light of scanning tunneling microscopy measurements [5,6]. Moreover, we further demonstrate spatial integration and electrical control of single-photon sources in field-switchable van der Waals devices. Our work firmly establishes 2D materials as a highly scalable material platform for integrated quantum photonics.
References:
[1] J. Klein et al., 2D Mater. 5, 1 (2017)
[2] J. Klein et al., Nature Comm. 10, 2755 (2019)
[3] J. Klein and L. Sigl et al., submitted (2020)
[4] J. Wierzbowski and J. Klein et al., Sci. Rep. 7, 12383 (2017)
[5] J. Klein et al., Appl. Phys. Lett. 115, 261603 (2019)
[6] E. Mitterreiter et al., Nano Lett. 20, 4437 (2020)

Due to intrinsic defects, transition metal dichalcogenides (TMD) monolayers have low (< 0.1%) photoluminescence quantum yield (PLQY), hindering their potential for optoelectronic device applications. Here, we report a passivation method of monolayer TMD using organic/transition metal oxide (TMO) mixtures with laser soaking. A 10 nm mixed layer of 3,3',5,5'-Tetra[(m-pyridyl)-phen-3-yl]biphenyl (BP4mPy) and MoO₃ is vacuum deposited onto a MoS₂ monolayer, which is followed by laser soaking for tens of minutes in ambient. For 1:4 vol% BP4mPy:MoO₃, the passivated MoS₂ monolayer shows up to 250 times enhancement in PL intensity as compared to exfoliated MoS₂. The PLQY reaches ~ 5 % at excitation power density < 10 W/cm² and rolls off as power density increases, reaching 1 % at 10³ W/cm². These values are comparable to those achieved by recently reported superacid-treated¹ samples with PLQY reaching 3 % at 10 W/cm² and 1 % at 10² W/cm². In addition, the passivated MoS₂ is stable in air, vacuum, and solvents (e.g. acetone, isopropanol). Using low temperature PL and transient spectroscopy, we find that the mid-gap defect states of MoS₂ are eliminated by passivation, concomitant with a decrease of exciton-exciton annihilation.

Mechanisms of passivation are studied by varying the photon energy of the soaking laser, and as a function of the structures: 5 nm BP4mPy/d nm MoO₃/MoS₂, where d is varied from 3 to 12nm. The effectiveness of the passivation is found to be determined by the polaron-pair generation at the BP4mPy/MoO₃ interface and polaron-diffusion efficiency from the interface to MoS₂. This passivation method has been generalized to other organic/TMO capping layers, where organic molecules serve as the electron donor while TMO as the acceptor. It is also applicable to passivating other sulfur-based TMDs (e.g. WS₂), offering a potential platform for next generation 2D device applications.

This work was supported in part by the Army Research Office (ARO) under Award Number W911NF-17-1-0312 and Air Force Office of Scientific Research (AFOSR) Award Number FA9550-17-1-0208

Xiaheng Huang and Zidong Li contributed equally to this work.

[1] Amani, Matin, et al. "Near-unity photoluminescence quantum yield in MoS₂." Science 350.6264 (2015): 1065-1068.

Two-dimensional (2D) semiconductors, such as MoS₂, are promising for electronics due to their good mobility in ultrathin channels. However, current synthesis methods result in defective films which hinder their electrical performance and can influence doping and reactivity. Therefore, it is important to have a quick and nondestructive method to analyze defect densities in MoS₂.
Past studies used transmission electron microscopy [1] or scanning tunneling microscopy [2] to probe MoS₂ defects, but these methods are time consuming, destructive, and cannot be applied to large-area films. On the other hand, Raman spectroscopy has been used to probe graphene defects, particularly through the D-peak [3]. A similar Raman feature exists in MoS₂ as the LA(M) peak, however this only appears at very large defect densities (>10¹³ cm^-2) [4].
Here, we use Raman spectroscopy for rapid, large-area analysis of MoS₂ defect densities in the range of interest for electronic applications. We use a simple Raman peak fitting procedure as well as the phonon confinement (RWL) model [4-6] to compare disorder in monolayer MoS₂ prepared by exfoliation, chemical vapor deposition (CVD), and atomic layer deposition (ALD). Importantly, we uncover a correlation between Raman-based defect metrics and electrical mobility of MoS₂, for the first time.
Non-resonant MoS₂ Raman spectra (532 nm laser) typically have two peaks, E’ and A₁’. However, the E’ peak is comprised of two optical phonon modes (LO and TO) which almost overlap at the Gamma point. Thus, additional information on defects can be gained by fitting two E’ peaks, including the E’ shoulder. We use this approach and observe subtle changes in the Raman spectra of MoS₂ prepared by different methods, i.e. peak broadening, shifting, and increased E’ peak shoulder intensity, which correlate with increased defect densities [4,6]. Based on these, we find that exfoliated monolayer MoS₂ is the least defective, followed by CVD-grown, and ALD-grown MoS₂ is the most defective material.
To quantitatively understand how defects change the MoS₂ Raman spectra, we use the RWL model [4-6], which posits that defects spatially confine Gamma point phonons, broadening their k-space participation and thus broadening the Raman peak. Non-Gamma point phonons which participate depend on the phonon dispersion, and a steeper dispersion near the Gamma point causes increased Raman peak broadening and asymmetry due to defects. Here we used a fit to the MoS₂ phonon dispersion based on [7].
By comparing Raman data across various MoS₂ samples with the RWL model, we find that as the defect density increases, the E’ peak (with LO and TO modes) shifts and asymmetrically broadens. Fitting this model to Raman data from various MoS₂ sources, we can estimate the defect density in each sample and relate it to the relative E’ peak positions and intensities. Importantly, we correlate these changes in Raman features with electrical mobility data for various samples from our lab and from the literature, providing a rapid evaluation of the quality of a 2D semiconductor film. This is an important advance, which allows us to distinguish between samples with, e.g. 10 and 30 cm²/V/s mobility using Raman analysis, without the need for extensive fabrication and electrical measurements. Our approach could also be extended to other 2D semiconductors, in particular WSe₂, which also has a steep phonon dispersion near the Gamma point.
[1] J. Hong et al., Nat. Comm. 6, 6293 (2015)
[2] P. Vancso et al., Sci. Rep. 6, 29726 (2016)
[3] A. Ferrari and D. Basko, Nat. Nano 8, 235 (2013)
[4] S. Mignuzzi et al., Phys. Rev. B 91, 195411 (2015)
[5] H. Richter et al., Solid State Comm. 39, 625 (1981)
[6] W. Shi et al., Chin. Phys. Lett. 33, 057801 (2016)
[7] H. Tornatzky et al., Phys. Rev. B 99, 144309 (2019)

Electric double layer (EDL) gating using a single-ion conductor is compared to a dual-ion conductor using both finite element modeling and experimentally on two-dimensional (2D) crystals based field effect transistors (FETs). A modified version of the Nernst-Planck Poisson (mNPP) equations are used to calculate the ion density per unit area in a parallel plate capacitor geometry with a bulk ion concentration 215 ≤ c_bulk ≤ 1650 mol/m³. With electrodes of equal size at a 2 V potential difference, the EDL ion density of the single-ion conductor is ~ 7 x 10¹³ ions/cm², which is approximately 50% of the ion density induced in the dual-ion conductor. However, this difference is reduced to 8% when the electrode at which the cationic EDL forms is 10 times smaller than the anionic electrode. Thus, for a FET gated by a single-ion conductor, it is especially important to have a large gate-to-channel size ratio to achieve strong ion doping. I will also show that the numerical modeling results can be approximated by a simple analysis of capacitors in series, where the EDLs are modeled as capacitors with thickness estimated by the sum of the Debye screening length and the Stern Layer. The modeled ion densities are validated by Hall-effect measurements on graphene FETs gated by a polyethylene oxide (PEO)-based single-ion conductor. The sheet carrier density (n_S), which is reported for the first time for a single-ion conductor, is ~ 2 x 10¹³ cm^-2 at V_g = 2 V, which is 3.5 times smaller than the predicted value, and on the same order of magnitude as the n_s measured for a PEO-based dual-ion conductor on the same graphene. Our results suggest single-ion conductor-gated 2D FETs can achieve n_S up to 10¹⁴ cm^-2, which is theoretically predicted to be able to induce sufficient strain required to achieve the semiconducting 2H to metallic 1T phase transition in MoTe₂. Therefore, in the last part I will discuss the ongoing work focused on the fabrication of single-ion conductor-gated MoTe₂ FETs with suspended channels, and the in-situ characterization of the phase transition induced by the strain via the bending of single-ion conductor under an applied field.

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

Electrostatic doping by ions relies on the formation of an electric double layer (EDL) at the electrolyte/channel interface where the image charges induced by the ions serve as dopants in the channel. Electrostatic doping can induce sheet carrier densities in two-dimensional (2D) crystals exceeding 10¹⁴ cm^-2 without altering the intrinsic bandgap, and the doping is reconfigurable between p- and n-type or p-n/n-p junctions by redistributing ions under the field. However, in conventional electrolytes such as ionic liquids/gels, a gate voltage must be continuously applied to hold the ions in place and therefore set the doping state. What would be useful is a “gateless” electrostatic doping wherein the doping could be programmed by ions and then locked into position via a trigger until the ions could be unlocked and reconfigured again on demand. Herein, we report a newly developed doubly polymerizable ionic liquid (DPIL) that can perform ion-locking by a chemical trigger to achieve the gateless electrostatic doping. The structure of DPIL is similar to the ionic liquid, [EMIM][TFSI], but with polymerizable functional groups on both charged species to perform ion-locking. Before polymerization, the DPIL monomers behave as a typical electrolyte with mobile ions that can be drifted by an applied field to form EDLs. To lock the ions, and therefore the EDL into place, either UV or thermal initiated polymerization is used. Once locked, the electric field is removed and EDL does not dissipate. Here, ion locking to create a gateless, lateral p-n junction is demonstrated on graphene.¹ The junction is set by an applied field and DPIL is polymerized to lock the doping; the junction is measured at room temperature. The results are compared to a common locking approach: thermally quenching a polyethylene oxide-based electrolyte to a temperature below the glass transition temperature (T_g) and then measuring the junction at low temperature. Both approaches require only the source/drain terminals to create the junction, and a gate is not required to maintain the doping after ion-locking. Backgated transfer characteristics of both approaches show two current minima – the signature of a p-n junction in graphene. We have also made progress towards reconfigurable doping (i.e., locking and unlocking) by incorporating thermally-labile linkers into DPIL to allow ions to be released again after polymerization.²


[1] Liang, J., Xu, K., Arora, S., Laaser, J. and Fullerton-Shirey, S. "Ion-Locking in Solid Polymer Electrolytes for Reconfigurable Gateless Lateral Graphene p-n Junctions." Materials 13.5 (2020): 1089.
[2] Arora, S., Liang, J., Fullerton-Shirey, S. and Laaser, J. "Triggerable Ion Release in Polymerized Ionic Liquids Containing Thermally-Labile Diels-Alder Linkages." ACS Materials Letters (2020).

Surface functionalization with organic electron donors (OEDs) is an effective doping strategy for two-dimensional (2D) materials, which can achieve doping levels beyond those possible with conventional electric field gating. While the effectiveness of surface functionalization has been demonstrated in many 2D systems, the doping efficiencies of OEDs have largely been unmeasured, which is in stark contrast to their precision syntheses and tailored redox potentials. Here, using monolayer MoS₂ as a model system and an organic reductant based on 4,4’-bipyridine (DMAP-OED) as a strong organic dopant, we establish that the doping efficiency of DMAP-OED to MoS₂ is in the range of 0.63 to 1.26 electrons per molecule. We also achieve the highest doping level to date in monolayer MoS₂ with a carrier density of 5.8 ± 1.9 × 10¹³ cm^-2 by surface functionalization with DMAP-OED at the maximum functionalization conditions. This is four times greater than the previous best system based on benzyl viologen, demonstrating that DMAP-OED is a stronger dopant than benzyl viologen. The measured range of the doping efficiency is in good agreement with the values predicted from first-principles calculations.
We show that multiple factors, such as reduction potential, size, and binding mode of the dopant molecule, as well as interactions between the molecules and 2D materials and among the molecules themselves play an important role when considering the design of strong molecular dopants. The ability to dope MoS₂ beyond its degenerate limit via an organic super electron donor provides opportunities to access correlated electronic phases, such as superconductivity at the degenerate level, and offers an alternative to ionic gating and lithium intercalation for carrier density control, which may produce unwanted electrochemical reactions. Our work provides a basis for the rational design of OEDs for high-level doping of 2D materials.

Transition metal dichalcogenides (TMDCs) has attracted great attention as a promising semiconductor of next generation nanoelectronics due to unique properties such as atomically flat nature and tunable band gap property. To utilize TMDCs for complementary logic circuit and various homo/heterjuction devices, controlling of the conductivity and the carrier type is necessary. In order to modulate carrier concentration and carrier type, surface charge transfer doping (SCTD) on TMDCs using functionalized organic molecules has been extensively studied. However, the molecular dopant could limit the electrical performance of TMDCs like in graphene [1] leading to charged impurity scattering. In this regard, a thorough understanding of the effect of SCTD on the charge transport of TMDCs is essential.
In this presentation, we report the effect of surface charge transfer doping on the transport properties of tungsten diselenide (WSe₂) by using three types of p-type molecular dopants, 2,3,5,6-tetrafluoro-7,7,8,8-tetracyanoquinodimethane (denoted as F₄-TCNQ), molybdenum tris(1,2-bis(trifluoromethyl)ethane-1,2-dithiolene) (denoted as Mo(tfd-COCF)₃), and Tris(4-bromophenyl)ammoniumyl hexachloroantimonate (denoted as magic blue). These strong molecular acceptors have different electron affinities, and thus the doping strength among the molecular dopants vary significantly, which we characterized with WSe₂ field-effect transistors (FETs). In particular, we measured four-point probe mobility at variable temperature from 10 K to 300 K to investigate the doping effect on the intrinsic charge transport of WSe₂. Comparing to pristine WSe₂, the mobility of molecular doped WSe₂ FETs is significantly reduced at low temperature regime. This result implies that the dopants can induce impurity scattering in the transport of WSe₂ FETs. It is found that magic blue molecule induces the least impurity scattering in WSe₂ FETs among the three molecular dopants. This study can help developing suitable molecular dopants for controlling the electrical properties of TMDC devices by understanding dopant-induced effects on the charge transport.

[1] Chen et al. Nat. Phys. 4, 377 (2008)

Doping of MoS₂ can introduce novel functionality in terms of electronic, optical, magnetic, or mechanical properties. Experimental resolution of the doped structures have been challenging and often inconclusive, showing an important role for theory. In this work we study Ni-doped MoS₂, motivated specifically by its exploration as a solid lubricant in the low-temperature and low-pressure environment in space, where liquid lubricants fail [MR Vazirisereshk, A Martini, DA Strubbe, and MZ Baykara, Lubricants 7, 57 (2019)]. Some preliminary studies suggest that doping with transition metals (especially Ni) can improve tribology performance by reducing friction and wear, but the structure and mechanisms remain unclear. We use density functional theory (DFT) calculations to determine the structure and properties of Ni-doped MoS₂ in bulk and monolayer form, in various polytypes and doping levels. We consider formation energies for different sites for Mo or S substitution, intercalation, or adsorption, to create a phase diagram in chemical potential space to inform synthesis efforts. Our calculations identify structural changes on doping and which are the most stable structures. Stable Ni-doped MoS₂ phases and cases where phase segregation is favorable are identified. We compute infrared and Raman spectra to enable experimental identification of the doped structures coming from various growth methods. We additionally calculate elastic properties and potential energy for sliding to connect to atomic-force microscopy friction experiments, and use our DFT data to parametrize classical force fields for larger-scale reactive molecular dynamics simulations that can directly address friction and wear on larger length and time scales.

Doping provides a key scheme for tuning materials’ functional properties. The atomic-scale control of the position, distribution, and concentration of dopants is crucially important in engineering the material’s behavior. Two-dimensional (2D) crystals provide an extremely rich materials platform to study the behaviors of dopant atoms in different lattices and their impacts on material’s properties. While there has been remarkable progress in the synthesis and atomic-scale characterization of doped 2D transition metal dichalcogenides (TMDs), most studies have focused on isotropic TMDs with a hexagonal structure. However, anisotropic TMD material systems with a 1T’ structure offer a more versatile platform to study the behaviors of dopant atoms. For example, while there is only one atomic site for metal atoms in hexagonal TMDs with the 1H structure, there are two different atomic sites for metal atoms in anisotropic TMDs with the 1T’ structure. Here, we dope an anisotropic TMD crystal and probe individual and coupled dopant atoms in its lattice using aberration-corrected scanning transmission electron microscopy (STEM) with sub-Ångstrom resolution. Statistical analyses on the atomically-resolved images reveal the preference of dopant atoms for specific atomic sites. Using density functional theory calculations, we develop a model to describe the preferred distribution of atomic species.

In recent years, van der Waals layered materials, especially their two-dimensional (2D) forms (monolayers, bilayers, and few-layers), have attracted significant attention as novel high-performance structures for a variety of applications, including thermoelectric devices, e.g. structures that are based on efficient conversion between heat and electricity. Thermoelectric properties and the performance of different 2D materials, such as SnSe, Bi₂Te₃, and MoS₂, have been studied extensively. Owing to its unique properties, such as high carrier mobility (1000 cm²(Vs)⁻¹ at room temperature), a tunable thickens-dependent bandgap (0.3–2.0 eV), a large Seebeck coefficient (335 μV K⁻¹ at room temperature), and a strong in-plane anisotropy, black phosphorus (BP) and its 2D form, phosphorene, are among the most prospective thermoelectric materials. Recent investigations have indicated that the thermoelectric properties of BP can be further enhanced by alloying or doping. For example, a very high ZT value of 6.0 has been predicted by first-principle calculations in black phosphorous with 1/4 of its P atoms substituted with Sb atoms. Here, motivated by these works, we have synthesized a series of As_xP_1-x alloys and conducted a comparative study of their structural and thermoelectric properties. Raman measurements of these samples show the presence of 3 types of atomic vibrational modes arising from As-As bonding (< 270 cm^-1), As-P bonding (270 cm^-1 - 350 cm^-1), and P-P bonding (> 350 cm^-1), respectively. For a wide range of As concentration, these materials are found to crystallize in the orthorhombic puckered structure and are alloy analogs of BP. Furthermore, it is found that the crystal structure of these As_xP_1-x alloys expands and their vibrational modes red-shift systematically with the increase of As. These changes are accompanied by the systematic changes of the electrical transport properties, especially the thermoelectric power. Temperature-dependent transport measurements show that, at room temperature, the samples with lower and intermediate As concertation exhibit very high thermopower (as high as 803 μV/K for x = 0.2 compared to 323 μV/K for BP) and high power factor (as high as 317 μWm^-1K^-2 for x = 0.5 compared to 22 μWm^-1K^-2 for BP). Overall, the study suggests that the thermoelectric properties of BP can be greatly improved by alloying, resulting in comparable or even better performance than some of the reported outstanding thermoelectric materials. Additional benefits of As_xP_1-x alloys include enhanced durability and higher resistance against degradation than those of BP. Finally, their unique in-plane anisotropy provides an opportunity for decoupling thermal and electronic transport pathways and therefore further enhancing thermoelectric properties.

Functional van der Waals layered materials, which exhibit interesting phenomena such as magnetism and ferroelectricity, have been proposed for use in next-generation nanoscale devices and sensors. Metal thiophosphates are an interesting class of 2D van der Waals gapped materials that contain a common structural framework where altering the cation composition can induce different types of ferroic ordering, including ferroelectricity and magnetism. making this family arguably the 2D equivalent od complex oxides.The compound CuCrP₂S₆ is a promising 2D material that evinces multiferroic behavior where the Cu⁺ and Cr⁺³ cations are responsible for antiferroelectric and antiferromagnetic ordering, respectively, which are theoretically predicted to couple. In this study, we use X-ray diffraction and Raman spectroscopy to map out these phase transitions in the range 70-400 K. The antiferroelectric phase transition is complex and shows a gradual transition to complete antipolar order with an intermediate quasi-antipolar step. X-ray diffraction studies reveal evidence for negative thermal expansion across the antipolar phase transition around 140 K. This is accompanied by a drastic reduction in rotational and translational mode frequencies of the anion groups in CuCrP₂S₆. Our temperature-dependent structural data provides an important reference for subsequent research into this promising 2D multiferroic material.

Understanding metastability and phase transitions of 2D materials requires long and large-scalemolecular dynamics simulations, but this is hindered by lack of accurate force fields capable ofdescribing the vast variety of configurations and phases. Machine learning teachniques are attractivefor developing force field models because these powerful function-fitting schemes are capable ofreaching first principles accuracy at a small fraction of computational cost. Gaussian process (GP)regression is one promising technique among them, whose remarkable competitive advantage is thebuilt-in uncertainty quantification based on Bayesian posterior inference, which can be used tomonitor the quality of model predictions. A current limitation of existing GP force fields is that theprediction cost grows linearly with the size of the training data set, making accurate GP predictionsslow. In this work, we exploit the special structure of the kernel function to construct a mapping ofthe trained Gaussian process model, including both forces and their uncertainty predictions, ontoa spline functions of low-dimensional structural features. This method extends previous proposalsof mapping only the forces by incorporating uncertainty at constant scaling cost, accelerating theactive learning workflow for training of Bayesian force fields. To demonstrate the capabilities of thismethod, we construct a force field for 2D stanene and perform large scale dynamics simulation ofits structural evolution as a function of termperature. We discover an unusual phase transformationmechanism of stanene, where the 2D metastable structure distorts, buckles and transforms into 3Dtin or a liquid phase depending on the temperature. By monitoring the uncertainty of the predictionwe are able to systematically improve the force field during the simulation, demonstrating a pathto autonomous learning of fast models for studying phase transformations of complex systems

MoTe2 has received much attention due to its unique physical properties that make it good candidate for spintronics, thermoelectric and piezoelectric applications. MoTe2 is stable in both the 2H and 1T` phase. In general, the 1T`phase displays metallic behavior, making MoTe2 also interesting candidate for in-plane interconnects when combined with 2D semiconductors. In this study, we successfully synthesize large-area few-layers MoTe2 using a simple tellurization CVD process. Subsequently, the as-grown transition metal ditelluride was chemically modified using an in situ laser-assisted method recently developed in our group. The entire process takes place within a home-made mini chamber with an optical quartz window, where the samples are exposed to a laser beam in the presence of a controlled environment rich in different chalcogen atoms. For an optimized set of parameters (e.g. exposure time and laser power), the tellurium atoms are replaced by sulfur atoms. The chemical exchange process is monitored in real time via Raman spectroscopy. The site-selective replacement of the chalcogen atoms generate metal-semiconductor heterojunctions. 2D field effect transistors were fabricated to study the electrical response of these lateral heterojunctions. Additionally, we performed first-principle calculations for different chalcogen compositions in the ternary system MoTe2(1-x)S2x and found that the band structure evolves with increasing sulfur composition revealing a systematic transition from metal to semiconductor.

Blue phosphorene, a new allotrope of 2D phosphorus material, has been theoretically predicted to be energetically as stable as black phosphorene with energy difference less than 2meV. It has a wide fundamental bandgap (> 2 eV) compared to black phosphorene, tunable gap depending on the number of layers, semiconducting-semimetal transition under strain, possible high carrier mobility, higher in-plane rigidity, and superconductivity with metal intercalation, which would become a worthy contender in the emerging field of post-graphene 2D electronics, such as photodetectors, transistors, and other novel devices. The next task is to synthesize the newly predicted blue phosphorene and to utilize its unique properties in the next generation of nanolithography devices. However, it is a big challenge to simply exfoliate blue phosphorene from its bulk counterpart since blue phosphorus can only exist at high pressure (>5GPa). Alternatively, recent efforts by epitaxial growth method have produced a single layer of blue phosphorus, but it strongly depends on the surface structures of Au, Al. and Te substrates. Therefore, it is extremely desirable if we can develop a new pathway to produce freestanding blue phosphorene. Here, we proposed a novel kinetic pathway to grow layered blue phosphorene from layered black phosphorene via Li intercalation, based on first principle study. This study points out that Li atoms could play as ‘catalysts’ to drive P atoms in the ‘reactive region’ undergoing a bond-breaking process and lead to a local structural transformation from orthorhombic lattice to an assembly of parallel narrow nanoribbons with rhombohedra-like symmetry. After delithiation, these nanoribbons are self-mended and a layered blue phosphorene is formed. The interlayer distance of the obtained blue phosphorene layers was given by ~ 3.56 Å, indicating a monolayer blue phosphorene could be obtained by mechanical exfoliation. This theoretical study not only reveals the novel transition mechanism from layered black phosphorene to layered blue phosphorene induced by Li interaction, but also provides a fundamental guidance for synthesizing blue phosphorene.

Atomically-thin (2D) semiconductors have shown great potential as the fundamental building blocks for next-generation electronics. However, all the 2D semiconductors that have been experimentally made so far have room-temperature electron mobility lower than that of bulk silicon. Although materials-specific explanations have been made [1,2], the "universality" of the low mobility is not well understood. Here, by using first-principles calculations and re-formulating the transport equations, we show that the universally low mobility of 2D semiconductors originates from the large “density of scatterings”, which is intrinsic to 2D system. The “density of scatterings” characterizes the number of phonons that can interact with the electrons, and can be fully determined from the electron and phonon band structures without knowledge of electron-phonon coupling strength. Our work reveals the underlying physics limiting the electron mobility of 2D semiconductors, and offers a descriptor to quickly assess the mobility.[3]
[1] L. Cheng, C. Zhang, Y. Liu*
“The Optimal Electronic Structure for High-Mobility 2D Semiconductors: Exceptionally High Hole Mobility in 2D Antimony”
J. Am. Chem. Soc., 2019, DOI: 10.1021/jacs.9b05923
[2] L. Cheng and Y. Liu*
“What Limits the Intrinsic Mobility of Electrons and Holes in Two Dimensional Metal Dichalcogenides?”
J. Am. Chem. Soc., 2018, 140, 17895-17900
[3]. L. Cheng, C. Zhang, Y. Liu, in preparation.

Ion beams in high-resolution imaging and patterning techniques can be used efficiently to manipulate and characterize 2D materials. Achieving higher precision control of the structure and properties necessitates a detailed understanding of the excited electron dynamics occurring in the material in response to the irradiation. Unfortunately, most existing knowledge concerning an energetic ion’s interaction with a material is limited to bulk. While many real-life radiation effects, such as electron emission, inherently involve surface processes also for thick materials, the case is particularly compelling for ion-irradiated 2D materials: Their myriad potential applications oftentimes rely on precise control of defects that can be achieved with ion irradiation. In this context, first-principles simulations are an ideal tool to characterize the elusive fundamental physics of ion-irradiated surfaces and inform ion beam microscopy and patterning techniques for 2D materials.

Here we discuss our recent application of real-time time-dependent density functional theory simulations to two examples: (i) As an ion approaches and traverses a surface, it is expected to exchange charge with the material and transition into a steady-state. We model this pre-equilibrium behavior for protons with 0.5 - 1.5 atomic units of velocity impacting few-layer aluminum sheets. This allows us to interpret an enhancement of energy deposition in the sheet, observed from our simulation results, as being tied to surface plasmons and we derive electron escape depths with important implications for optimization of focused ion beam techniques. (ii) We study energy transfer, charge induced, and electrons emitted for monolayer and trilayer graphene as H⁺, He²⁺, Si⁴⁺, Si¹²⁺, and Xe⁸⁺ ions with velocities of 0.5 - 4.6 au traverse the material. In particular, we study the emerging excited charge dynamics and its dependence on projectile velocity, trajectory, species, and charge as well as graphene thickness, with important implications for defect formation through particle radiation.

Transition metal dichalcogenides (TMDC) are known to show great potential for next-generation electric and optoelectronic devices. In TMDC family, ZrS₂ has shown a superior electrical properties and chemical catalytic properties. However, native oxidation remains a major issue to improve its long-term stability and thus an active area of research. Here, we develop first-principles informed reactive force field for Zr/O/S and perform reactive molecular dynamics simulations to study oxidation of ZrS₂ and MoS₂ to elucidate the reaction mechanisms of oxidation process. Our results show a significantly larger oxidation rate for ZrS₂ in comparison with MoS₂. We find that the oxidation pathway of bulk ZrS₂ involves adsorption of oxygen on ZrS₂ surface, followed by amorphization and oxygen transport into bulk, leading to a continuous oxidation. In contrast, bulk MoS₂ remains inert due to the absence of oxygen transport into bulk. Our results show a valuable insight into the atomistic-scale mechanisms of native oxide formation and provide guidance for processing devices based on MoS₂ and ZrS₂.

This work was supported as part of the Computational Materials Sciences Program funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award Number DE-SC0014607. Computations were performed at the Argonne Leadership Computing facility under the DOE INCITE and Aurora Early Science programs and at the Center for High Performance Computing of the University of Southern California.

Although common in the periodic table, monoelemental metals appear severely under-represented in known families of 2D materials. Their recent realization in air-stable 2D form (e.g. Ga, In, and Sn), epitaxial to SiC and capped by graphene [1], suggests an immense chemical space of possible 2D metals and alloys that could expand the physics of 2D metals beyond what is known for graphene and Nb/Ta chalcogenides, including unusual nonlinear optical response [2] and 2D superconductivity with doped-semiconductor characters. Here we present a first-principles computational survey predicting the favorabilities of all metals in the periodic table in intercalating graphene/SiC. All experimentally known metal intercalant structures agree with predicted 2D metal ground states. This stability survey not only reveals conspicuous trends related to metal cohesive energies and metal-silicon bonding, but also identifies special groups of metals stabilized by opening a bandgap. We discuss conditions required for gapping, including electron counts and in-plane mirror symmetry breaking. From this gapping stabilization, alloying rules unique to 2D metals are derived.

[1] N. Briggs, Nat. Mater. 19, 637 (2020)
[2] M. A. Steves, arXiv:2004.01809 (2020)

The design of 2D materials and thin-film devices requires the ability to grow high-quality films consistently and systematically. The selection of substrates is critical to this goal, as a poor match will result in undesired defects and other difficulties during synthesis. To systematically identify potential substrate materials, we apply a topological algorithm and data mining approach to the Materials Project database in search of low-energy surfaces that can act as substrates. Inspired by the recent high-throughput searches for exfoliable 2D materials, we develop a topological algorithm to cleave bonds in fully periodic crystals. If the bond cleaving generates a surface, we isolate the cleaved monolayer and calculate the energy needed to break these bonds. This algorithm is applied to all metastable crystals in the MaterialsProject database with five or fewer atoms in the primitive cell, identifying 4,703 unique surfaces across 2,136 crystals. We compare the cleavage energy to commonly used substrates to identify over 3,000 surfaces with cleavage energy below (0001) AlN. We follow by characterizing trends between cleavage energy and the precursor phase bandgap. We summarize the breadth of materials using the precursor bandgaps and lattice parameters of cleaved surfaces, establishing a computational substrate database. We test the applicability of this database by searching for epitaxial matches to (100) BaSnO₃, identifying several matches including (001) Rb₂NiO₂, (001) NiO, and (001) CaSe. We finalize by looking forward to ways to improve the computational prediction of thin-film and substrate interactions. We make this data freely available at MaterialsWeb.org and the software used to cleave the surfaces available in the open-source software package MPInterfaces.

The discovery of 2-dimensional (2D) materials, such as CrI3, that retain magnetic ordering at monolayer thickness has resulted in a surge of both pure and applied research in 2D magnetism. In this talk, I will detail our magneto-Raman spectroscopy study on multilayered CrI3, focusing on two additional features in the spectra that appear below the magnetic ordering temperature and were previously assigned to high frequency magnons. Instead, we conclude these modes are actually zone-folded phonons. We observe a striking evolution of the Raman spectra with increasing magnetic field applied perpendicular to the atomic layers in which clear, sudden changes in intensities of the modes and their polarization dependence are attributed to the interlayer ordering changing from antiferromagnetic to ferromagnetic at a critical magnetic field. Our work highlights the sensitivity of the Raman modes to weak interlayer spin ordering in CrI3.

Two-dimensional magnetic van der Waals materials have attracted much interest recently. Magnetism in low dimensional systems is an interesting topic for the fundamental physics, and atomically thin magnetic materials are promising candidates for novel spintronic devices. Antiferromagnetic 2-dimensional materials are particularly interesting both for fundamental physics and also for antiferromagnetism-based spintronic devices. However, traditional research tools such as neutron scattering to probe antiferromagnetic ordering cannot be employed for atomically thin materials due to the small sample volume. Although magneto-optical Kerr effect measurements can be used to monitor the magnetic ordering in ferromagnetic materials, the lack of net magnetization precludes the use of the Kerr effect in probing antiferromagnetic ordering. Optical spectroscopy is becoming increasingly important for the study of antiferromagnetic 2-dimensional materials. Raman spectroscopy, for example, has been established as an invaluable tool to probe the magnetic transition in antiferromagnetic van der Waals materials as it has been found that the magnetic ordering can be reliably monitored by Raman spectroscopy [1]. Furthermore, recent spectroscopic studies revealed a novel coherent state in some of these materials stabilized by the antiferromagnetic ordering. In this presentation, I will review recent achievements in the study of antiferromagnetism in 2 dimensions using optical spectroscopy. FePS₃ exhibits an Ising-type antiferromagnetic ordering down to the monolayer limit, in good agreement with the Onsager solution for a 2-dimensional order-disorder transition [2]. The transition temperature remains almost independent of the thickness from bulk to the monolayer limit, indicating that the weak interlayer interaction has little effect on the antiferromagnetic ordering. On the other hand, NiPS₃, which shows an XXZ-type antiferromagnetic ordering in bulk, exhibits antiferromagnetic ordering down to 2 layers with a slight decrease in the transition temperature, but the magnetic ordering is suppressed in the monolayer limit [3]. Furthermore, an almost resolution-limited, sharp optical transition was observed in photoluminescence and optical absorption measurements on NiPS₃, which is interpreted as being due to a novel excitonic state coupled with the antiferromagnetic ordering [4]. Furthermore, a Heisenberg-type antiferromagnet MnPS₃ also exhibits ordering down to 2 layers [5] with a weak dependence of the Néel temperature on the thickness [6].



References
[1] Kim K, Lee J-U, Cheong H, Nanotechnology (2019); 30, 452001.
[2] Lee J-U, et al., Nano Letters (2016); 16, 7433.
[3] Kim K, et al., Nature Communications (2019); 10, 345.
[4] Kang S, et al., in press.
[5] Kim K, et al., 2D Materials (2019); 6, 041001.
[6] Lim S, et al., submitted.

Electrical manipulation of spin in a magnetic material is a fundamental challenge for future spintronics. In this respect, recent discoveries of gate-tunable magnetism in ferromagnetic two-dimensional (2D) materials such as CrI₃ and Fe₃GeTe₂ demonstrate that they are an excellent platform for the study of interplay between charge and magnetic order. In this talk, we will discuss electrostatic control of Cr₂Ge₂Te₆ (CGT), a van der Waals ferromagnetic semiconductor, in an electric double-layer transistor (EDLT) geometry [1]. We show that degenerately electron-doped CGT exhibits peculiar magnetoresistance (MR) curves with pronounced hysteresis and abrupt switching features, which is characteristic of spontaneous magnetic order. Remarkably, the hysteresis persists up to 200 K, well above the Curie temperature of 61 K for pristine CGT, indicating enhancement of exchange energy by free carriers. Further, we find that the easy axis of the doping-induced ferromagnetic state lies in the 2D plane, in contrast to the out-of-plane easy axis of the pristine material. We will discuss the role of doping on the magnetic anisotropy and prospects for observing similar effects in other magnetic 2D materials.

[1] I. Verzhbitskiy et al. "Controlling the magnetic anisotropy in Cr₂Ge₂Te₆ by electrostatic gating" Nat. Electron. In Press (https://doi.org/10.1038/s41928-020-0427-7).

Inspired by the recent discovery of ferromagnetism in two-dimensional atomically thin layers, such as chromium triiodide (CrI₃)^1,2, chromium germanium telluride (Cr₂Ge₂Te₆)³, and other van der Waals materials^4–6 has moved the emphasis of studies from bulk crystals to low-dimensional materials. Two-dimensional semiconductors, including transition metal dichalcogenides, are of interest in electronics and photonics but remain nonmagnetic in their intrinsic form. Previous efforts to form two-dimensional dilute magnetic semiconductors utilized extrinsic doping techniques⁷, bulk crystal growth⁸ or liquid precursor based chemical vapor deposition⁹, detrimentally affecting uniformity, scalability, or Curie temperature (T_C).

Here, we demonstrate successful in situ substitutional doping of Fe atoms into MoS₂ monolayers via low-pressure chemical vapor deposition (LPCVD)¹⁰. We uncover an unambiguous Fe-related spectral feature in the luminescence of Fe:MoS₂ monolayers, which is stable up to room temperature. The origin of the Fe-related luminescence peak is further supported by results obtained from a density functional theory (DFT) calculation of dipole-allowed transitions. In addition, we find that monolayer Fe:MoS₂ displays ferromagnetism by probing the hysteresis in the magnetic circular dichroism (MCD) of the Fe-related emission. Moreover, by using nitrogen-vacancy center magnetometry and magnetization measurement using superconducting quantum interference devices (SQUID), we provide clear evidence for room temperature ferromagnetism in our synthesized Fe:MoS₂ monolayers and also quantitatively determine a 0.5 ± 0.1 mT local magnetic field.

Reference:
1. Huang, B. et al. Nature 546, 270–273 (2017).
2. Jiang, S., Shan, J. & Mak, K. F. Nat. Mater. 17, 406–410 (2018).
3. Gong, C. et al. Nature 546, 265–269 (2017).
4. Deng, Y. et al. Nature 563, 94–99 (2018).
5. Bonilla, M. et al. Nat. Nanotechnol. 13, 289–293 (2018).
6. O’Hara, D. J. et al. Nano Lett. 18, 3125–3131 (2018).
7. Coelho, P. M. et al. Adv. Electron. Mater. 5, 1900044 (2019).
8. Li, B. et al. Nat. Commun. 8, 1–7 (2017).
9. Yun, S. J. et al. Adv. Sci. 7, 1903076 (2020).
10. Fu, S. et al. Nat. Commun. 11, 2034 (2020).

Magnetism is a topic of a wide interest since the discoveries of motors/generators, through magneto-resistance and up to modern times, where low dimensional materials offer a support for new magnetic phenomena. The talk will focus on the influence of magnetic moments and magnetism on the optical magneto-properties of semiconductors in an ultimate two-dimensional limit found in van der Waals transition metal phosphorous tri-chalcogenides. A few types of magnetic properties will be discussed: the long-range magnetic order, ferromagnetism, anti-ferromagnetism or special spin textures; an interfacial developed Rashba spin-orbit effect; nuclear spin Overhauser effect; magnetic polaron, all gaining special stabilization by the size confinement and a shape anisotropy. The mentioned intrinsic fields lead to a lift of energy or momentum degeneracy at band-edge states with selective spin orientation in the ground or/and excited state, being of a special interest in emerging technologies of spin-electronics and quantum computation. The lecture will include the study of long-range magnetic order and valley effects in single layer of metal phosphor tri-chalcogenide compounds. Metal phosphor tri-chalcogenides with the general chemical formula MPX₃ (M=metal, X=chalcogenide) closely resembling the metal di-chalcogenides, but the metal being paramagnetic elements, while one-third of them are replaced by phosphor pairs. The metal ions within a single layer produce a ferromagnetic or anti-ferromagnetic arrangement, endowing those materials with unique magnetic and magneto-optical properties. Most recent magneto-optical measurements will be reported, exposing the existence of valley degree of freedom in a few MPX₃ (e.g., FePS_3, MnPS₃), that reveals a protection of the spin helicity of each valley however, the coupling to an anti-ferromagnetism lifts the valleys' energy degeneracy. The phenomenon was also examined in magnetically doped diamagnetic MPX₃ layers. The results indicated the occurrence of coupling between photo-generated carriers and magnetic impurities and the formation of magnetic polaron.

References:
[1] A.K. Budniak, N.A. Killilea, S.J. Zelewski, M. Sytnyk, Y. Kauffmann, Y. Amouyal, R. Kudrawiec, W. Heiss, E. Lifshitz; Small, 2020, 16 (1), 1905924

Since the first reports on intrinsically magnetic two-dimensional (2D) materials in 2017 [1,2], the price-to-pay for accessing their monolayers is the lack of ambient stability and that magnetic ordering persists only at cryogenic temperatures.
We discovered in a mineral aggregate – mainly composed of hematite, magnetite, and chalcopyrite – soft layers of which macroscopic flakes easily could be peeled off that stuck to a permanent magnet. Employing mechanical exfoliation, we succeeded in thinning and transferring micrometer sized – mainly hexagonally shaped – flakes to SiO₂ substrates. Energy-dispersive x-ray spectroscopy (EDS) revealed magnesium and silica as major components of the flakes. Raman spectroscopy indicated the presence of hydroxide groups, pointing towards talc, a hydrated magnesium phyllosilicate mineral. Long-term EDS and Raman revealed that in the flakes about 10 % of the Mg atoms are substituted by Fe. With atomic force microscopy, a minimum flake thickness of 1 nm was determined indicating cleavage down to a talc monolayer. Combined magnetic force microscopy and superconducting quantum interference device magnetometry measurements imply that Fe-rich talc exhibits superparamagnetic behavior in the explored Fe concentration range. As convincingly demonstrated by electron diffraction, the Fe-rich flakes are also showing long-term stability under ambient conditions in contrast to the 2D magnets reported so far.
[1] C. Gong, et al., Nature 546, 265 (2017).
[2] B. Huang, et al., Nature 546, 270 (2017).

We identify through chemical substitutions and density-functional theory (DFT) calculations a family of 27 metastable single-layer transition metal monochalcogenides, MX (M = Mn, Co, Ni, Cu, Rh, Pd, Ag, Cd, Au; X = S, Se, Te), whose structures are based on the monolayer form of bulk layered CuTe. DFT calculations with the generalized-gradient PBE and hybrid HSE06 functionals of these 20 materials in four candidate 2D structures determine the thermodynamic stability and groundstate structure and characterize their electronic properties. We predict that the CuTe parent compound is indeed the most stable structure of this family. Thirteen of these two-dimensional (2D) materials display sufficiently low formation energies to be stable as free-standing 2D materials and could be mechanically exfoliated from their bulk parent phases, while several more could be grown on substrates. All the stable 2D materials are metallic, and 2D MnTe, CoS, and CoSe are magnetic. We determine that MnTe exhibits a particularly large magnetocrystalline anisotropy, a preferred out-of-plane magnetization, and strong magnetostriction. These materials may be suitable for applications as metallic contacts and wires in nanoscale or magnetic layers in semiconductor and spintronics devices.

Two dimensional (2D) Janus Transition Metal Dichalcogenides (TMDs) are a new class of atomically thin polar materials. In these materials, the top and the bottom atomic layer are made of different chalcogen atoms. To date, several theoretical studies have shown that a broken mirror symmetry induces a colossal electrical field in these materials, which leads to unusual quantum properties. Despite these new properties, the current knowledge in their synthesis is limited only through two independent studies; both works rely on high-temperature processing techniques and are specific to only one type of 2D Janus material - MoSSe. Therefore, there is an urgent need for the development of a new synthesis method to (1) Extend the library of Janus class materials. (2) Improve the quality of 2D crystals. (3) Enable the synthesis of Janus heterostructures.
The central hypothesis in this work is that the processing temperature of 2D Janus synthesis can be significantly lowered down to room temperatures by using reactive hydrogen and sulfur radicals while stripping off selenium atoms from the 2D surface. To test this hypothesis, a series of controlled growth studies were performed, and several complementary characterization techniques were used to establish a process–structure-property relationship. The results show that the newly proposed technique, namely Selective Epitaxy and Atomic Replacement (SEAR) is effective in reducing the growth temperature down to ambient conditions. The proposed technique benefits in achieving highly crystalline 2D Janus layers with an excellent optical response. Further studies herein show that this technique can form highly sophisticated lateral and vertical heterostructures of 2D Janus layers. Overall results establish an entirely new growth technique for 2D Janus layers, which pave ways for the realization of exciting quantum effects in these materials such as Skyrmion formation, 2D ferromagnetic ordering, topological superconductivity, and Majorana Zero Modes.

Janus transition metal dichalcogenide (TMD) differs from conventional TMDs by its out-of-plane mirror asymmetry and intrinsic dipole moment. The dipole moment of Janus TMD monolayer, MoSSe as an example, affects the interlayer coupling of their van der Waals heterostructures (HSs). Here, we present the unexpected interfacial interactions of Janus MoS₂/MoSSe HS by direct synthesis and twist stacking using low-frequency Raman spectroscopy. The interlayer coupling strength of Janus HS with 2H and 3R stackings is found to be strongly enhanced compared with their pristine MoS₂ counterparts through the linear chain model and density functional theory (DFT) calculations. Both experimental and DFT results indicate that the enhancement is attributed to the reduced interlayer distance due to the intrinsic dipole moment and the compressive strain of Janus MoSSe. On the other hand, low-frequency Raman modes of twisted MoS₂/MoSSe heterobilayers exhibit characteristic dependence on the twisting angles that the shear and breathing modes are both observable only at high-symmetry angles. All these spectroscopic features along with DFT calculations serve as a fingerprint of the strong interlayer coupling in Janus HSs and indicate the potential in tuning the interlayer coupling strength through Janus heterostacking.

Conductive layered metal-organic frameworks (MOFs) are a new family of 2D materials which offer electrical conductivity in addition to the other known fascinating properties of MOFs including permanent porosity and high surface area. Making progress towards application of these materials as field effect transistors, semiconductors and supercapacitors depends on a deep understanding of the structural, electronic and dynamical behavior of 2D MOFs at the atomistic level. Here, we answer to the open question of the structural stability of 2D MOFs subject to humidity with emphasis on Co₃(HHTP)₂ and Cu₃(HHTP)₂, HHTP=2,3,6,7,10,11-hexahydroxytriphenylene, which show stark structural differences in aqueous solution. Static electronic structure methods are too perfect for reflecting the flexible nature of these layered materials while ab initio molecular dynamics (AIMD) simulations are too expensive for providing a reliable picture of the complex dynamics in these systems. To tame the flexible and ever-changing layered structures of 2D MOFs, we develop ab initio parametrized force fields (AIFF) that allow large-scale/long-time simulations of the dynamics of both dry and hydrated MOFs. Our molecular dynamics simulations reproduces the structural properties of both MOFs as well as selective hydrolysis of metal nodes in hydrated Co₃(HHTP)₂ vs. intact metal nodes in hydrated Cu₃(HHTP)₂, in agreement with experimental reports. Our developed AIFFs reveal that the reason behind this behavior is the presence of \textit{intrinsic defect sites} in dry Co₃(HHTP)₂. Quantum mechanical calculations on cluster models cut from crystal structures confirm the higher tendency of Co²⁺ for coordination to water molecules compared to Cu²⁺ which prefers HHTP linkers. Our multi-faceted strategy paves the way toward simulation of realistic MOF-based materials and their interface with confined water molecules which is especially relevant to designing more robust water stable materials with desired properties and applications.

Introduction of a highly stable metal organic framework (MOF) in the late 90s has inspired a wide range of interest in these porous structures. Their high surface area and permanent porosity has stimulated a spectrum of applications for gas adsorption/storage, water treatment, heterogeneous catalysis, and even defunctionalization of chemical warfare agents. Synthesis of conductive 2D layered MOFs, where extended π-π conjugation creates possible intra- and inter-layer charge transfer path, opened another venue of potential applications for these intriguing materials. Tuning electronic properties of 2D MOFs for any specific application such as supercapacitors, field-effect transistors and electronically-induced chemical sensors can be a complicated challenge due to the flexibility of the layered structure that is composed of weak van der Waals interactions. The continuous strive for unravelling the electronic characteristics of 2D MOFs is most illustrated in the ongoing researches on Ni₃(HITP)₂, HITP= 2,3,6,7,10,11-hexaiminotriphenylene, which since its synthesis in 2014 has been the subject of a continuous debate on being a metallic or semiconductor material. This debate arise from the fact that most theoretical studies and experimental deductions neglect the flexibility and structural dynamics of 2D MOFs, the two characteristics which should be fully addressed before trying to tune their electronic properties for any specific applications. Here, we employ a range of theoretical tools to elucidate different structural aspects of 2D MOFs and their individual effect on the electronic properties and overall conductivity of the material. By obtaining the 1x1x1 Ni₃(HITP)₂ initial cell from experimental results, we construct a 2x2x2 unit cell and optimize it with periodic electronic structure calculation to get the most stable coordination geometry for metal site. According to this optimization, we develop an ab initio force field (AIFF) based on the most accurate quantum chemical calculations. Classical molecular dynamics based on the developed AIFF elucidate different structural deformations in specific environments. Then, we apply ab initio molecular dynamics (AIMD) to mimic same structural dynamics individually and determine their effects on electronic properties with subsequent periodic electronic structure calculations. Taking advantage of a combination of theoretical and computational tools, we create a general methodology to understand the electronic properties of 2D MOFs at an atomistic level.

Hexagonal boron nitride (BN) is an excellent 2D nanomaterial due to possessing with impressive mechanical property (e.g. Young’s modulus ~1 TPa ), thermal stability up to 800 ^oC, high thermal conductivity (600 W/m-K), and electrically insulative properties (band gap ~5.97 eV) (Boldrin et al., 2011). These make BN a suitable reinforcement filler for the application of electrically insulating but thermally conductive polymer composite materials. While polymer nanocomposite offers a wide range of application including packaging to the automotive due to its exceptional low cost and lightweights with improved performances, one of the main challenges is the weak interfacial molecular interaction between reinforcement fillers and polymer matrix, affecting nanocomposite properties (Wang et al., 2014).
In this work, we report a novel method to functionalize BNs by coupling agent silane (3-aminopropyltriethoxy) in order to assess the role of silane functional groups at the interface of polyethylene (PE) and BNs. We studied the effect of non-modified (pBN), and silane modified BNs (sBN) on the melt-extruded PE-BN composite performances by evaluating mechanical and thermal properties. While, both pBN and sBN nanomaterials enhance PE-BN nanocomposite performances, sBN have comparatively higher molecular interaction between PE matrix and BN fillers, resulting a significant improvement to the nanocomposite performances when compared with pBN composites. In comparison with PE materials, addition of 5 wt. % BN increases tensile modulus of elasticity by 37.4 and 42.2 %, tensile strength by 14.5 and 26.9 %, flexural modulus of elasticity by 24 and 30.5 %, and storage modulus by 80 and 90.6 % for pBN and sBN nanocomposites, respectively. In addition to the mechanical reinforcement, thermal properties were also evaluated for PE-BN nanocomposites. While thermal stability of PE-pBN and PE-sBN nanocomposites was comparable, coefficient of thermal expansion was decreased by 11.7 and 20.3 % at 5 wt. % pBN and sBN loading, respectively. These results confirm enhanced molecular interaction of BN fillers with PE polymer matrix at the interface, affecting mechanical and thermal properties of the nanocomposites. The enhanced interfacial molecular interaction further reduced phonon scattering, resulting an increased thermal conductivity (14% increment for sBN composites when compared with pBN). However, low thermal stability of silane molecules makes it less attractive for limiting PE-BN nanocomposite thermal degradation. With 5 wt.% BN loading, PE-pBN nanocomposites show 4 ^oC increase in maximum degradation temperature while PE-sBN is increased by 2 ^oC. These polymer nanocomposites with enhanced mechanical and thermal properties can be an excellent choice for application in insulation materials, particularly for cables in automotive vehicles.

Much of the current interest in atomically-thin transition metal dichalcogenide (TMD) semiconductors such as MoS₂ and WSe₂ is driven by the physics of coupled spin & valley degrees of freedom and the potential for new spin- and valley-based devices. This talk will discuss recent optical studies that probe the valley-related physics of electrons, holes, and excitons in monolayer TMD semiconductors.

In a first study [1,2], we measure the polarized magneto-absorption spectra of high-quality exfoliated TMD monolayers in very large magnetic fields to 91 tesla, which allows to follow the diamagnetic shifts of not only the 1s ground state of the neutral exciton but also its excited 2s, 3s, ..., ns Rydberg states. The energies and diamagnetic shifts provide a direct determination of the effective (reduced) exciton masses and also the dielectric properties of these monolayer semiconductors – fundamental material parameters that until now were available only from theoretical calculations. Moreover, we also measure other important material properties, including exciton binding energies, exciton radii, and free-particle bandgaps. These results provide essential and quantitative parameters for the rational design of opto-electronic van der Waals heterostructures incorporating 2D semiconductor monolayers.

In a second study, we introduce an entirely passive, noise-based approach for exploring intrinsic valley dynamics in monolayer TMD semiconductors [3]. Under conditions of strict thermal equilibrium, we use optical Faraday rotation to “listen” to the thermodynamic fluctuations of the valley polarization in a Fermi sea of resident electrons or holes, due to their spontaneous scattering between the K and K’. Frequency spectra of this valley noise reveal narrow Lorentzian lineshapes and therefore very long intrinsic valley relaxation, especially for holes (~microseconds). These results provide quantitative measurements of intrinsic valley dynamics, free from any external perturbation, pumping, or excitation.

*In collaboration with Mateusz Goryca, Jing Li, & Andreas Stier (NHMFL), Nathan Wilson & Xiaodong Xu (University of Washington), and Cedric Robert, Bernhard Urbaszek, & Xavier Marie (INSA-Toulouse)

[1] M. Goryca et al., Nature Communications 10, 4172 (2019).
[2] A. V. Stier et al., Phys. Rev. Lett. 120, 057405 (2018).
[3] M. Goryca, N. P. Wilson, P. Dey, X. Xu, & S. A. Crooker, Science Advances 5, eaau4899 (2019).

Polaritons, hybrid light-matter excitations, enable nanoscale control of light. Particularly large polariton field confinement and long lifetimes can be found in graphene and van der Waals materials, opening new avenues for low-loss photonic applications. Here, we present our recent progress in the discovery on engineering of these nanoscale polaritons in 2D materials [1,2]. First, plasmon polaritons arising from graphene and topological insulators were investigated in graphene/Bi₂Te₃, topological insulator Bi₂Te₃, graphene nanoribbon and 3D graphene using either spectroscopic or real space imaging techniques [3-6]. We show how the plasmonic coupling happens in two Dirac materials, how high-order plasmonic modes are observed in 3D graphene structure, how multiple plasmonic modes at sub-wavelength are achieved in graphene nanoribbon and how edge chirality controls the plasmonic shift.
We would also like to highlight our progress on the observation of anisotropic and ultra-low-loss polariton propagation along the surface of natural vdW material α-MoO₃ [7]. We visualized and verified phonon polaritons with elliptic and hyperbolic in-plane dispersion, which have been theoretically predicted but never experimentally observed in natural materials before. To better utilize the low-loss directional polaritons, we developed a chemical intercalation method to reversibly control and switch the phonon polaritons while maintaining their long lifetimes [8]. Combined with photolithography, spatially controlled polariton switching was achieved with various in-plane heterostructure configurations and patterns. Most recently, we show how the hyperbolic polaritons in α-MoO₃ thin slabs were delicately manipulated by controlling the interlayer twist angle [9]. We experimentally observed tunable topological transitions from open (hyperbolic) to closed (elliptical) dispersion contours in twisted α-MoO₃ bilayers at a photonic magic twist angle. The theoretical analysis showed these transitions of in-plane hyperbolic polaritons are controlled by a topological quantity. At the transitions, the photonic dispersion of the resulting topological polaritons flattens, exhibiting low-loss tunable polariton canalization and diffractionless propagation.
In summary, biaxial van der Waals semiconductors like α-MoO₃[7, 9] and V₂O₅[10] represent an emerging family of material supporting exotic polaritonic behaviors and these natural-born hyperbolic materials offer an unprecedented platform for controlling the flow of energy at the nanoscale.

The isolation of a growing number of two-dimensional (2D) materials has inspired worldwide efforts to study their fundamental physical properties and integrate distinct 2D materials into van der Waals (vdW) heterostructures.^1-2 While the few atoms thick structure of 2D semiconductors naturally makes their electronic structure quantum confined, it also has profound implications on their optical properties.^3-4 Further being 2D in nature it also presents opportunities for engineering them into 0D and 1D quantum confined structures.⁵ My talk will discuss our group’s more recent work on optical phenomena and photonic devices made from 2D semiconductors. I will present how 1D nanostructuring of excitonic 2D semiconductors into nanophotonic dielectric gratings can enable exploration of new regimes of light-matter confinement including formation of hybrid exciton-plasmon-polariton states.⁶ I will extend this discussion of light-matter interaction to excitonic/dielectric superlattices which are scalable over large areas. I will extend this concept to other excitonic materials and show dynamic tunability of their optical properties. If time permits, I will also discuss new approaches and methods for nanostructuring 2D semiconductors down to ~5 nm lateral dimensions with superior control of crystallinity and structure.⁷ I will conclude by giving a broad perspective on future prospects of 2D materials from fundamental science to applications.
References:
1. Jariwala, D.; Marks, T. J.; Hersam, M. C., Nature Materials 2017, 16, 170-181.
2. Jariwala, D.; et al. ACS Nano 2014, 8, 1102–1120.
3. Brar, V. W.; Sherrott, M. C.; Jariwala, D., Chemical Society Reviews 2018, 47, 6824-6844.
4. Jariwala, D.; et al. ACS Photonics 2017, 4, 2692-2970.
5. Stanford, M. G.; Rack, P. D.; Jariwala, D., npj 2D Materials and Applications 2018, 2, 20.
6. Zhang, H.;….Jariwala, D., et al. Hybrid Exciton-Plasmon-Polaritons in van der Waals Semiconductor Gratings. Nature Communications 2020, (in press).
7. Kumar, P….Stach, E. A.; Jariwala, D., et al. Npj 2D Materials and Applications 2020, https://doi.org/10.1038/s41699-020-0150-2.

Tailoring the photoluminescence (PL) in two dimensional (2D) transition metal dichalcogenide (TMDs) using external factors is one of the remarkable interest for its use in emerging valleytronics, nanophotonic and optoelectronic applications.^1,2 Significant effort have been devoted to enhance or manipulate the excitonic emission in a monolayer MoS₂. However, it has limited to the nanoscale fundamental studies for nanoelectronics and photonics applications. Here, we will talk about a novel van der Waal nano- hybrid/heterojunctions system fabricated with a non-lithographic process to manipulate the PL emission, which is composed of a few layer MoS₂ integrated into a transparent semiconducting silver metaphosphate glass matrix. Successful isolation and formation of heterojunction revealed the preservation of phase integrity and the crystallinity. The heterojunctions demonstrated exotic intrinsic A- and B- excitonic peak emission. More interestingly we are able to tailor a dominant B- excitonic emission over A exitonic emission. A significant 6-fold enhancement in PL spectrum (van der Waals hetrojunctions) over a control sample was recorded (Figure 1).³ Furthermore, exciton plasmon coupling was under taken to demonstrate the enhancement in B-excitonic emission of the van der Waals nanoheterojunctions. Finally, an ultrafast time-resolved spectroscopy interpreted the plasmon-enhanced electron transfer that takes place in Ag nanoparticles-MoS₂ nanoheterojunctions is behind the enhancement of the excitonic emission. No doubt, the efficient coupling of exciton-plasmon and tunability of B- excitonic emission pave great attention in emerging valleytronic and light emitting devices working with B excitons.
References

1. Sajedeh Manzeli, D. Ovchinnikov, D. Pasquier, O. V. Yazyev, & A. Kis, Nat. Rev. Mater., 2, 17033 (2017).
2. G. M. Akselrod, T. Ming, C. Argyropoulos, T. B. Hoang, Y. Lin, X. Ling, D. R. Smith, J. Kong, and M. H. Mikkelsen, Nano Lett., 15, 3578-3584 (2015).
3. A. S. Sarkar, I. Konidakis, I. Demeridou, G. Kioseoglou and E. Stratakis, Under review, 2020, https://arxiv.org/abs/2005.02852

The reduced electrical screening in two-dimensional (2D) materials provides an ideal platform for realization of exotic quasiparticles, that are robust and whose functionalities can be exploited for future electronic, optoelectronic and valleytronic applications. Recent examples include an interlayer exciton, where an electron from one layer binds with a hole from another, and a Holstein polaron, formed by an electron dressed by a sea of phonons. Here, we report a new quasiparticle, ‘polaronic trion’ in a heterostructure of MoS₂/SrTiO₃. This emerges as the Fröhlich bound state of the trion in the atomically thin monolayer of MoS₂ and the very unique low energy soft phonon mode (≤ 7 meV, which is temperature and field tunable) in the quantum paraelectric substrate SrTiO₃ (STO), arising below its structural antiferrodistortive (AFD) phase transition temperature. This dressing of the trion with soft phonons manifests in an anomalous temperature dependence of photoluminescence emission leading to a huge enhancement of the trion binding energy (~70 meV). The soft phonons in STO are sensitive to electric field, which enables field control of the interfacial trion-phonon coupling and resultant polaronic trion binding energy. Lastly, these results suggest evidence of Rotational Frohlich Coupling a result of the counter intuitive interplay between the rotational axis of the MoS₂ trion and STO phonon mode. Polaronic trions could provide a platform to realize quasiparticle based tunable optoelectronic applications driven by many body effects.
References:
1. Sarkar, S., Goswami, S., Trushin, M., Saha, S., Panahandeh-Fard, M., Prakash, S., Tan, S.J.R., Scott, M., Loh, K.P., Adam, S. and Mathew, S., 2019. Polaronic Trions at the MoS2/SrTiO3 Interface. Advanced Materials, 31(41), p.1903569.
2. Trushin, M., Sarkar, S., Mathew, S., Goswami, S., Adam, S. and Venkatesan, T., 2019. Evidence of Rotational Fr\" ohlich Coupling in Polaronic Trions. arXiv preprint arXiv:1911.09118.
3. https://www.advancedsciencenews.com/discovery-of-an-unusual-quasiparticle-in-a-2d-material/
4. https://www.sciencedaily.com/releases/2019/08/190827095049.htm

Recognizing color is a primary means of object identification – both in biological and non-biological systems. In the latter case, i.e. conventional photodetection or imaging, arbitrary color-identification is not an inherent function. On one hand, a color-camera can disperse incident light into its red, green and blue (RGB) components (using filters) and merely record the voltage-response the RGB-filter-covered detectors in a 0-255 scale. However, either a classification algorithm or a human observer of the regenerated image is needed to recognize the color. For pure machine-based color recognition, the set of recorded RGB responses can be used to estimate the nearest color of the incident light without the observer’s mediation. Spectral estimation on the other hand, usually carried out using dispersion-based spectrometers, is the process of breaking the incident light into its wavelength components via diffraction-grating and finding the intensity at each wavelength. These two types of tools, camera and spectrometers, however, cannot be used interchangeably and each have their limitations. We report on a dispersion-free 2D materials-based and machine learning-enabled color and spectrum-estimator, that can mimic the color-estimation ability of the eye by learning to perceive new colors, as well as spectral estimation capability in the spectral range of 350 nm – 800 nm. This technique, where we use optical transmittance of a few transition-metal-dichalcogenide (TMD)-based thin-film filters -present an alternative for the estimation problem, and can reduce the manufacturing costs and the design complexities of the conventional color-estimation and spectrum-estimation tools, and also introduce the learning capacity to the estimator.
Through a comprehensive simulation scheme that led to the selection of an ideal set of filters, we have shown the importance of excitonic features of the transmittance curves in the estimation problem, and have designed and fabricated the thin-film filters based on outcome of these simulations. We have also studied the efficacy of various machine learning algorithms and identified the key advantages and limitations of each algorithm for color and spectrum estimation. Finally, by modeling the temporal drift of spectral transmittance of the filters over the period of one year, we have shown that it is possible to overcome the drift-induced inaccuracies over extended times.

Two-dimensional (2D) materials are promising candidates for optoelectronics. Previous studies have shown that light absorption in 2D materials can be enhanced by generating multilayers using layer-by-layer stacking or integration with cavity resonators. However, these approaches have critical limitations as they may lead to optical properties that differ from the intrinsic properties of monolayers and they are still governed by the anisotropic thin film absorption with little control over light with different polarizations and incident angles. Here, we show that adding three-dimensional (3D) textures with controlled geometry to 2D materials, demonstrated with monolayers of transition metal dichalcogenides (TMDs), is a powerful solution to these challenges. For this, monolayer TMDs (MoS₂, WS₂, and WSe₂) were conformally grown on surfaces with nanoscale half-spherical textures, producing wafer-scale optical films with distinct geometry at different length scales: globally they behave as an optically flat substrate, whereas they possess isotropic bubble-like surfaces at the sub-wavelength limit. Spectrally-resolved, angle- and polarization-dependent absorption and reflection measurements confirm that all our TMD films with 3D textures have enhanced absorption for all visible wavelengths with a more pronounced enhancement for higher incident angles and p-polarization that flat optical films do not absorb strongly. This suggests that our TMD monolayers provide flat optoelectronic films with enhanced and isotropic light absorption, which can be precisely predicted by the geometric design of the surface texture parametrized by the aspect ratio of the domes in our experiment. This approach will advance the development of integrated optoelectronic circuitry with customizable light absorption.

Transition metal dichalcogenide materials (TMDCs) are an emerging type of van der Waals semiconductors that possess unique electronic and optical properties^1,2 making them promising candidates for future photonic devices. Previous studies focused on monolayer/few layer structures, while, at the same time, properties of bulk materials and their applications have received little attention. Here we study theoretically light interaction with and propagation in bulk TMDCs materials. We show that owing to their high refractive index³, TMDCs may pave the way to a new generation of optical components with higher light confinement and smaller footprint, when compared to traditional Si and III/V photonic devices.
We begin our analysis with a study of light guiding in a simple planar waveguide made of bulk TMDCs (MoS₂, MoSe₂, WS₂ and WSe₂) and conventional semiconductors (Si, InP and GaAs). In our analysis we focus on the near-infrared and telecom frequencies, that is, below the band gap of the semiconductors, where materials are transparent. By analyzing the dispersion and mode properties of such waveguides, we find that the waveguide footprint may be reduced almost twice when compared to counterparts made of Si. Next we extended our analysis to other waveguide structures, including ridge, rib and slot waveguides. We show that deeply subwavelength and highly confined light guiding is possible in waveguides made out of TMDCs. These findings may pave the way to novel subwavelength optical devices based on TMDC materials.
We further discuss light interaction with photonic crystals made of TMDC materials. Specifically, we have designed and analyzed a 2D photonic crystal slab with a cavity in it. We find that the size of such a cavity can be made significantly smaller (~30%) than that for a comparable Si or GaAs structures, while demonstrating higher resonator quality factor. Small device size and significant concentration of electromagnetic field have profound implications for quantum and nonlinear optics. Specifically, as we show, such cavities demonstrate a threefold enhancement of Purcell factor when compared to analogous Si resonators. We then demonstrate that strong optical confinement in a small volume leads to a strong light-materials interaction and can be utilized to create more energy efficient active devices, such as modulators. In particular, we demonstrate that two-fold reduction in both cross-sectional area and switching energy consumption are possible with TMDC structures.
Finally we discuss a potential roadmap for integrated photonics based on TMDC structures. We show that owing to an enhanced light-materials interaction in TMDC nanostructures novel opportunities for highly integrated and energy efficient photonic devices emerge.

References:
1. Mak, K. F.; Shan, J. J. N. P., Photonics and optoelectronics of 2D semiconductor transition metal dichalcogenides. 2016, 10 (4), 216.
2. Xia, F.; Wang, H.; Xiao, D.; Dubey, M.; Ramasubramaniam, A. J. N. P., Two-dimensional material nanophotonics. 2014, 8 (12), 899.
3. Wilson, J. A.; Yoffe, A. J. A. i. P., The transition metal dichalcogenides discussion and interpretation of the observed optical, electrical and structural properties. 1969, 18 (73), 193-335.

Two-dimensional flat MoS₂ retains an outstanding interest as active material to manipulate light via modulation or harvesting, owing to its thickness dependent optical response. Starting from this background, MoS₂ layers nanostructured in anisotropic layouts gain an extra-control of the optical characteristics enabling fabrication of novel metasurfaces or flat-optic configurations for light harvesting [1]. Here we illustrate two possible methodologies to reduce ultra-thin MoS₂ to different anisotropic structural fashions. One consists of rippling MoS₂ by means of pattern-assisted chemical vapour deposition [2]. This way, we prove MoS₂ can be continuously grown on a one-directional pre-patterned substrate by means of chemical vapor deposition or sulfurization schemes therein mimicking the anisotropic shape feature [3]. This results in a strongly dichroic optical response that is concomitant with an anisotropic strain redistribution in the MoS₂ layer (revealed by 2D Raman mapping and Kelvin-probe force microscopy) [2] and with a peculiar exciton dynamics (revealed by transient optical spectroscopy) [4]. As an alternative, a second layout is based on physically separated MoS₂ stripes with nanoscale thickness (few nm) produced by sulfurization of a Mo ribbon-like pattern prefabricated by laser interference lithography. Spatial confinement in the stripe pattern results in an enhanced ultrabroadband amplification of the optical absorption throughout the visible spectrum following a guided-mode anomaly. This layout is also presented as a suitable platform where to engineer hybrid plasmonic systems as, for instance, coming from the interplay of MoS₂ nanostripes and Au nanoparticles. Overall, we show that MoS₂ (and expectedly other 2D transistion metal dichalcogenides with similar features) is strikingly suitable to anisotropy design expanding the edges of light-matter interactions with potential in applications for optical modulation, large-area flat optics, and light harvesting.
[1] C. Martella et al., Adv. Mater. (2018) 1, 1705615
[2] C. Martella et al., Adv. Mater. (2017) 29, 1605785
[3] C. Martella et al., Nano Res. (2019) 12, 1851
[4] A. Camellini et al., ACS Photon. (2018) 5, 3363

For the last few years we have been witnessing significant progress in micro and nano photonics due to successful fabrication of atomically thin optoelectronic devices based on two-dimensional (2D) layered materials and their vertical stacks [1-5]. Considering near infrared (NIR)/short-wave infrared (SWIR) photodetection, we find a theoretical study reported in [6] which describes various possible combinations of group-6 and group-7 monolayer TMDs. Among different combinations of type-II van der Waals (vdW) heterointerfaces, owing to efficient generation, separation, and collection of charge carriers, it has been found that both ReS₂/WSe₂ and ReS₂/MoSe₂ emerge as potentially promising candidates for the next generation ultrathin optoelectronic devices [6].
However, the analysis of electronic transmission through ReS₂/WSe₂ and ReS₂/MoSe₂ heterostructures is required for further insights on their conductance at near-equilibrium. In this work, employing density functional theory (DFT) and nonequilibrium Green’s function (NEGF) combination, we have computed zero bias transmission spectra of the two-port devices with monolayer ReS₂/monolayer WSe₂ and monolayer ReS₂/monolayer MoSe₂ channels. In order to conduct first-principles based DFT calculations, we have utilized the "QuantumATK" software package [7].
For a small energy range (-2 eV to 2 eV) near the Fermi level, we find that the electronic transmission through ReS₂/MoSe₂ heterointerface is slightly better than that in ReS₂/WSe₂. Besides, the near-direct bandgap and superior optical absorption of the constituent MoSe₂ layer are the other key features which make ReS₂/MoSe₂ heterostructure suitable for NIR/SWIR photodetection [6]. However, the electronic structure calculations reveal that the larger band offset values at the interface can ensure better separation of charge carriers for ReS₂/WSe₂ vertical stack.

* This work was funded by the Department of Science & Technology, Govt. of India through the grant DST/SJF/ETA-01/2016-17. D.S. acknowledges the Department of Electrical Engineering, Indian Institute of Technology (IIT) Bombay for the Institute Post Doctoral Fellowship.

References:
[1] F. H. L. Koppens et al., Nature Nanotechnology, 9, 780-793, 2014.
[2] T. C. Berkelbach and D. R. Reichman, Annu. Rev. Condens. Matter Phys., 9, 379-396, 2018.
[3] The International Technology Roadmap For Semiconductors 2.0, "Outside System Connectivity," 2015.
[4] A. K. Geim and I. V. Grigorieva, Nature, 499, 419-425, 2013.
[5] C-H. Lee et al., Nature Nanotechnology, 9, 676-681, 2014.
[6] D. Saha, A. Varghese, and S. Lodha, ACS Appl. Nano Mater., 3, 820-829, 2020.
[7] QuantumATK version 2018.06, Synopsys Quantumwise A/S, Available at http://quantumwise.com/.

Crystalline metallic films are attracting significant attention within the nanophotonics community as a material platform for plasmonics that offers light concentration and manipulation on extreme subwavelength scales with lower intrinsic losses than for plasmons in amorphous metal films. Recent advances in nanotechnology now enable the fabrication of such crystalline films with nanometer-scale thickness, down to the level of only several atomic monolayers, further boosting plasmon confinement for the miniaturization of next-generation electronic and photonic devices. However, in very thin crystalline metallic films, quantum mechanical effects play a non-negligible role in the dynamics of electrons and their interaction with external electromagnetic fields, manifesting in the spectral features associated with its linear optical response that depend on film thickness and crystallographic orientation. These effects are anticipated to become even more significant when dealing with nonlinear optical phenomena, which necessitate intense optical fields that access additional transitions among quantized electronic states.
Here we reveal quantum and finite-size effects in the nonlinear optical response associated with plasmons supported by few-atom-thick crystalline noble metal films. In particular, we employ a quantum-mechanical description that captures the main features in the electronic band structure associated with the various crystal facets of noble metals to simulate their nonlinear optical response by extending the random-phase approximation to higher orders. Our results demonstrate strong signatures of film thickness and crystallographic orientation in the linear plasmonic response that are amplified in nonlinear optical processes such as harmonic generation and two-photon absorption, with additional effects attributed to surface states, electron spill out, and the directional band gap associated with bulk atomic-layer corrugation. These findings are of key importance in the characterization of atomically-thin crystalline films and their applications for nonlinear plasmonic nanophotonic devices.

Layered semiconductors attract significant attention due to their diverse physical properties controlled by their composition and the number of stacked layers. Herein, large crystals of the ternary layered semiconductor chromium thiophosphate (CrPS₄) are prepared by a vapor transport synthesis. Optical properties are determined using photoconduction, absorption, photoreflectance, and photoacoustic spectroscopy exposing the semiconducting properties of the material. A simple, one-step protocol for mechanical exfoliation onto transmission electron microscope grid is developed [1,2] and multiple layers are characterized by advanced electron microscopy methods, including atomic resolution elemental mapping confirming the structure by directly showing the positions of the columns of different elements’ atoms. CrPS₄ is also liquid exfoliated and in combination with colloidal graphene, an ink-jet printed photodetector is created. This all-printed graphene/CrPS₄/graphene heterostructure detector demonstrates specific detectivity of 8.3×10⁸ (D*). This study shows a potential application of both bulk crystal as well as individual flakes of CrPS₄ as active components in light detection, when introduced as ink printable moieties with a large benefit for manufacturing [1].

References:
[1] A.K. Budniak, N.A. Killilea, S.J. Zelewski, M. Sytnyk, Y. Kauffmann, Y. Amouyal, R. Kudrawiec, W. Heiss, E. Lifshitz; Small, 2020, 16 (1), 1905924
[2] M. Shentcis, A.K. Budniak, R. Dahan, Y. Kurman, X. Shi, M. Kalina, H.H. Sheinfux, M. Blei, M.K. Svendsen, Y. Amouyal, F. Koppens, S. Tongay, K.S. Thygesen, E. Lifshitz, F.J.G. de Abajo, L.J. Wong, I. Kaminer; Under revision in Nature Photonics

Acknowledgments:
This work was supported by the European Commission via the Marie-Sklodowska Curie action Phonsi (H2020-MSCA-ITN-642656)
This work was performed within the grant of the National Science Centre Poland (OPUS 11 no. 2016/21/B/ST3/00482).
S.J.Z. also acknowledges the support within the ETIUDA 5 grant from National Science Center Poland (no. 2017/24/T/ST3/00257).

A promising class of lead-free semiconductors for photovoltaic applications include the bismuth halides, such as bismuth iodide [1], bismuth oxyiodide [2] and the double perovskite Cs2AgBiBr6 [3]. Although these materials have been predicted to exhibit defect tolerance, as seen in lead-halide perovskites, and already display improved stabilities and long charge carrier lifetimes, the power conversion efficiencies of the corresponding devices have not reached the level of lead-based perovskites. Potential reasons for this are explored, for example, the disconnected nature of the bismuth halide octahedra in the crystal structure, which limits carrier mobility, and the lower levels of absorption due to indirect bandgaps.
Phonons have been shown to play a crucial role in carrier transport and hot carrier cooling in the lead halides [4], and there are many reports of strong carrier-phonon coupling, especially in low-dimensional morphologies [5]. The effect of phonons is expected to be even greater in the bismuth based perovskites due to their indirect bandgaps, as carriers will emit phonons to cool to the band extrema separated in momentum space from the excitation point, as well as to recombine. The combined effects of quantum confinement and coupling to phonons could allow carriers to overcome this distance in momentum space. The relative rates of these processes are vital for device efficiency, therefore we probe the early-time excited states in many bismuth halide semiconductors using transient absorption, Raman and photoluminescence spectroscopy. Overall, this work indicates that bismuth-based materials have the potential to be used in efficient optoelectronic devices, but there is a need to account for the effects of strong carrier-phonon coupling and localisation of electronic states on carrier scattering rates. We therefore present charge carrier-lattice interaction strength as an important design criterion for efficient next-generation solar cells.

[1] R. Brandt et al., J. Phys. Chem. Lett. 6, 4297–4302 (2015).
[2] R. Hoye et al., Adv. Mater. 1702176 (2017).
[3] R. Hoye et al., Adv. Mater. Interfaces 1800464 (2018).
[4] M. Price et al., Nat. Commun. 6, 8420 (2015).
[5] F. Thouin et al., Nature Materials 18, 349–356 (2019).

Polaritons in 2D materials have been extensively studied over the past decade due to their fundamental interest and as a platform for applications in telecommunications and sensing [1]. The wavelength of these polaritons is generally small compared to that of a photon of the same frequency [2], making them very attractive to manipulate light at deep-subwavelength distances. However, it simultaneously introduces a challenge in the coupling between propagation light and plasmonic modes, since the momentum mismatch between the two makes the in/out-coupling of this process intrinsically weak [2].

In this work, we address some fundamental limits in the coupling of radiation to 2D polaritons [3]. We study the scattering properties of 0D and 1D scatterers over a 2D or finite-thickness layer and quantify the coupling of light-to-polaritons cross-section for this process as a function of the scatterer effective polarizability. We find a generalization of the optical theorem to 2D polaritons, which allows us to determine the maximum possible value of their effective polarizability, which is remarkably independent of material. We present our results in the form of simple and rigorous analytical expressions, delivering an exact quantitative measure of photon-to/from-polariton coupling and polariton-to-polariton scattering. To that end, we first introduce a universal characterization of 2D SPs and their associated optical fields in terms of a single parameter – the ratio of their wavelength λ_p to the film thickness – regardless of the physical nature of the material’s response. This allows us to determine the efficiency of different optical scattering channels involving incident light and polaritons, expressed in terms of universal light−polariton coupling strengths and fundamental limits to the scatterer polarizability. We write the result as a universal curve that is applicable to any layer thickness in the electrostatic approximation. Additionally, this result leads to a maximum light-to-polariton coupling cross-section which we explore, particularly by directly comparing the coupling efficiency of 0D and 1D scatterers and finding that the latter are up to 100 times more efficient than the former. We formulate our results for both 0D and 1D scatterers, including material edges, and present them in simple closed-form expressions. We finally propose coupling light to plasmons using a graphene edge as a configuration which maximizes the coupling efficiency, and address the case of scatterers with intrinsic losses to propose an optimum distance between the surface and the scatterer which maximizes the efficiency.

We find the maximum possible photon-to-polariton cross section to be of the order of λ_p³/λ₀ and λ_p²/λ₀ for point and line scatterers, respectively (λ₀ is the light wavelength). For graphene, the limits to in-coupling are quantified by the maximum possible ratio between the numbers of generated plasmons and incident photons N_plasmon/N_photon∼α³ and α² for point and line scatterers, where α≈1/137 is the fine structure constant. Besides their fundamental interest, our results provide useful tools for the design of optical devices involving the in/out-coupling of propagating light and 2D SPs.

[1] J. García de Abajo, "Graphene Plasmonics: Challenges and Opportunities," ACS Photonics 1, 135 (2014).
[2] A.H. Castro Neto, F. Guinea, N.M.R. Peres, K.S. Novoselov, and A.K. Geim, "The electronic properties of graphene," Rev. Mod. Phys. 81, 109 (2009).
[3] E.J.C.Dias and F.J. García de Abajo, "Fundamental Limits to the Coupling between Light and 2D Polaritons by Small Scatterers." ACS Nano 13, 5184 (2019).

Transition metal dichalcogenides (TMDs) are ideal for exploring fundamental physics and applied optics as they are semiconductors with a direct bandgap in the monolayer limit. Chemical vapor deposition (CVD) allows growing wafer scale thin films of TMDs. The improved optical quality of MoS₂ monolayers with a reduced transition linewidth below 5 meV at low temperature (T = 4 K) by reducing dielectric disorder allows combining CVD grown MoS₂ monolayers to form multilayers to access new functionalities [1,2].

We examine the correlation between the stacking order and the interlayer coupling of valence states in high quality MoS₂ as grown CVD bilayers and artificially stacked bilayers from CVD monolayers. We show that hole delocalization over the bilayer in 2H stacking is
allowed and results in strong interlayer exciton absorption and also in a larger A-B exciton separation as compared to 3R bilayers, where both holes and electrons are confined to the individual layers [2,3]. Comparing white light reflectivity spectra for 2H and 3R stacking allows extracting an interlayer coupling energy of about 50 meV. Obtaining very similar results for as-grown and artificially stacked bilayers is promising for assembling large area van der Waals structures with CVD material, using inter layer exciton formation and A-B exciton separation as indicators for interlayer coupling [2].

In bilayer MoS₂ the interlayer exciton has an out-of-plane static electric dipole. We show that as a result the interlayer exciton transition energies are widely tunable in applied electric fields, which is impossible for excitons in monolayers.

We show tuning over 120 meV of interlayer excitons with a high oscillator strength in bilayer MoS₂ due to the quantum-confined Stark effect [4]. We optically probed the interaction between intra- and interlayer excitons as they were energetically tuned into resonance. Interlayer excitons interact strongly with intralayer B excitons, as demonstrated by a clear avoided crossing, whereas the interaction with intralayer A excitons is substantially weaker. Our observations are supported by density functional theory (DFT) calculations, which include excitonic effects.

[1] S. Shree, A. George et al., 2D Materials,7, 1 (2019).
[2] I. Paradisanos, S. Shree, Nature Communications, 11, 2391 (2020).
[3] I. C. Gerber, E. Courtade, S. Shree et al., Physical Review B, 99, 035443 (2019).
[4] N. Leisgang, S. Shree, I. Paradisanos, L. Sponfeldner, Nature Nanotechnology, 2612, 63 (2020).

Following the graphene breakthrough, many other 2D materials have been discovered and extensively studied due to their exceptional physicochemical properties arising from the unique layered structure. Therein, 2D graphitic carbon nitride (g-C₃N₄) nanosheet has attracted dramatically increasing attention as a photocatalyst because of its nontoxicity, low cost, high stability and large surface, etc. However, g-C₃N₄ is still suffering from fast recombination of photo-induced charge carriers and insufficiently broad photoresponse range.^1-2 Herein, we will present some of our recent development in efficient and broadband 2D g-C₃N₄ nanosheet-based photocatalysts. One example is about our synthesized nanocomposites based on plasmonic Au nanoparticles (NPs) and in-situ synthesized lanthanide-doped NaYF₄ on g-C₃N₄. It showed enhanced ultraviolet (UV)-, visible- and near infrared (NIR)-light photocatalytic activity in the degradation of organic pollutants.³ The used Au and rare-earth metals in this work are still expensive. To utilize solar energy more efficiently and cost-effectively, a 0D/2D heterojunction based on NIR-responsive quantum dots loaded g-C₃N₄ nanosheets was constructed, which showed a record NIR photocatalytic activity in wastewater treatment.⁴ The use of heavy-metal based QDs constitutes a big concern in the environmental application. To utilize solar energy more environmentally-friendly, a completely metal-free 2D/2D heterojunction of black phosphorus (BP)/g-C₃N₄ was synthesized.⁵ The 2D/2D heterojunction showed high activity and long-term stability in solar H₂ evolution. It is noteworthy that we developed an innovative, “organic ice”-assisted exfoliation method,⁶ which led to the high-yield and high-quality few-layer BP with largely reduced energy inputs and processing time. In addition to yielding novel and interesting materials and properties, the current work also provides physical insights that can contribute to the future development of broadband photocatalysts and their applications in energy and environmental areas.

1. Q. Zhang, et al., J. Materiomics 2017, 3, 33-50.
2. Q. Zhang, et al., Nanoscale Horiz. 2019, 4, 579-591.
3. Q. Zhang, et al., ACS Catal., 2017, 7, 6225-6234.
4. Q. Zhang, et al., Appl. Catal., B 2020, 270, 118879.
5. Q. Zhang, et al., Adv. Funct. Mater. 2019, 1902486.
6. Q. Zhang, et al., United States Patent, Application No. 62/685,371; International Patent, Application No. PCT/CA2019/050813.

Liquid phase exfoliation has become an important top down production technique giving access to large quantities of nanosheets in colloidal dispersion. Importantly, this is a highly versatile technique that can be applied to numerous layered materials from graphite to transition metal dichalcogenides, h-BN, III-VI semiconductors, hydroxides etc. All materials can be exfoliated in a similar way using aqueous surfactant or suitable solvents as stabilisers. Nanosheets in dispersion are extremely polydisperse with broad lateral size and thickness distributions. To narrow size and thickness distributions, liquid cascade centrifugation has proven to be a powerful tool for efficient size selection yielding nanosheet dispersions with well-defined dimensions.
Building on these established methodologies of exfoliation and size selection, it became possible in recent years to adress some fundamental questions regarding the exfoliation process, such as: In which way is the yield, nanosheet size and layer number dependent on the liquid medium? What is a good descriptor for the exfoliation efficiency and what does it depend on? What is the role of defects? In this talk, these fundamental questions will be addressed and a picture of the current understanding will be drawn.

Layered transition metal dichalcogenides (TMDs) exhibit a rich diversity of structural polymorphs defined by the coordination symmetry of transition metal atoms.¹ Tailoring the lattice symmetry within individual TMD layers enables disparate electronic properties ranging from semiconducting (2H phase) to metallic and semimetallic (1T/1T’ phases) for the same material composition. The metastable 1T/1T’ polymorphs are generally obtained by destabilisation of the thermodynamically stable 2H phase, for instance, via electron transfer. The metastable 1T/1T’ phases are particularly appealing for the heterogeneous catalysis as they demonstrate higher electrical conductivity and higher density of catalytically active sites compared to the semiconducting 2H counterparts.² Yet to date, obtaining the metastable 1T/1T’ phases of high purity and in measurable quantities remains challenging. In our previous work, we demonstrated a successful colloidal synthesis of the 1T’ phase of WSe₂.³ Here, we report on how the designed colloidal approach can be expanded to synthesise W_xMo_1-xSe₂ (x = 0-1) nanostructures. Although colloidal MoSe₂ nanoflowers are attained in the 2H phase, alloying with tungsten, which under the same synthesis conditions forms the 1T’ phase of WSe₂, allowed us to stabilise the 1T’ phase in the ternary W_xMo_1-xSe₂ nanostructures. Moreover, the W_xMo_1-xSe₂ nanostructures display tunable 1T’/2H crystal phase ratio that depends on their chemical composition. A set of the W_xMo_1-xSe₂ (x = 0-1) working electrodes was tested for the catalytic hydrogen evolution reaction (HER). The ternary 1T’/2H W_xMo_1-xSe₂ (x ~0.5) nanoflowers demonstrate an enhanced catalytic activity and faster reaction kinetics compared to either the pristine 2H MoSe₂ or the 1T’ WSe₂ nanoflowers. We suggest that the presence of the metastable 1T’ phase in the ternary W_xMo_1-xSe₂ (x ~0.5) nanoflowers may cause the observed improvement in the catalytic performance.

References:
1. Yang, H., Kim, S. W., Chhowalla, M. & Lee, Y. H. Structural and quantum-state phase transition in van der Waals layered materials. Nat. Phys. 13, 931–937 (2017).
2. Tang, Q. & Jiang, D. Mechanism of hydrogen evolution reaction on 1T-MoS2 from first principles. ACS Catal. 6, 4953–4961 (2016).
3. Sokolikova, M. S., Sherrell, P. C., Palczynski, P., Bemmer, V. L. & Mattevi, C. Direct solution-phase synthesis of 1T’ WSe2 nanosheets. Nat. Commun. 10, 712 (2019).

Black phosphorus (BP), which was discovered by Bridgman back in 1914, has recaptured attention due to its promising physical and chemical properties. BP is the thermodynamically stable allotrope of phosphorous under ambient conditions. It is a layered material that can be exfoliated into a monolayer two-dimensional (2D) material similar to graphene. As a rising star in 2D materials, few-layer phosphorene is considered a strong competitor against other 2D materials and can be applied in an increasing number of fields such as energy storage and biomedical devices. To realize its real properties, the wide application of phosphorene nanosheets depends on the development and optimization of exfoliation methods. Fabricating BP with a high level of quality, uniformity, and producibility is of crucial importance for its large-scale application. Among various top-down and bottom-up exfoliation methods, electrochemical exfoliation has emerged as an attractive approach for exfoliating 2D materials with high quality and high yields. However, they are multi-step and time-consuming procedures, and thus less attractive for practical applications. It is, therefore, necessary to develop alternative techniques that can fabricate and deposit phosphorene nanosheets on substrates in a facile, single-step, scalable, and eco-friendly manner for energy storage devices. In this study, we propose a novel and straightforward two-in-one process to exfoliate bulk BP into phosphorene nanosheets in DI water, which are then dragged electrophoretically to be deposited on a conductive substrate. The procedure is based on the mechanism of bipolar electrochemistry (BPE), which is based on applying a sufficiently high voltage to generate electrochemical reactions between two feeding electrodes and a conductive bipolar electrode placed wirelessly between them. The difference in the electric potential between the solution and the bipolar electrode drive redox reactions on the cathodic and anodic poles of the bipolar electrode. Due to the concept of BPE, in general, the induced voltage on the two poles of bulk BP depends on the applied voltage, length of the BP electrode, and distance between the two feeding electrodes. After 24 hours of BPE, bulk BP did not show any noticeable change; however, obvious deposition of a thin film on the feeding electrodes of the bipolar cell can be observed. The exfoliated-and-deposited BP nanosheets were characterized with different microscopic and spectroscopic techniques (SEM, Raman, XPS, and HR-TEM analysis), revealing thin layers of 2D phosphorene with the orthorhombic crystal structure and lateral dimensions up to a few hundreds of nanometers. SEM images showed the high-quality 2D morphology of phosphorene nanosheets on the feeding electrodes. TEM images confirmed that the crystal structure of BP appears to be not affected by the bipolar exfoliation, which confirms a low defect concentration in the produced phosphorene nanosheets. Owing to the 2D structure of the exfoliated-and deposited phosphorene, the electrochemical performance of the phosphorene nanosheets was evaluated for capacitive energy storage application in a two-electrode symmetric configuration (EIS, CV, and GCD). The deposited 2D phosphorene nanosheets delivered high specific capacitance of 11 mF cm^-2 at the scan rate of 2 mV s^-1 and specific capacitance of 8.57, 7.62, 6.66, 5.0, 4.65, 4.01, 3.32, and 2.86 mF cm^-2 at the scan rate of 5, 10, 20, 50, 100, 200, 500, and 1000 mV s^-1, respectively, which is superior than most of the 2D materials-based devices such as MXene, 2D MnO₂, graphene or graphene oxide. This non-toxic, straightforward, and inexpensive novel method could be applied for other 2D materials for energy storage and biomedical application. In this conference, detailed novel bipolar exfoliation and deposition of phosphorene nanosheets and its electrochemical characterization will be presented.

A pair of electron and hole across the interface of semiconductor heterostructure can form a bound quantum state of the interlayer exciton. In a coupled interface between atomically thin van der Waals layers (vdW), the Coulomb interaction of the interlayer exciton increases further. In this presentation, we will discuss observing interlayer exciton formation in semiconducting transition metal dichalcogenide (TMDC) layers. Unlike conventional semiconductor heterostructures, charge transport in of the devices is found to critically depend on the interlayer charge transport, electron-hole recombination process mediated by tunneling across the interface. We demonstrate the enhanced electronic, optoelectronic performances in the vdW heterostructures, tuned by applying gate voltages, suggesting that these a few atom thick interfaces may provide a fundamental platform to realize novel physical phenomena. Furthermore, complete experimental control of density, displacement and magnetic fields in our graphene double layer system enables us to explore the rich phase diagram of several superfluid exciton phases with the different internal quantum degrees of freedom.

One of the most alluring features of 2D materials is their tunability, as their carrier density, effect mass, band gap, and electronic phases can be tuned in situ by external parameters such as electric field, magnetic field and strain. Here we will present transport studies of high quality 2D materials. For instance, we demonstrate large tunable SOC and zero-field spin-splitting in atomically thin InSe with unprecedented mobility. From beating patterns in quantum oscillations, we establish that the SOC parameter a is thickness-dependent; it can be continuously modulated over a large range by an out-of-plane electric field, achieving zero-field splitting tunable between 0 and 20 meV. Surprisingly, a could be enhanced by an order of magnitude in some devices, suggesting that SOC can be further manipulated by variations in interlayer spacing induced by stacking and/or electrostatic compression. Our work highlights the extraordinary tunability of SOC in 2D materials, which can be harnessed for in operando spintronic and topological devices and applications.

Many-electron interaction effects dominate many spectroscopic properties of reduced-dimensional systems, leading often to manifestation of novel phenomena not seen in the bulk. In this talk, I present some of our recent work on electron interaction effects in atomically thin quasi 2D materials: (1) strongly bounded correlated multi-particle excitations, such as trions and bi-excitons, in quasi 1D and 2D materials; (2) giant excitonic effects in shift currents, a nonlinear optical phenomenon, in non-centrosymmetric quasi 2D semiconductors; and (3) excitons and band renormalization in optical field driven angle-resolved photoemission spectroscopy (ARPES). Studies of these novel phenomena are made possible because of newly developed theoretical methods which incorporate higher-order many-electron correlation effects from first principles, using an interacting Green’s function approach.

In a magnetic topological insulator, nontrivial band topology conspires with magnetic order to produce exotic states of matter that are best exemplified by quantum anomalous Hall (QAH) insulators and axion insulators. Up till now, such magnetic topological insulators are obtained by doping topological insulators with magnetic atoms. The random magnetic dopants, however, inevitably introduce disorders that hinder further exploration of topological quantum effects in the material. We resolve this dilemma by probing quantum transport in MnBi₂Te₄ thin flake—a topological insulator with intrinsic magnetic order. In this layered van der Waals crystal, the ferromagnetic layers couple anti-parallel to each other, so bulk MnBi₂Te₄ is an antiferromagnet. Atomically thin MnBi₂Te₄, however, becomes ferromagnetic when the sample has odd number of septuple layers (a septuple layer represents a single structural unit in the out-of-plane direction). We observe zero-field QAH effect in a five-septuple-layer specimen; an external magnetic field further enhance the QAH quantization by forcing all layers to align ferromagnetically. MnBi₂Te₄ therefore becomes the first intrinsic magnetic topological insulator exhibiting QAH effect.

The availability of intense tabletop terahertz-frequency pulses and simple field-enhancement structures that enable peak electric fields of tens of MV/cm have ked to demonstrations of highly nonlinear THz-induced responses from a wide variety of materials. THz-driven electronic, magnetic, and structural phase transitions and domain switching have illustrated novel possibilities for control over collective properties and dynamics. In experiments on single-layer and few-layer MoTe₂, we have found that irradiation with just one sufficiently strong THz pulse induces an irreversible phase transition [1]. The initial, noncentrosymmetric 2H phase disappears, as shown by the disappearance of characteristic Raman lines and second harmonic generation. After further THz irradiation, the topological insulator phase appears, revealed by its distinct Raman spectrum. Single-shot measurements of the MoTe₂ evolution as a function of time after THz irradiation reveal complex dynamics over many-picosecond time scales. Theoretical calculations show that the transition is induced through THz-induced liberation of carriers which stabilize the new phase and reduce the energy barrier between the two phases. The result illustrates the prospects for THz control over complex multiphase landscapes in quantum materials.

[1] “Terahertz-driven irreversible topological phase transition in two-dimensional MoTe₂,” J. Shi, Y.-Q. Bie, W. Chen, S. Fang, J. Han, Z. Cao, T. Taniguchi, K. Watanabe, V. Bulović, E. Kaxiras, P. Jarillo-Herrero, and K. A. Nelson, arXiv:1901.13609 (2019).

Selective patterning and functionalization of graphene can produce spatially-defined properties. Buckling or wrinkling of graphene on polymeric substrates enables a direct approach to tune the physical properties without lithographic steps, and the resulting curvature of the wrinkles can control local chemical reactivity. However, most buckling methods have been limited by the range of wavelength tunability, only global control of the wrinkled patterns, and delamination and cracks in the graphene. This talk will describe a scalable approach to achieve area-specific reactivity and patterning of conformal graphene wrinkles. We will discuss how a fluoropolymer layer sandwiched between graphene and different polymer substrates can facilitate crack-free and switchable graphene nanostructures. Such patterned areas with different curvatures show different reactivities based on a plasma fluorination reaction. Our approach for large-area, fine control over graphene wrinkle topographies has prospects for other two-dimensional electronic materials and optoelectronics and plasmonics applications.

Sustainable energy storage devices demand new materials for energy storage systems which alleviate their cost and increase longevity. A prominent energy storage technology for modern applications is Lithium-ion batteries. In this regard, two-dimensional (2D) materials attracted substantial and considerable attention due to their unique properties and abundant potential in various applications, including batteries¹. Controllable synthesis of 2D materials with high quality and better efficiency is essential for their large scale applications, such as their functions as electrode material in batteries. The methods for synthesis of two-dimensional materials include exfoliation, solvothermal, and chemical vapor deposition (CVD) routes, among which, CVD offers better quality, efficiency, and consistency. In this perspective, the family of 2D transition metal di-chalcogenides (TMDCs) gained significant attention for their ability to store metal ions such as Li, Na, and Mg. TMDCs of group 6 transition metals (MX₂ where M = Mo, W, and X = S, Se, Te) have a lamellar structure (space group P6₃/mmc) similar to graphite but with larger interlayer spacing. Tungsten di-chalcogenides are significant here because of the larger size of W that provides a further alteration of the 2D structure and has been successfully utilized for various applications such as photodetectors, field-effect transistors (FETs), etc. The typical group 6 TMDC lattice structure show W atoms confined in a trigonal prismatic coordination sphere neighbored to Se atoms. Because of its very high density (~9.32 g cm⁻³), WSe₂ also has a high volumetric capacity thereby proving it as a prospective LiB electrode material².
Our work focuses on a facile synthesis of 2H-WSe₂ on W foil in the ambient-pressure chemical vapor deposition system using only Argon as carrier gas during the reaction. Extensive characterization of the bulk and exfoliated material confirm that the as-synthesized 2H-WSe₂ is layered (i.e. 2D). XRD (X-Ray Diffraction) confirms the phase while HRSEM (High-Resolution Scanning Electron Microscopy), HRTEM (High-Resolution Scanning Electron Microscopy), and AFM (Atomic Force Microscopy) clarifies morphology of the material. FIB-SEM (Focused-Ion Beam Scanning Electron Microscopy) estimates the depth of the 2H-WSe₂ formation on W foil around 5-8 μm and Raman/UV-Vis measurements prove the quality of the exfoliated 2H-WSe₂. The redox processes of Lithium-ion Battery (LiB) studies show an increase in capacity until 500 cycles. On prolonged cycling the discharge capacity till the 50th cycle at 250 mA/g the material shows a stable capacity of 550 mAh/g. These observations indicate exfoliated 2H-WSe₂ has promising applications as LiB electrode material.

References
(1) Xia, H.; Xu, Q.; Zhang, J. Recent Progress on Two-Dimensional Nanoflake Ensembles for Energy Storage Applications. Nano-Micro Lett. 2018, 10 (4), 1–30. https://doi.org/10.1007/s40820-018-0219-z.
(2) Share, K.; Lewis, J.; Oakes, L.; Carter, R. E.; Cohn, A. P.; Pint, C. L. Tungsten Diselenide (WSe₂ ) as a High Capacity, Low Overpotential Conversion Electrode for Sodium-Ion Batteries. RSC Adv. 2015, 5 (123), 101262–101267. https://doi.org/10.1039/C5RA19717A.

To alleviate the adverse impacts of utilizing fossil fuel reserves on the environment (global warming, greenhouse emissions), extensive research is being devoted to the development of alternative/renewable energy systems, which include: photovoltaics, fuel cells, batteries, etc. Among them, fuel cells are an excellent solution for energy conversion, where, proton exchange membrane (PEM) fuel cells have a wide variety of applications. Some of these applications are being portable systems, automotive, etc., where an important reaction occurring at the cathode is the oxygen reduction reaction (ORR). At the current stage in fuel cell technology, platinum (Pt)-based materials are the most practical catalysts¹, which, undoubtedly are the prime choice as the cathode catalyst for the electro-reduction of oxygen but Pt-based catalysts are expensive. To reduce such costs in commercially viable fuel cells, extensive research is focused on developing alternative catalysts. Silver-based catalysts show a comparatively high thermodynamic stability and excellent catalytic activity over a wide pH range. The TMCs (O, S, Se, and among them especially Se-based chalcogenides) have attracted significant attention since the work on selenium-based cluster compounds. Selenium-based metal-rich TMCs catalysts exhibit good electrocatalytic activity which is mainly attributed to Se because it helps to improve the electronic structure of the metal by preventing the electrochemical oxidation of metals in most transition metal chalcogenides, as seen for the cases of cobalt selenide and ruthenium selenide.
Here, we present a high yield, industrially scalable, and the first-time report on the synthesis of silver selenide from silver foil and selenium powder using a two furnace atmospheric pressure-chemical vapor deposition (AP-CVD) system. We grew low temperature, LT-β-Ag₂Se on Ag foil in the AP-CVD system using thermal annealing and leading to a hollow fern-like morphology material, and its first-ever reported application as an ORR cathode catalyst. The exfoliated β-Ag₂Se is characterized using XRD, EDS, HRTEM, AFM, and HRSEM, to determine its stoichiometry and its hollow layered fern-like morphology. Differential scanning calorimetry (DSC) further confirms that the final product is the low-temperature phase β-Ag₂Se. Electrochemical oxygen reduction reaction (ORR) of exfoliated β-Ag₂Se exhibits sharp oxygen reduction current with an onset potential of 0.88 V/RHE under alkaline condition, indicating oxygen reduction property of β-Ag₂Se. The large value of limiting current (3 mA cm^-2) as well as small Tafel slope (68.5 mV dec^-1) corroborates β-Ag₂Se as a superior layered transition metal chalcogenide cathode material suitable for such energy-related applications (typically fuel cells and/or metal-air batteries).

References
(1) Ostroverkh, A.; Johánek, V.; Dubau, M.; Kúš, P.; Khalakhan, I.; Šmíd, B.; Fiala, R.; Václavu, M.; Ostroverkh, Y.; Matolín, V. Optimization of Ionomer-Free Ultra-Low Loading Pt Catalyst for Anode/Cathode of PEMFC via Magnetron Sputtering. Int. J. Hydrogen Energy 2019, 44 (35), 19344–19356. https://doi.org/10.1016/j.ijhydene.2018.12.206.

2D transition metal dichalcogenide MoTe₂ has recently gained much attention due to its wide-ranging electronic states including the 2H semiconducting state and 1T′ semimetallic state at room temperature, and the T_d Weyl semimetallic state and superconductivity at low temperatures. Due to small energy differences between the polymorphs, phase transition between distinct electronic states can be achieved easily, making MoTe₂ a model system for fundamental studies of phase transformation and structure-property relationship as well as for potential applications such as low-power, non-volatile phase-change memory. Currently however, large-scale synthesis of high-crystalline MoTe₂ thin films with precise thickness control is largely lacking.
Here, we demonstrate cm²-scale synthesis of 2H-MoTe₂ thin films with layer control down to a monolayer and grain size that spans several microns. Layer precision is achieved by controlling the initial thickness of the precursor MoO_x thin films, which are deposited on sapphire and SiO₂ substrates by atomic layer deposition (ALD) and subsequently converted to MoTe₂ through annealing in a tellurium-rich atmosphere. During the reactions, it is found that the precursor-substrate interface is critical in determining the uniformity in thickness and grain size of the resulting MoTe₂ films: Robust, uniform film growth is achieved on sapphire down to a monolayer of MoTe₂ while incomplete conversion and island growth of MoTe₂ is observed on SiO₂. Surface characterization of the MoTe₂ films using atomic force microscopy indicates that the MoTe₂ films converted on sapphire are uniform in its thickness to within one layer thickness variation. Furthermore, using dark-field transmission electron microscopy, we show that the films are highly crystalline with grains spanning over tens of microns. Thus, we show that by precisely controlling the number of ALD cycles for the precursor MoO_X films on sapphire, we can achieve MoTe₂ thin films with thickness control.
The synthesis strategy we present decouples the layer control from the variabilities of growth conditions, creating a robust growth method that is applicable to other transition metal dichalcogenides. Our MoTe₂ films thus enable systematic studies of layer-dependent properties of MoTe₂ and other transition metal dichalcogenides.

Two-dimensional (2D) transition metal dichalcogenides (TMDCs) with desired properties are ideal candidates for novel applications as 3D integrated devices on Si CMOS, sensors, and optoelectronics.^1–4 Due to the low yield of mechanical exfoliation, metalorganic chemical vapor deposition (CVD) has emerged one of the most promising approaches to synthesize WSe₂ monolayers because of its ability to tightly control source concentrations and scale to wafer size. In order to achieve single crystals monolayers, however, the orientation of domains must be controlled at the nucleation stage so that they grow laterally and coalesce without the formation of high angle or inversion domain boundaries.

In this study, we demonstrate the orientation-controlled growth of WSe₂ on 2” diameter c-plane sapphire by metalorganic chemical vapor deposition (MOCVD) in a cold-wall reactor. We used a multi-step growth protocol to independently control the WSe₂ nucleation density and domain lateral growth.⁵ By controlling the growth temperature and ratio of H₂Se and W(CO)₆, the WSe₂ triangular domains grow epitaxially on the sapphire substrate with >85% of the domains exhibiting a preferred crystallographic direction after 5 minutes of growth. With continued lateral growth time up to 30 minutes, the WSe₂ monolayer becomes fully coalesced with a low surface coverage (2%) of bilayer domains. Dark-field TEM combined with high resolution imaging of WSe₂ removed from the sapphire substrate, demonstrates that the films are nearly single crystal with a low density of inversion domain boundaries. Low temperature (4K) photoluminescence of the WSe₂ monolayer transferred to oxidized silicon reveals emission peaks associated with biexcitons, negative trions, and neutral excitons for which the peak width ranges from 8 to 11 meV. The results demonstrate the potential of MOCVD to produce wafer-scale epitaxial TMD monolayers with structural and optical properties approaching that of single crystal exfoliated flakes.


(1) Acerce, M.; Voiry, D.; Chhowalla, M. Metallic 1T Phase MoS2 Nanosheets as Supercapacitor Electrode Materials. Nat. Nanotechnol. 2015, 10 (4), 313–318. https://doi.org/10.1038/nnano.2015.40.
(2) Late, D. J.; Doneux, T.; Bougouma, M. Single-Layer MoSe2 Based NH3 Gas Sensor. Appl. Phys. Lett. 2014, 105 (23), 3–7. https://doi.org/10.1063/1.4903358.
(3) Tan, H.; Fan, Y.; Zhou, Y.; Chen, Q.; Xu, W.; Warner, J. H. Ultrathin 2D Photodetectors Utilizing Chemical Vapor Deposition Grown WS2with Graphene Electrodes. ACS Nano 2016, 10 (8), 7866–7873. https://doi.org/10.1021/acsnano.6b03722.
(4) Akinwande, D.; Huyghebaert, C.; Wang, C. H.; Serna, M. I.; Goossens, S.; Li, L. J.; Wong, H. S. P.; Koppens, F. H. L. Graphene and Two-Dimensional Materials for Silicon Technology. Nature 2019, 573 (7775), 507–518. https://doi.org/10.1038/s41586-019-1573-9.
(5) Zhang, X.; Choudhury, T. H.; Chubarov, M.; Xiang, Y.; Jariwala, B.; Zhang, F.; Alem, N.; Wang, G. C.; Robinson, J. A.; Redwing, J. M. Diffusion-Controlled Epitaxy of Large Area Coalesced WSe2 Monolayers on Sapphire. Nano Lett. 2018, 18 (2), 1049–1056. https://doi.org/10.1021/acs.nanolett.7b04521.

Due to relativistic effects, heavy metals (Pb, Bi, etc.) demonstrate strong spin orbit coupling (SOC) effects, making them promising elemental precursors to broken spin degeneracy and non-trivial topology. Atomically thin heavy metal layers exhibit large Rashba effects, translating to large spin Hall angles suitable for spin pumping, memory, and other spintronic applications. Atomically thin lead (Pb) in particular may support superconductivity along with non-trivial topology, potentially serving as a route to exotic superconducting states. However, typical synthesis techniques do not provide a platform for scalable, stackable, and air-stable 2D metals, necessary for ex-situ device work. To answer this challenge, we demonstrate the synthesis of an air-stable 1ML-Pb with strong (~500meV) spin splitting within a graphene/silicon-carbide interface through confinement heteroepitaxy. We expect this material to exhibit a large spin Hall angle suitable for spintronic applications.

Tungsten disulfide (WS₂) has a tunable bandgap predicted to go from its measured value of 1.3 eV (indirect) in bulk to 2.1 eV (direct) in monolayer form, and is thus of potential interest for photodetectors. In this work, we report an atmospheric pressure chemical vapor deposition (APCVD) approach to synthesize large domain size WS₂ on SiO₂ (University Wafers) substrates, using a custom-designed and locally fabricated 12-zone horizontal split furnace (Quazar Technologies) and solid phase precursors: tungsten trioxide (WO₃) and elemental sulfur (S). A series of experiments were carried out at temperatures between 900-1100 °C and in the presence of a carrier gas (Ar) flow (0-400 sccm). Thermal coupling between neighboring heating zones for precursors were controlled using tuned distances between precursor boats. At optimized conditions, growths of monolayer (edge size~ 63 μ m), multilayer spiral and pyramidal structural WS₂ were observed on the substrate. Raman measurements were carried out with a laser source (Horiba Labram, 514 nm, 1mW) and spot size of 4 μ m. In-plane (E¹_2g) and out-of-plane mode (A_1g) peaks at 352/cm and 420/cm confirmed monolayer growth. In other samples, bilayer and multilayer WS₂ growth was observed with Raman shifts 353/cm, 420/cm and 354/cm,421/cm. Field emission scanning electron microscopy (FEI Quanta 200 F SEM, 10 kV) and optical imaging (Olympus BX53M) confirms locally uniform growth of sharp edged WS₂, with some visible upstream dependence of edge sizes (variation from 10 μ m to 60 μ m). Multilayer growth and non-coincidental triangles are observed at the center of the substrate with large edge sizes exceeding 60 μ m. Use of vapor trapping techniques increases area coverage, though growth is then primarily multilayer, consistent with expectations from precursor supply limits during synthesis. Work is continuing towards our aim of achieving large edge size monolayer WS₂ suitable for photodetector applications.

Novel organic two-dimensional materials synthesized from polymerization of molecular units have recently attracted much attention, due to their versatile and exotic chemical, mechanical, electronic and magnetic properties. However, large scale and facile synthesis of 2D polymers in homogeneous solution, albeit common for conventional linear polymers, remains challenging despite rapid progress in this field. Previously, two-dimensional, covalently bonded polymers have been made by: reversible crystallization under hydrothermal conditions, templated growth on flat surfaces or interfaces, and polymerization after non-covalent assembly. These methods are relatively slow and hard to scale up. In comparison, irreversible reactions in a homogeneous solution phase is attractive due to its potential to synthesize large quantities of chemically stable product. Unfortunately, such reactions are more likely to produce amorphous 3D structures rather than 2D polymers. The lack of restriction on bond rotation can turn a 2D polymer into 3D, and those defects are permanent due to the irreversibility of reaction. Moreover, a 3D polymer grows much faster than a 2D polymer with the same size.

In this work, we propose and compare two possible mechanisms for making two-dimensional polymers from irreversible reactions in solution phase: the autocatalysis effect, and the bond-planarity effect. Autocatalysis could be achieved by self-templated growth on the surface of as-formed 2D molecules, which increases the rates of 2D growth; while bond-planarity would limit the bond rotation and therefore reduce the probability of 3D growth. By modelling the reaction networks as systems of differential equations, we are able to calculate the reaction time, the size distribution function of the product, and the yield of 2D polymer. The autocatalysis and bond planarity effects are parametrized by two factors: β and γ, respectively.

The results show that both mechanisms can significantly increase the size and yield of 2D polymer, and suppress the formation of 3D product. Without the aid of either effect, the average size of 2D polymer will be minuscule (<10), and the yield is negligible. In contrast, when either of the two effect is strong enough (β>10⁴ and γ>10³), a yield close to unity and a much larger average size (>100) could be achieved for 2D polymer. We find that bond-planarity effect would limit the bond rotation and therefore restore the behaviour of 2D polymerization confined to a plane. On the other hand, autocatalysis could dramatically increase the rates of 2D growth, which leads to sizes even bigger than the ideal 2D polymerization without competition from 3D growth. We also study the combined effect of both mechanisms, and obtain further enhancement in the average size (>1000) and near-unity yield, indicating a potential synergy between the two factors.
Our calculation show for the first time from theory, the feasibility of producing two-dimensional polymers from irreversible polymerization in solution. The framework is independent of the detailed pathways of reactions and structures/functionalities of molecules, hence the results are expected to be general and applicable for all types of 2D polymerization process that happen irreversibly in solution. Latest experimental work from our group has successfully demonstrated the synthesis of mechanically strong 2D polymer from irreversible polymerization in solution, via the bond-planarity mechanism, which provides validation for our calculation.

Polymers that extend covalently in two dimensions have attracted recent attention as a means of combining the mechanical strength and in-plane energy conduction of layered nanomaterials with the low densities, synthetic processability, and organic composition of their one-dimensional counterparts. Efforts to date have proven successful in forms that do not allow that full realization of these properties, such as polymerization at flat interfaces or fixation of monomers in immobilized 2D lattices. A frequently employed synthetic route is to introduce microscopic reversibility, at the cost of bond stability, to achieve 2D crystals after extensive error corrections. Herein we demonstrate a synthetic route to 2D irreversible polycondensation directly in the solution phase, resulting in covalently bonded 2D polymer platelets that are chemically stable, acid-soluble, and allowed to be further fabricated into highly oriented thin films and membranes. The resulting films of 30-nm thickness exhibit exceptional film strength at 2.52 GPa as well as a highly oriented nature, demonstrated by polarized in-plane photoluminescence study. This synthetic route provides new opportunities to 2D polymers with applications to composite materials and molecular sieving membranes.

The 2D layered materials are the potential components in next-generation semiconductor devices and quantum devices. For the integration of 2D layered materials into such functional devices, it is necessary to develop a high-quality large-area growth process.
In this talk, I will introduce the MOCVD growth of 2D layered materials, including transition metal dichalcogenides, metal oxides, and metal halides. Furthermore, I will discuss the MOCVD grown 2D layered materials and their heterostructures that could potentially be utilized in a single-photon source for quantum devices.

Two-dimensional materials such as transition metal dichalcogenides (TMDs) are intensively investigated because of their potential applications in future electronics. So far mainly group six (Mo/W) TMDs have been investigated, which show thickness depend electronic and optical properties. Metal-to-semiconductor transitions, high mobilities, and high potential for various sensing applications, now raised interest to the group 10 (Pt/Pd) TMDs or Nobel Metal Dichalcogenides (NMDs).
In this presentation, the low temperature synthesis of various TMDs by thermally assisted conversion (TAC) is presented. [1] The composition and morphology of the resulting large-scale layers are investigated by a multitude of characterization techniques including Raman spectroscopy, [2] AFM and X-ray photoelectron spectroscopy (XPS). In particular, the low temperature TAC synthesis PtSe2 potentially allows back end of line (BEOL) integration compatible with silicon technology. The effects of growth on the underlying substrates or investigated by TOF-SIMS and transmission electron microscopy. Further, as pre-pattered structures can be grown by the TAC, which allows to fabricate electronic devices using standard micro-fabrication technology. [3] PtSe2 has shown potential for a range of novel electronics devices. Examples for high performance chemical sensors, [1] IR-photodetectors[4] and MEMS[5] devices based on PtSe2 will be presented.
References
[1] Yim et al. ACS Nano, 10 (10), 9550 2016.
[2] O’Brien et al., 2D Materials, 3, 021004, 2016.
[3] Yim, et al. NPJ 2D Materials and Applications 5 (2) 2018.
[4] Yim, et al. Nano Letters, 3 (18), 1794 2018
[5] Wagner et al. Nano Letters, 8 (6), 3738 2018

Understanding and control of surfaces and interfaces is necessary to exploit the potential of 2D metal chalcogenides in electronics and optical devices. The ubiquity and scalability of vapor-phase processing in particular provides a wealth of opportunities, from the growth of van der Waals heterostructures to the chemical tailoring of interfaces to optimize device performance, including devices made from "composite" 2D materials such as those printed from inks. This talk will describe research to leverage interfaces for devices via two routes: atomic layer deposition (ALD) of metal oxides to modify dichalcogenide properties and pulsed chemical vapor deposition of ferroelectric monochalcogenides. In prior work, tuning of ALD oxide stoichiometry was shown to tune the Fermi level of MoS₂. [1] More recently, in situ monitoring of MoS₂ and MoSe₂ transistor performance during deposition of an oxide dielectric enabled the disentangling of changes in carrier concentration and mobility during the alternating dosing of the metal-organic precursor and oxidant.[2] Intriguingly, we find a substantial enhancement in field effect mobility with sub-monolayer coverage. Variable temperature transport measurements indicate that charged impurity scattering is reduced by screening even prior to formation of a complete dielectric, pointing to the importance of seeding conditions in maximizing mobility. Vapor deposition processes also provide scalable routes to van der Waals heterostructures whose properties can be engineered by band-offsets and dimensionality. Building on prior work demonstrating low-temperature epitaxy of SnS on MoS₂[3], a range of processing conditions were explored to understand how SnS film morphology is influenced by substrates including amorphous SiO₂, mica, MoS₂, and hexagonal boron nitride (hBN). Control over film thickness and morphology is particularly important to harness the in-plane ferroelectricity of SnS for devices. Large area monolayer regions are readily achieved on MoS₂ substrates by targeting temperatures and purge times that enhance selective growth at edges. Due to the relative inertness of the vdW surface of hBN, modified surface preparation and pulsing schemes are necessary to favor lateral growth of SnS, and monolayers growth is more challenging. However, isolated islands of epitaxial SnS on hBN are relatively easy to achieve, providing a path towards highly scaled memristive devices based on ferroelectric switching.

[1] Chem. Mater. 2018, 30, 3628. [2] ACS Appl. Electron. Mater. 2020, 2, 1273. [3] ACS Appl. Mater. Interfaces 2019, 43, 40543.

Two-dimensional layered transition metal dichalcogenides (TMD) such as MoS₂ and WS₂ are of interest for nanoelectronics because of their promising electronic characteristics and their predictable properties even in the few- and monolayer regime. The merits of TMDs for transistor applications have already been demonstrated by fabrication of field-effect transistors (FET) mainly using intrinsic MoS₂. However, doped variants of these TMDS are essential for the fabrication of high-performance devices and the realization of advanced transistors concepts. To allow adoption in actual device fabrication workflows, both the doped and the undoped material has to be fabricated with scalable synthesis methods. So far, it has proven difficult to synthesize doped TMDs with good control over the carrier density. Consequently, control over the doping profile is even farther out of reach.
Here we present a synthesis method for Al doped MoS₂ based on plasma-enhanced atomic-layer deposition (ALD) with good control over both the carrier density and doping profile. Using this method p-type doped MoS₂ was deposited with an precisely tunable carrier density between 10¹⁷ and 10²¹ cm^-3. The Al dopant was introduced using an ALD supercycle concept which consisted of N MoS₂ ALD cycles followed by 1 AlS_x ALD cycle. With this approach the Al concentration is controlled simply by varying the N parameter and the supercycle can be repeated as often as needed to deposit the desired film thickness. Both the doped and undoped films showed virtually no N, C, and O impurities and the Mo:S ratio was not impacted by the Al incorporation. The films were crystalline and showed the typical 2D layered structure, even at high Al concentration, as observed with cross-section transmission electron microscopy (TEM) imaging. This, combined with the minimal impact of the Al doping on the mobility, suggests that the introduction of Al does not disrupt the layered nature of the TMD and it acts like a well behaved dopant.
The excellent control over the doping profile of this approach was demonstrated in a combined TEM and energy-dispersive X-ray spectroscopy (EDX) study. A stack of 5 alternatingly doped and undoped regions, each 10 nm thick, was synthesized for characterization. The crystal structure was not interrupted by the transition between doped and undoped regions. On the other hand, the Al concentration showed a sharp transition between the doped and undoped regions, demonstrating the control over the doping profile.
To conclude, p-type doped MoS₂ was synthesized with a widely tunable carrier density and good control over the doping profile. Compatibility with large-scale fabrication was ensured by the conformallity, uniformity, and the sub-nm thickness control inherent of ALD. This approach is therefore of interest for the fabrication of nanoelectronics based on doped MoS₂. Moreover, p-type doped TMDs are of interest in other fields including photovoltaics and catalysis.

Hexagonal boron nitride (h-BN) with its atomically flat structure, excellent stability and large band gap energy (~ 6 eV) is an ideal insulator for 2D electronics and an outstanding material for the growth and encapsulation of transition metal dichalcogenides (TMdCs). Fabrication of h-BN thin films is dominated by exfoliation of multi-layer or bulk crystals and chemical vapor deposition (CVD) via thermal decomposition and surface nucleation of single-source precursors such as amino boranes and borazine. While mechanical exfoliation and chemical transfer procedures typically deliver h-BN flakes of several tens of micrometers that come along with significant interfacial impurities, a precise thickness control in the Ångström region by CVD grown thin films is hardly achievable. ALD, a vapor phase deposition technique based on self-limiting gas-surface reactions, is the ideal choice for a precise thickness control and high uniformity over areas in the centimeter regions. Therefore, we have developed ALD processes based on DFT calculations regarding the kinetics of film formation to grow h-BN layers in the centimeter region with a precise thickness control relevant for electronic applications. The synthetic access was achieved by two complementary half-reactions of B and N precursors with respect to adsorption energy calculation for each decomposition reaction step confirmed by mass spectroscopy. In addition to an insight into the growth mechanism, parameter-dependent layer thickness optimization was obtained.

Miniaturization over three-dimensions is very attractive for future on-chip technologies where efficiencies need to be optimized over millimetre sized areas. This is a new challenge, as device fabrication is mainly developed to achieve planar-geometries. Direct Ink Writing is a cost-effective, industrially scalable route that brings the possibility of making electrodes and energy storage devices in a three-dimensional geometry.
Here, we demonstrate 3D printed supercapacitors from highly concentrated, water-based inks. The active materials are atomically thin sheets of transition metal dichalcogenides, either exfoliated from bulk or obtained via direct synthesis in solution. To ensure rapid prototyping as well as streamlined manufacturing, we formulated viscoelastic slurries with tailored rheological properties allowing printability and mechanical integrity of the printed structures in various resolutions.
The printed architectures, from woodpile and serpentine structures to interdigitated electrodes, are extended over a few mm in three-dimensions and present structural integrity with as low as 100microns sized struts. Following electrode optimisation via thermal annealing, we formed high-surface area mechanically robust networks of 2D nanosheets. They combine electrochemical surface controlled kinetic processes to deliver high amounts of energy, in short bursts, to power electronics as standalone or in conjunction with other energy conversion/storage devices

The formulation of inks based on 2D materials could provide new functional behaviors to printed electronics while retaining scalable manufacturing processes. Recent advances in solution-based exfoliation have demonstrated that the electronic properties of single semiconducting 2D nanosheets can be preserved during ink processing, but there remain several challenges to the production of thin film transistors based on printed 2D materials. In addition to minimizing interface resistance and preserving bulk-like mobilities, the microstructure of the composite material must be optimized. To rationally guide the development of 2D inks and related processing, we integrated device modeling with electrical and scanning probe characterization of transistors with “composite” 2D channels. Experiments and finite element modeling were guided by a gate-dependent resistor network model that quantifies performance trade-offs determined by nanosheet thickness, degree of overlap, and interface resistivity. We measured the electrical characteristics of model MoS₂/MoS₂ homojunctions created by mechanical exfoliation to validate model parameters. Kelvin probe force microscopy (KPFM) was used to measure the surface potential profiles in transistor channels as a function of applied bias conditions and morphological characteristics such as sheet thickness. Finite element device simulations, using parameters derived from experiments, were used to interpret both the experiments and trends observed in the resistor network model. We find that screening of the upper portions of overlapping nanosheets produces current crowding at junctions between the edge of one nanosheet and the basal plane of another. Reducing the nanosheet thickness (i.e. the number of layers) distributes the current over larger areas and facilitates efficient charge transfer, increasing the effective mobility and on/off ratio, providing further motivation to develop exfoliation schemes that preserve intrinsic monolayer properties. Detailed analysis of potential drops measured in KPFM provides indirect evidence that weakly screened trapped charges at edge states further limit charge transfer, suggesting that passivation and/or functionalization schemes targeting the edges of nanosheets could greatly improve transistor performance.

Rapid direct writing of geometrically complex patterns of two dimensional (2D) quantum materials is critical for applications in a variety of areas, including flexible and wearable electronics and sensors. However, the available synthesis methods (e.g., CVD, MOCVD, MBE) are neither capable of direct writing nor possible to integrate onto flexible substrates due to high growth temperatures. Here, we demonstrate a rapid laser-based crystallization and complex patterning of 2D quantum materials (MoS₂, WS₂) on flexible substrates. A thin layer of solid-state stoichiometric amorphous 2D film is first pulsed laser deposited onto the flexible substrates – e.g., polydimethylsiloxane (PDMS) and polyethylene terephthalate (PET) - followed by a time-resolved infrared laser crystallization to controllably couple a precise amount of energy to locally heat and crystallize the amorphous 2D materials. The influence of pulse duration, frequency, and the number of pulses on the crystallization of 2D materials is investigated, and the spectroscopy and structural investigations verify the quality of crystallized 2D materials on the flexible substrates. Moreover, ultrathin 2D devices on stretchable PDMS substrate are fabricated, and their electrical characteristics under tensile strains and stretching cycles are measured. This novel method opens up a new opportunity for the future 2D materials-based wearable, transparent, and flexible optoelectronic and photonic devices.

Mr. Zabihollah Ahmadi is currently a Ph.D. student in the Department of Electrical and Computer Engineering (ECE) at Auburn University (AU). His research in the Laser-Assisted Science and Engineering (LASE), Lab under the supervision of Prof. Mahjouri-Samani, focuses on developing novel laser-based approaches for the synthesis and processing of 2D quantum materials and heterostructures on the flexible platform for optoelectronic and photonic applications.

2D materials and their heterostructures provide an exciting playground for pioneering science. In this talk I will demonstrate how atomic resolution scanning transmission electron microscopy (STEM) imaging and analysis is being used to support the advancement of these systems.

For example, I will discuss new results revealing the structural relaxation that occurs when two transition metal dichalcogenide (TMDC) monolayers are misoriented by a small twist (rotation angle). Many high profile experimental measurements of the exciting physics present in twisted TMDCs have necessarily neglected the potential for local structural relaxation. We have used atomic resolution STEM to reveal that TMDCs bilayers twisted to a small angle, less than ~3 degrees, reconstruct into energetically favourable stacking domains separated by a network of stacking faults, and that this behaviour is more complex than is observed in graphene [1]. For crystal alignments close to 3R stacking, we find that a tessellated pattern of mirror reflected triangular 3R domains while for alignments close to 2H stacking, stable 2H domains dominate, with nuclei of an earlier unnoticed metastable phase limited to ~ 5nm in size. This appears as a kagome-like pattern at twist angles less than 1 degree, transitioning to a hexagonal array of screw dislocations separating large-area 2H domains as the twist angle approaches 0 degrees. Tunneling spectroscopy measurements reveals that such reconstruction creates strong piezoelectric textures and pseudo-magnetic fields, opening new avenues for engineering of 2D material properties on the nanometre scale [1].

As the range of 2D materials increases rapidly, many of those with the most the promising magnetic or optoelectronic properties are found to be unstable in air or moisture, which makes fabrication challenging and hampers their use. Even relatively stable 2D crystals like MoSe2 and WSe2 are found to have perfect interfaces only when processed in an inert environment [2]. These materials can be stabilised to ambient conditions by encapsulating with stable 2D layers, such as graphene and hexagonal boron nitride. [3] For example, we have applied the encapsulation approach to study the presence and behaviour of point and extended defects in the 2D monochalcogenides, GaSe and InSe which are of interest due to their relatively high electron mobilities, strong second harmonic generation and quasi-direct bandgap when reduced to the monolayer limit.[4] Nonetheless the properties of InSe and GaSe crystals often degrade on exposure to air, light, and/or moisture, thought to be the result of oxidation induced defects. We have used atomic resolution STEM together with electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy to characterise individual point and extended defects in InSe and GaSe [5]. We observe that both materials contain multiple point and extended defects, even when exfoliated in argon and encapsulated in graphene. Point vacancies are observed to be healable under the electron beam, while extended defects and stacking faults have symmetries not found in bulk samples and which could be used to tune the properties of 2D post-transition-metal monochalcogenide materials for optoelectronic applications. We also demonstrate the nanopatterning of black phosphous using electron beam lithography via etching protocols developed in the STEM [6], and recent work revealing oxidation induce improvement of superconductivity in few layer TaS2 [7].

References
[1] A. Weston et al, Nature Nanotechnology (2020) doi: 10.1038/s41565-020-0682-9
[2] A.P. Rooney et al, Nano Letters, 17, 5222, (2017)
[3] Y. Cao et al, Nano Letters, 15, 8, 4914-4921 (2015)
[4] M.J. Hamer et al., ACS nano 13 (2), 2136-2142 (2019)
[5] D G Hopkinson et al, ACS nano, 13 (5), 5112-5123 (2019)
[6] N. Clark et al, Nano Letters, 18 (9), 5373-5381 , (2018)
[7] J Bekaert, et al, Nano Letters, 20 (5), 3808-3818 (2020)

We apply a multiscale modelling approach to study moiré superlattice in twisted homo- and heterobilayers of transition metal dichalcogenides (TMD), taking into account the interlayer hybridisation of the electronic orbital and lattice reconstruction due to stacking-dependent adhesion. First of all, we develop DFT-parametrized interpolation formulae for interlayer adhesion energies of MoSe₂, WSe₂, MoS₂, and WS₂ with both parallel and antiparallel orientation of their unit cells and arbitrary offset of the honeycomb lattices in the adjacent layers. Then, we combine those interpolation formulae with elasticity theory and analyze the bilayer lattice relaxation into mesoscale domain structures. We find that 3R and 2H stacking domains develop for, respectively, bilayers with parallel (P) and antiparallel (AP) orientation of the monolayer unit cells, separated by a network of dislocations, for twist angles θ <θ_P∼2.5 and θ <θ_AP∼1. Such lattice reconstruction has been verified by STEM imaging. We also show that the triangular domain structures of P-oriented homobilayers would manifest itself in local tunnelling characteristics of marginally twisted. For AP bilayer, we show that the deformation of the lattices around domain walls (which resemble twist dislocations oriented along the planes of in bulk 2H crystals) generate piezo-electric charges, reaching local density up to ±0.5x10¹²e/cm² at the junctions of the honeycomb domain wall network. Finally, we use DFT modelling of bandstructure of bilayers with various stacking configurations and interlayer distances to develop and parametrise interpolation formulae for the interlayer hoping between the layers for the relevant parts of the Brillouin zone, and, then, establish the electronic structure of the bilayeracross the moire supercell, taking into account the piezolectric potentials.

Interlayer interactions in stacked 2D crystals are one of key ingredients in exhibiting their qualitative distinct physical properties, differing from their single layer physics [1]. For example, a well-known 2D quantum spin Hall insulating transition metal dichalcogenide (TMDC) can show either topological Weyl metallic phase or trivial one depending on a minute stacking order difference [1,2]. Recent advances in fabricating stacked 2D crystals enable us to perform controlled studies on interesting electronic, magnetic and topological properties in low dimensional heterostructures. In this talk, I will first discuss interplay between intralayer correlation, interlayer interaction, and magnetism in layered materials [3-5]. To understand magnetic properties, I will introduce a new computationally efficient method to calculate various correlation effects in layered materials [3]. Another interesting examples revealing intriguing interlayer interactions will be discussed for graphene bilayer systems [6,7] and TMDCs [8]. For a graphene bilayer system with a rotation angle of 30 degrees, it is shown that a quasicrystalline order through a perfect incommensurate interlayer interaction shows localized 12 fold resonant states with fractal scaling [6]. For a bilayer MoS2 system, I will demonstrate a intrinsic coupling between strain and shear in the system that can alter its low energy Raman modes, thereby providing a way to obtain its almost all fundamental mechanical constants through light-matter interactions.
References
1) Y.-W. Son, Nature 565, 32 (2019).
2) H.-J. Kim, S.-H. Kang, I. Hamada, and Y.-W. Son, Phys. Rev. B 95, 180101 (R) (2017).
3) S. Lee and Y.-W. Son, arXiv.org:1911.05967 (2019).
4) S. Lee and Y.-W. Son, in preparation (2020).
5) E. Ko and Y.-W. Son, in preparation (2020).
6) S. J. Ahn et al, Science 361, 782 (2018).
7) P. Moon, M. Koshino, Y.-W. Son, Phys. Rev. B 99, 165430 (2019).
8) J. Lee et al., Nat. Comm. 8, 1370 (2017).

In the monolayer limit transition metal dichalcogenides (TMDs) such as MoS₂ and WS₂ have direct bandgaps, leading to significant interest in the use of these materials for opto-electronic applications. Their internal photoluminescence (PL) quantum efficiency (QE) is a sensitive measure of their quality. While some III-V thin films and II-VI quantum dot structures can achieve near-unity values of this metric, most semiconductor materials have QEs which are much smaller. We have discovered that treatment of sulfur-based TMDs with an organic superacid dissolved in a non-polar solvent suppresses non-radiative recombination at low pump powers, leading to near 100% QE. [1]. Similar, near-unity, QE values can be achieved via electrostatic gating by lowering the Fermi level to near mid-gap to suppress the formation of non-radiative trions [2].
Tuning of band parameters with strain is a common motif in semiconductor devices. Chemical vapor deposition of some monolayer TMDs on thermal expansion coefficient mismatched substrates can induce strains on the order of 1%, in spite of the weak van der Waals interaction [3]. Larger strains can be envisioned: Au grows epitaxially on MoS₂ with a predicted mismatch strain of 10% [4]. Further, the strain is predicted to be confined to the monolayer in contact with the Au, while other layers in the structure remain unstrained. This effect weakens the bonding between the top layer and the bulk [5], which enables exfoliation of large areas of monolayer material with defined patterns [6,7]. Further optimization of this nano-manufacturing process has improved the yield to nearly 50% and has led to the ability to form defined TMD heterostructures [8].

¹M. Amani, D.-H. Lien, D. Kiriya, J. Xiao, A. Azcatl, J. Noh, S.R. Madhvapathy, R. Addou, S. KC, M. Dubey, K. Cho, R.M. Wallace, S.-C. Lee, J.-H. He, J.W. Ager, X. Zhang, E. Yablonovitch, and A. Javey, Science 2015, 350, 1065.
²Lien, D.-H.; Uddin, S. Z.; Yeh, M.; Amani, M.; Kim, H.; Ager, J. W.; Yablonovitch, E.; Javey, A. Science 2019, 364, 468–471.
³Ahn, G. H.; Amani, M.; Rasool, H.; Lien, D.-H.; Mastandrea, J. P.; Ager III, J. W.; Dubey, M.; Chrzan, D. C.; Minor, A. M.; Javey, A. Nat. Commun. 2017, 8, 608.
⁴ Y. Zhou, D. Kiriya, E.E. Haller, J.W. Ager, A. Javey, and D.C. Chrzan, Phys. Rev. B 93, 054106 (2016).
⁵Sun, H.; Sirott, E. W.; Mastandrea, J.; Gramling, H. M.; Zhou, Y.; Poschmann, M.; Taylor, H. K.; Ager, J. W.; Chrzan, D. C. Phys. Rev. Mater. 2018, 2, 094004.
⁶S.B. Desai, S.R. Madhvapathy, M. Amani, D. Kiriya, M. Hettick, M. Tosun, Y. Zhou, M. Dubey, J.W. Ager, D. Chrzan, and A. Javey, Adv. Mater. 28, 4053 (2016).
⁷Gramling, H. M.; Towle, C. M.; Desai, S. B.; Sun, H.; Lewis, E. C.; Nguyen, V. D.; Ager, J. W.; Chrzan, D.; Yeatman, E. M.; Javey, A.; Taylor, H. ACS Appl. Electron. Mater. 2019, 1, 407–416.
⁸Nguyen, V.; Gramling, H; Towle C.; Li, W; Lien, D-H; Kim, H; Chrzan, D. C.; Javey, A; Xu, K; Ager, J; Taylor, H. J. Micro Nano-Manufacturing 2019 7 041006.

We investigate the strain-driven localization of electrons and enhanced photoluminescence (PL) response of trions and excitons in monolayer MoS₂ strained by SiO₂ nanopillars. Using spatially resolved PL and Raman mapping, we observe the well-recognized red shift in the PL and Raman on the apex of the pillars due to the strain response, together with an overall 6-fold increase in the PL intensity and an increased fraction of radiative trions. The increased trion fraction in the strained areas is attributed to more efficient electron strain-funneling than exciton funneling, which increases the trion production relative to the A excitons. The large overall PL intensity increase on the pillars is due to a combination of interference enhancements and charge and exciton funneling. Using time-resolved PL, we also study radiative lifetimes of excitons and charge carriers to explore the effect of strain on the intrinsic radiative lifetime of each species which leads to increased production of trions on the strained site . The results demonstrate that strain-induced local energy minima affect diffusion of excitons and free charge carriers differently, as in the case of non-uniformly strained WS₂ [1]. Our results show that strain control could be used in addition to the electric field, as part of the toolbox for quasiparticle manipulation in 2D systems.
[1] M. G. Harats, J. N. Kirchhof, M. Qiao, K. Greben, and K. I. Bolotin, “Dynamics and efficient conversion of excitons to trions in non-uniformly strained monolayer WS₂,” Nat. Photonics, 2020, doi: 10.1038/s41566-019-0581-5.

In-plane and out-of-plane deformation of 2D materials, especially of van der Waals (vdW) heterostructures, occurs readily during the fabrication process and it modifies their optical and electronic properties. Bubbles that form between layers in vdW heterostructures have been viewed as an inconvenience, with research focusing on how to avoid them during fabrication. On the other hand, practical applications of bubbles have been recently reported including quantum dot-like emitters and single photon emitters. Therefore, understanding mechanical properties of bubbles is necessary to manipulate their geometry and optoelectronic properties. Bubbles formed by monolayers are found to exhibit universal shape and aspect ratio depending on the elastic stiffness of each monolayer. However, studies on bubbles formed by multilayers have not been established yet due to non-negligible bending modulus and unconventional lubricant interfaces.
Here, we present that the geometry of bubbles formed by hexagonal boron nitride (hBN) multilayers is strongly correlated to the thickness of multilayers, which enables us to manipulate the aspect ratio of bubbles by the thickness. Since the bending rigidity of 2D multilayers is no more negligible as in the case of monolayers, two phenomena in multilayer bubbles are observed, which are distinct to monolayer bubble with fixed aspect ratio: the aspect ratio of multilayer bubble (1) decreases with the thickness of multilayer and (2) increases with the radius of bubbles. In addition, significant enhancement of luminescence is observed by cathodoluminescence from the bubbles in ultraviolet region owing to a large band gap nature of hBN. Moreover, Fabry-Perot interference is created where the pattern largely depends on the geometry of bubbles. We believe that optical properties and/or strain could be efficiently manipulated by designing the geometry of multilayer bubbles with desirable thickness.

The inspection of Friedel's law in ultrafast electron diffraction (UED) is important to gain a comprehensive understanding of material atomic structure and its dynamic response. Here, monoclinic gallium telluride (GaTe), as a low-symmetry, layered crystal in contrast to many other 2D materials, is investigated by mega-electron-volt UED. Strong out-of-phase oscillations of Bragg peak intensities are observed for Friedel pairs, which has dynamically violated the Friedel's law. As evidenced by the preserved mirror symmetry and supported by both kinematic and dynamic scattering simulations, the intensity oscillations are provoked by the lowest-order longitudinal acoustic breathing phonon. Our results provide a generalized understanding of Friedel's law in UED, and demonstrate that by designed misalignment of surface normal and primitive lattice vectors, coherent lattice wobbling and effective shear strain can be generated in crystal films by laser pulse excitation, which is otherwise hard to achieve and can be further utilized to dynamically tune and switch material properties.

Owing to the spatial and dielectric confinement, monolayers of transition metal dichalcogenides (TMDs) undergo strong many-body Columbic interactions yielding into large binding energies for the exciton (>0.1 eV) and higher order exciton-complexes compared to the conventional semiconductors. However, local environments experienced by these tightly bound excitons often drive the optoelectronic properties of these TMDs. These TMDs often exhibit low photoluminescence quantum yield (PLQY) and short intrinsic radiative lifetime due to the presence of large impurities, defects and efficient phonon assisted non-radiative Auger recombination processes. Currently researchers in the community are exploring various routes to modulate the bandgaps and tailor optoelectronic properties of these TMDs to make them more suitable for the real device applications.
In this context, an interesting avenue continuously evolving in this field is that modulation of the bandgap (even upto 0.5 eV) of these semiconductors by inducing strain (tensile & compressive).This has understood by means of controlling the mechanical stress on those monolayers and studies in this line are largely limited within the PL properties and dynamics. While the effects of strain on the fate of those excitons (in ultrashort timescale) including excitonic many-body events remains less unexplored.
To unravel the effect of stress on those excitons and their dynamics, we first study the excitonic properties of high-quality monolayer MoS2, WS₂ & WSe2 (grown by laser-assisted evaporation technique; analogous to CVD-method) samples both under substrate and suspended condition employing ultrafast optical (namely transient absorption, TA) spectroscopy as a major tool. Optical characterization data reveals that these as-synthesized monolayer flakes in fused silica substrates possess highly red-shifted absorption/PL w.r.t. the typical excitonic position. This characteristic low-band-gap signature is attributed to the effect of strain that presumably arises from the thermal expansion coefficient mismatch between the TMDs and substrate. It is quite common to observe this sort of unavoidable strain on TMDs during the CVD growth process. This is further corroborated when the strains were released from those monolayers by means of exfoliating them from the as-grown substrate. These suspended monolayer samples exhibit an enhanced PL intensity and the abs./PL position shifts back near to the typical excitonic position. TA spectroscopic measurement (for a representative case of WS2 in fused silica) reveals that under low-excitation density, sharp excitonic bleach signal (~ 640 nm) decays quickly within 200 ps, consistent with earlier reports. Interestingly, under higher excitation density, even though spectral features remain similar, but the overall TA signal becomes long-lived and hardly decays within 2000 ps. This slowing down signature at higher exciton density is found to be similar for MoS2 and WSe2 as well. This observation contrasts with the typically observed faster bleach recovery dynamics (due to exciton-exciton annihilation events) at higher excitation fluence for large no of semiconductors including TMDs. This slow dynamic is anticipated as a typical signature of strain induced formation of indirect exciton. As the intervalley scattering dominates at higher exciton density, large number of direct excitons quickly scatter into the other-valley (indirect) and thus contribution of the long-lived TA signal increases. This is further evident from the TA measurements of suspended (strain-reduced) samples in which slowing down effect of the TA signal (at higher exciton density) is relatively suppressed. This interesting finding of controlling the population of direct and indirect excitons by means of optical excitation will open exciting areas in potential optoelectronic applications.

Scanning microwave microscopy (SMM, often called scanning microwave impedance microscopy, sMIM) is a scanning probe implementation that can directly probe free carriers with nanometer spatial resolution using tip-localized GHz fields. With the subsurface sensitivity inherent to near-field techniques and the ability to measure electrically isolated systems, SMM is well-suited for studying electronic inhomogeneities in 2D semiconducting materials and their devices. We begin by highlighting SMM measurements of photogenerated free carriers in transition metal dichalcogenide lateral heterostructures.

We then focus on the application of SMM to studying ambipolar field effect transistors of 2D films of the 1D van der Waals material tellurium (tellurene). Tellurene can be solution processed, exhibits high bipolar mobilities, and is stable under ambient conditions, which makes it an attractive alternative to conventional 2D semiconductors. We combine SMM with a differential implementation to image spatial variations in conductivity as well as the associated carrier type as a function of the applied backgate voltage in tellurene devices. We reveal spatial inhomogeneities in carrier type and conductivity and show that the measured backgate-dependent device conduction minimum in fact arises from the simultaneous coexistence of n-type and p-type regions in an overall minimum conductivity configuration. We further show that the spatial inhomogeneities in conductivity and carrier type in four-terminal devices results in differing transport characteristics between different terminals across the same device in well-defined backgate-controlled p-type and n-type conduction channels. While these results highlight the challenges facing the reproducible fabrication of 2D material devices, the possibilities of solution-based doping and deterministic weighting of multiple inputs into the same device hold future promise.

2D materials can be assembled into various vertical heterostructure stacks to emit strong electroluminescence. However, until now, most work has been done using mechanical exfoliated materials, with little insights gained into the operation mechanisms when using synthetically grown crystals by chemical vapour deposition (CVD). In particular, the operation lifetime/duration of electroluminescence and subsequent failure mechanisms are poorly understood, mainly due to lack of abundance in device numbers that can shed statistical significance into the ensemble response. Here, we demonstrate that all 2D vertical layered heterostructures (Gr:hBN:WS2:hBN:Gr) using CVD grown materials can generate strong red electroluminescence (EL). Here, graphene acts as a transparent electrode, WS2 works as an ultrathin photo-active material, and h-BN serves as a tunnelling barrier for carrier injection. Tens of the devices are produced across a single chip, with the success rate of working devices over 90%. EL starts to be detected at bias values of ~5 V, with a bright red EL dot located at the junction area. Continuous operation of the devices generate EL for more than 2 hours without significant degradation of WS2 under high bias. Further investigation on breakdown tests reveal that the dominant reason for the device failure is local heating accumulation on top-graphene electrodes. This study provides a viable way for wafer-scale fabrication of high performance 2D LED arrays for ultra-thin and flexible optoelectronic devices and detailed insights about the mechanisms of failure and operation limits of EL devices made with CVD grown 2D materials within ambient conditions.

Increasing the lifetimes and yields of photoexcited charge carriers in quantum-confined heterostructures is an important goal for realizing efficient next-generation energy conversion and storage devices. One-dimensional single-walled carbon nanotubes (SWCNTs) and two-dimensional transition metal dichalcogenides (TMDCs) are fascinating materials classes with possible applications in photovoltaic and photoelectrochemical technologies. However, the photophysics in heterostructures containing these materials are not well understood. Here, we explore a model Type-II heterojunction with (6,5)-SWCNTs as the electron donor and MoS₂ as the acceptor, where the energy level offsets at these interfaces provide a promising strategy to separate charge and overcome the ultrafast decay of tightly bound excitons that are inherent to these materials. We find that the TMDC microstructure affects the charge photogeneration pathways in MoS₂/SWCNT heterojunctions, where a small fraction of multilayer islands in monolayer MoS₂ leads to longer-lived charge carriers and higher carrier yields, despite the emergence of a defect-associated spectral signature. We combine transient absorption spectroscopy with time-resolved microwave photoconductivity to explore the interplay between defect states and the generation of mobile carriers. Global fitting of our transient absorption spectra enables us to separately analyze exciton, charge, and defect evolution on the MoS₂ and SWCNT layers, and microwave photoconductivity distinguishes between free and trapped charge. Our comparisons between multilayer-containing and monolayer-only MoS₂/SWCNT heterostructures provide valuable insight for understanding the impact of local TMDC thickness variations on charge generation pathways.

Devices made of two-dimensional (2D) van der Waals (vdW) materials, including graphene, transition metal dichalcogenides (TMDs, e.g. MoS₂, MoSe₂, WeS₂ etc.) and beyond-graphene materials (e.g. phosphorene, silicene, germanene, etc.), hold tremendous potential for next-generation nanoelectronics applications. The prospect of vertically stacking¹--both homogeneous and heterogeneous--vdW materials has opened numerous avenues to engineer them for superior device performance. However, practical realization of electronic devices made of 2D vdW materials is limited by the inefficient thermal dissipation from the 2D material to the substrate. The high interface thermal resistance can cause unreliable performance and even lead to device failure. Significant research interests have been invested towards improving the electronic properties such as mobility, subthreshold slope, etc.,^2-4 but heat dissipation has received less attention.
Heat removal in 2D field-effect transistors is mainly cross-plane through the substrate, owing to the small thermal healing length⁵ and large lateral/vertical aspect ratio.⁶ Heat dissipated by electrons is carried from the 2D material to the substrate via flexural out-of-plane lattice vibrations (ZA phonons) and the efficiency of heat transfer is quantified by thermal boundary conductance (TBC). Our previous studies suggest that the TBC of graphene and MoS₂ monolayers on various substrates can be improved by encapsulation.⁶ However, encapsulation is not a viable alternative to improve TBC in optoelectronic devices. A recent study demonstrated lowering of in-plane thermal conductivity of 2D materials via phonon-electron interactions, opening a new avenue to modulate heat transfer by electrostatic gating;⁷ however, the impact of electron-phonon coupling on cross-plane heat transfer through the 2D-substrate interface is not yet fully understood.
Here we demonstrate the role of phonon-electron scattering as a tool to tune TBC of 2DvdW materials on SiO₂ substrate. We calculate the electronic bandstructures and phonon dispersions of 2D materials from first principles Density Functional Theory using the QUANTUM ESPRESSO package.⁸ Then we compute phonon-electron rates () from Fermi’s Golden rule. is combined with the anharmonic phonon rates in our thermal model⁹ to obtain TBC of the 2D-3D interface. We found that TBCs of MoS₂ and MoSe₂ are improved by 21.4% and 23.6%, respectively, when the carrier density is varied from 10¹⁰ cm^-2 to 10¹⁴ cm^-2 at 300K. On the other hand, WS₂ and WSe₂ show a weaker dependence on phonon-electron coupling with an improvement of 2.4% and 5.4%, respectively, which is due to dominant phonon-phonon scattering in these materials. The maximum enhancement in TBC due to phonon-electron coupling is obtained for TMDs composed of transition metals with smaller atomic mass and heavier chalcogen atoms. We also found that TMDs with smaller effective masses exhibit better TBC. From a fundamental perspective, we found that in all these 2D materials TBC is limited by the rate at which ZA phonons are replenished in the monolayer and not by the rate at which they scatter with the substrate. This study provides a novel tool to dynamically tune TBC by gating.
References: ¹A.K. Geim, I.V. Grigorieva, Nature 499, 419-425, 2013. ²Bolotin et al., Solid State Commun. 146, 351, 2008. ³Radisavljevic et al., Nat. Nanotechnol. 6, 147-150, 2011. ⁴Perello et al., Nat. Commun. 6, 7809, 2015. ⁵S.V. Suryavanshi and E. Pop, J.Appl. Phys. 120, 224503, 2016. ⁶Yasaei et al., Adv. Mater. Interfaces 4, 1700334, 2017. ⁷S.-Y. Yue, R. Yang, B. Liao, Phys. Rev. B 100, 115408, 2019. ⁸Giannozzi et al., J. Phys. Condens. Matter 21, 395502, 2009. ⁹C.J. Foss and Z. Aksamija, 2D Mater. 6, 025019, 2019.

In this work, we unravel the mechanism that affects exciton transport in a periodic potential created by stacking two lattice mismatched transition metal dichalcogenide (TMD) monolayers. The confined periodic excitonic states (moiré excitons^1–3) in the heterostructure creates a diffusion barrier that impacts the energy transport and dynamics of interlayer excitons (electron and hole spatially separated in different monolayers).
We experimentally quantify the diffusion barrier experienced by interlayer excitons in both R-type and H-type, h-BN encapsulated MoSe2/WSe2 heterostructures. We further support the observations with density functional theory (DFT) simulations and a three-level model that represents the exciton dynamics at various temperatures. By measuring the diffusion of interlayer excitons at various bath temperatures, we observe that the effective diffusivity of excitons can be significantly lower than their intrinsic diffusivity in the absence of the periodic potential. In fact, at low temperatures, most of the excitons remain immobile, ceasing exciton transport completely. We conclude that the strong localization of interlayer excitons at low temperature and the temperature activated Arrhenius type of diffusivity are inherent to the deep periodic potentials arising from the moiré superlattice. Based on the temperature dependent diffusivity and lifetime of interlayer excitons, we extracted the moiré potential of 20±4 meV for the H-type (54 degree twisted angle) heterostructure and 34±3 meV for the R-type heterostructure (5 degree twisted angle).
We believe the results will significantly advance the understanding of exciton transport in van der Waals heterostructures and will aid in the design of room-temperature efficient excitonic devices that leverage the various regimes of diffusivity based on twisted angle.
This work was supported through the National Science Foundation (NSF) grant No. DMR- 1904541 and Air Force Office of Scientific Research (AFOSR) Award Number FA9550-17-1-0208.
1. Seyler, K. L. et al. Signatures of moiré-trapped valley excitons in MoSe 2 /WSe 2 heterobilayers. Nature 567, 66–70 (2019).
2. Tran, K. et al. Evidence for moiré excitons in van der Waals heterostructures. Nature 567, 71–75 (2019).
3. Jin, C. et al. Observation of moiré excitons in WSe2/WS2 heterostructure superlattices. Nature 567, 76–80 (2019).

In the last decade, entire families of novel two-dimensional (2D) materials have been theoretically predicted and subsequently synthesized. The combination of their ultrathin geometry, environmental sensitivity, and mechanical robustness make 2D materials an exciting framework from which improved device performance and novel functionalities can be realized. While much of the attention on 2D materials focused on their electronic, magnetic, and optical properties, heat dissipation in these materials remains a major concern for their future implementation. The primary pathway for heat removal in 2D-based devices is the thermal pathway from the 2D layer to its underlying 3D substrate. Since 2D-3D interfaces are governed by weak van der Waals interactions, the interface thermal pathways can be one to two orders of magnitude more resistive than 3D-3D covalently bonded interface counterparts. Therefore, developing solutions to boost 2D-3D thermal boundary conductance (TBC) and identifying 2D material properties that favor interface thermal transport are critical to the future implementation of 2D-based devices.
The TBC between graphene, boron nitride (BN), and transitional metal dichalcogenides on several substrates have been studied rigorously. However, more recently synthesized 2D materials such as buckled group IV, V, and III-V 2D materials have not been investigated. Previously [2D Mater. 6 (2019) 025019] we have shown that 2D-3D TBC depends strongly on the overlap of available phonon modes in the long-wavelength regime and found selection criteria for choosing the best substrate for TBC. Here we turn our focus to 2D material features and report theoretical predictions of TBC for silicene, blue phosphorene, black phosphorene, and boron arsenide (BAs) on SiO₂ (with comparisons to graphene and BN). Our TBC model is driven by ab initio phonon dispersions and takes into account the two mechanisms that govern TBC: (i) an external resistance between ZA phonons in the 2D layer and the substrate which depopulates the ZA phonon population of the 2D layer and (ii) an internal resistance between ZA phonons and other thermal energy carriers which replenishes the ZA phonon population. Weak internal scattering can lead to a large internal resistance which effectively reduces the overall TBC. However, encapsulation has been shown to increase the internal scattering of ZA phonons effectively reducing the internal resistance and increasing the overall TBC [Adv. Mater. 30 (2018) 1801629].
We predict the external TBC, which is an upper bound on the effective TBC, of 2D/SiO₂ of (a) graphene, (b) BN, (c) blue phosphorene, (d) black phosphorene, (e) silicene, and (f) BAs interfaces without encapsulation to be (a) 20, (b) 32, (c) 70, (d) 80, (e) 100, and (f) 130 MW.m^-2.K^-1, respectively. On the other hand, the effective TBC values are approximately 50% ((a) 54%, (b) 52.5%, (c) 45%, (d) 51.6%, (e) 48%, and (f) 50.3%, respectively) of the external TBC because of additional internal resistance due to a slow repopulation of ZA phonons. However, upon encapsulation, which increases the damping and repopulation of ZA phonons, nearly half ((a) 26%, (b) 25.7%, (c) 21%, (d) 26%, (e) 20.1%, and (f) 16.6%) of those losses are regained at room temperature. Furthermore, the effective TBC of encapsulated samples shows a far weaker temperature dependence than the uncoated counterparts which is most significant at low temperatures where a near order of magnitude increase in the total TBC for temperatures below 50 K is seen for some cases. We also see that the TBC can vary over nearly an order of magnitude depending on the phonon bandwidth of the ZA branch. Our results help highlight the role of encapsulation on the temperature dependence of TBC and we find that 2D materials with narrower ZA branch bandwidths correlate strongly to high TBC.

This work reports a systematic study of electrochemical Li intercalation and high-pressure studies of black phosphorous (BP). In-situ Raman spectroscopy is used to analyze the time evolution of vibrational modes of BP under lithiation during galvanostatic discharge in a dedicated electrochemical cell. At a constant current, discharge time accounts for the number of Li ions transferred to BP cathode, hence providing good control over the composition of LixP compound. Black phosphorous has three fundamental vibrational modes. The study shows that all three fundamental vibrational modes, A1g, B2g and A2g red-shift as a result of lithiation and in-plane modes (B2g and A2g) are red-shifting about 1.6 times more than out-of-plane mode A1g. The results also indicate the structural expansion measured using electron diffraction of ex-situ lithiated samples. Further analysis using optical and electron microscopy shows also that the intercalation of BP is highly anisotropic, where channels along the zigzag direction are found to be the easy direction for intercalation. The diffusion along these channels also leads to a formation of weakly connected nanoribbons, which could be used as an alternative route to fabricate phosphorene nanoribbons. X-ray diffraction on intercalated samples indicates a reduction of thickness as lithiation weakens interlayer bonding, thus resulting in a partial exfoliation of BP flakes. The experimental results show a random filling of Li atoms between BP layers rather than a periodic filling assumed in the first principle calculations based on density function theory which explains the discrepancy between the theoretical and experimental Raman peak positions. Furthermore, studying vibrational modes of BP under high-pressure is important and adds to the knowledge on mechanical properties and strain engineering in this material which could also lead to a structural phase transition towards blue phosphorous, as predicted by first principle calculations. A diamond anvil cell (DAC) is used to apply high pressure (up to ≈15 GPa) on BP. In this method, sample is stressed against two transparent diamond anvils via a liquid pressure transfer medium. Transparency of the diamond anvils allow in-situ Raman spectroscopy to monitor changes in phonon modes under pressure. High-pressure in-situ experiments are underway on both pristine and Li intercalated BP samples, also with a systematic study on the effect of different pressure media to study the possible structural phase changes of this intriguing material under high-pressure.

We report a family of two-dimensional hybrid perovskites (2DHPs) based on phenethylammonium lead iodide ((PEA)₂PbI₄) that exhibit complex cation-dependent optical properties. By replacing the 4-position (para) H on the phenyl moiety of the phenethylammonium cation with F, Cl, CH₃, or Br, we increase the length of the cation but leave the cross-sectional area unchanged, resulting in structurally similar PbI₄^2- frameworks with increasing interlayer spacing. Longer cations result in broader, blue-shifted excitonic absorption spectra with reduced or eliminated structure, indicating greater energetic disorder. Photoluminescence spectra are largely invariant and insensitive to cation length, suggesting polaron formation stabilizes a structural and electronic minimum. Temperature-dependent linewidth analysis reveals excitons couple to a vibration on the organic framework that is weakly sensitive to these cation substitutions, and Raman spectra and electronic structure calculations support the presence of such a cationic mode. Despite carriers being confined to the inorganic framework, the length of the organic cation alters the optical and electronic properties of 2DHPs.

If instead we replace the 2-position (ortho) H with F, Cl, or Br, we increase the cation’s cross-sectional area. Unlike the 4-position substitutions, these single-atom substitutions substantially change the observable number of and spacing between discrete resonances in the excitonic absorption and PL spectra and drastically increase the amount of hot exciton PL that violates Kasha’s rule by over an order of magnitude. To fit the progressively larger cations, the inorganic framework distorts and is strained, reducing the Pb-I-Pb bond angles and increasing the 2DHP band gap. Correlation between the 2DHP structure and steady-state and time-resolved spectra suggests the complex structure of resonances arises from one or two manifolds of states, depending on the 2DHP Pb-I-Pb bond angle (as)symmetry, and the resonances within a manifold are regularly spaced with an energy separation that decreases as the mass of the cation increases. The uniform separation between resonances and the dynamics that show excitons can only relax to the next-lowest state are consistent with a vibronic progression caused by a vibrational mode on the cation. These results demonstrate that simple changes to the cation can be used to tailor the properties and dynamics of the confined excitons without directly modifying the inorganic framework.

We live in an era where space tourism finally becomes a reality. Space tourism is expected to become a large industry within the next decade along with the more and more private sector companies promoting it. These ambitious goals can only be achieved through the development of technologies, which are resilient to damage due to high, persistent fluxes of ionizing radiation in the extreme space environment. In this regard, two-dimensional materials (2DMs) have attracted great attention in the electrical device research for space applications due to potential use as a replacement of traditional semiconductors substrates for next-generation electronic devices and circuits. Studying the effects of proton irradiation on MoS₂ can help us to understand the response of optoelectronic devices fabricated from these materials operating in radiation hard environments such as space radiation environment, high-altitude flights, military aircraft, satellites, nuclear reactors, and particle accelerators. In this research, proton irradiation-induced effects on monolayer MoS₂ were studied by emulating a certain space radiation environment using a relatively low energies (150 and 590keV) proton beam with the same fluence of 1×10¹² protons/cm². The pristine and irradiated monolayer MoS2 samples were characterized by comparing the phenomenological changes in the optical and morphological properties. The crystal quality was examined with x-ray diffraction (XRD) analysis and micro-Raman spectroscopy. The possible effects on defects/ traps were probed using photoluminescence (PL) spectroscopy. The surface compositional analysis was performed via x-ray photoelectron spectroscopy (XPS), and surface morphology was inspected using atomic force microscopy (AFM), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The MoS₂ material grown with the facilitation of sodium bromide by the chemical vapor deposition shows the negligible change in optical and morphological characteristics after the irradiation. Our results open the way for the use of robust and reliable MoS₂ based devices in space environments.

Charge scattering, impurities, wrinkles, and polymer residues among the substrate decline the electrical properties, such as carrier mobility and n-/p-type doping, has been a challenge for graphene-based electronic devices. Self-assembled monolayers (SAMs) have been widely used to tailoring the electronic structure of substrate-supported graphene due to the effective modifications such as the extremely flatten surface or dipole at the interface. Also, the fluoric self-assembly monolayer (FSAM) used to promoting the electrical properties due to the hydrophobic so that it can eliminate the residue of water molecules between the substrate and graphene during the transferring process. In this study, the trichloro(1H,1H,2H,2H-perfluorooctyl)silane (FDTS) was selected as the precursor followed by the formation of the FSAM layers through the dip-coating and spin-coating methods on arbitrary substrates such as Si, SiO₂/Si, PET, and glass. As the result, it was found out that the surface roughness of the FSAM/SiO₂ substrate significantly reduce from 2.96 to 0.29 nm with the optimized condition by dip-coating, on which the graphene mobility measured from Hall measurement increases about 1.77 times from 894.6 cm²/Vs to 1588cm²/Vs; the calculated mobility of graphene/FSAM with a top-gated field effect transistor (FET) is about 15 times higher than that of device without SAM-modified substrate. The increased mobility is mainly due to the lower surface roughness and dipole effect when graphene on FSAM layer, which is observed on other versatile substrates. The dipole at the interface of graphene or other 2D(MoS₂ and WSe₂) and FSAM layer were comprehensively studied, where it allows for altering the carrier transport properties especially in FETs. This method provides a new strategy through the substrate modification to enhancing the electrical performance on graphene or 2D-based FETs, which could pave a way toward next-generation high-performance nanoelectronics.

The availability of intense tabletop terahertz-frequency pulses and simple field-enhancement structures that enable peak electric fields of tens of MV/cm have ked to demonstrations of highly nonlinear THz-induced responses from a wide variety of materials. THz-driven electronic, magnetic, and structural phase transitions and domain switching have illustrated novel possibilities for control over collective properties and dynamics. In experiments on single-layer and few-layer MoTe₂, we have found that irradiation with just one sufficiently strong THz pulse induces an irreversible phase transition [1]. The initial, noncentrosymmetric 2H phase disappears, as shown by the disappearance of characteristic Raman lines and second harmonic generation. After further THz irradiation, the topological insulator phase appears, revealed by its distinct Raman spectrum. Single-shot measurements of the MoTe₂ evolution as a function of time after THz irradiation reveal complex dynamics over many-picosecond time scales. Theoretical calculations show that the transition is induced through THz-induced liberation of carriers which stabilize the new phase and reduce the energy barrier between the two phases. The result illustrates the prospects for THz control over complex multiphase landscapes in quantum materials.

[1] “Terahertz-driven irreversible topological phase transition in two-dimensional MoTe₂,” J. Shi, Y.-Q. Bie, W. Chen, S. Fang, J. Han, Z. Cao, T. Taniguchi, K. Watanabe, V. Bulović, E. Kaxiras, P. Jarillo-Herrero, and K. A. Nelson, arXiv:1901.13609 (2019).

Selective patterning and functionalization of graphene can produce spatially-defined properties. Buckling or wrinkling of graphene on polymeric substrates enables a direct approach to tune the physical properties without lithographic steps, and the resulting curvature of the wrinkles can control local chemical reactivity. However, most buckling methods have been limited by the range of wavelength tunability, only global control of the wrinkled patterns, and delamination and cracks in the graphene. This talk will describe a scalable approach to achieve area-specific reactivity and patterning of conformal graphene wrinkles. We will discuss how a fluoropolymer layer sandwiched between graphene and different polymer substrates can facilitate crack-free and switchable graphene nanostructures. Such patterned areas with different curvatures show different reactivities based on a plasma fluorination reaction. Our approach for large-area, fine control over graphene wrinkle topographies has prospects for other two-dimensional electronic materials and optoelectronics and plasmonics applications.

In a magnetic topological insulator, nontrivial band topology conspires with magnetic order to produce exotic states of matter that are best exemplified by quantum anomalous Hall (QAH) insulators and axion insulators. Up till now, such magnetic topological insulators are obtained by doping topological insulators with magnetic atoms. The random magnetic dopants, however, inevitably introduce disorders that hinder further exploration of topological quantum effects in the material. We resolve this dilemma by probing quantum transport in MnBi₂Te₄ thin flake—a topological insulator with intrinsic magnetic order. In this layered van der Waals crystal, the ferromagnetic layers couple anti-parallel to each other, so bulk MnBi₂Te₄ is an antiferromagnet. Atomically thin MnBi₂Te₄, however, becomes ferromagnetic when the sample has odd number of septuple layers (a septuple layer represents a single structural unit in the out-of-plane direction). We observe zero-field QAH effect in a five-septuple-layer specimen; an external magnetic field further enhance the QAH quantization by forcing all layers to align ferromagnetically. MnBi₂Te₄ therefore becomes the first intrinsic magnetic topological insulator exhibiting QAH effect.

2D materials and their heterostructures provide an exciting playground for pioneering science. In this talk I will demonstrate how atomic resolution scanning transmission electron microscopy (STEM) imaging and analysis is being used to support the advancement of these systems.

For example, I will discuss new results revealing the structural relaxation that occurs when two transition metal dichalcogenide (TMDC) monolayers are misoriented by a small twist (rotation angle). Many high profile experimental measurements of the exciting physics present in twisted TMDCs have necessarily neglected the potential for local structural relaxation. We have used atomic resolution STEM to reveal that TMDCs bilayers twisted to a small angle, less than ~3 degrees, reconstruct into energetically favourable stacking domains separated by a network of stacking faults, and that this behaviour is more complex than is observed in graphene [1]. For crystal alignments close to 3R stacking, we find that a tessellated pattern of mirror reflected triangular 3R domains while for alignments close to 2H stacking, stable 2H domains dominate, with nuclei of an earlier unnoticed metastable phase limited to ~ 5nm in size. This appears as a kagome-like pattern at twist angles less than 1 degree, transitioning to a hexagonal array of screw dislocations separating large-area 2H domains as the twist angle approaches 0 degrees. Tunneling spectroscopy measurements reveals that such reconstruction creates strong piezoelectric textures and pseudo-magnetic fields, opening new avenues for engineering of 2D material properties on the nanometre scale [1].

As the range of 2D materials increases rapidly, many of those with the most the promising magnetic or optoelectronic properties are found to be unstable in air or moisture, which makes fabrication challenging and hampers their use. Even relatively stable 2D crystals like MoSe2 and WSe2 are found to have perfect interfaces only when processed in an inert environment [2]. These materials can be stabilised to ambient conditions by encapsulating with stable 2D layers, such as graphene and hexagonal boron nitride. [3] For example, we have applied the encapsulation approach to study the presence and behaviour of point and extended defects in the 2D monochalcogenides, GaSe and InSe which are of interest due to their relatively high electron mobilities, strong second harmonic generation and quasi-direct bandgap when reduced to the monolayer limit.[4] Nonetheless the properties of InSe and GaSe crystals often degrade on exposure to air, light, and/or moisture, thought to be the result of oxidation induced defects. We have used atomic resolution STEM together with electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy to characterise individual point and extended defects in InSe and GaSe [5]. We observe that both materials contain multiple point and extended defects, even when exfoliated in argon and encapsulated in graphene. Point vacancies are observed to be healable under the electron beam, while extended defects and stacking faults have symmetries not found in bulk samples and which could be used to tune the properties of 2D post-transition-metal monochalcogenide materials for optoelectronic applications. We also demonstrate the nanopatterning of black phosphous using electron beam lithography via etching protocols developed in the STEM [6], and recent work revealing oxidation induce improvement of superconductivity in few layer TaS2 [7].

References
[1] A. Weston et al, Nature Nanotechnology (2020) doi: 10.1038/s41565-020-0682-9
[2] A.P. Rooney et al, Nano Letters, 17, 5222, (2017)
[3] Y. Cao et al, Nano Letters, 15, 8, 4914-4921 (2015)
[4] M.J. Hamer et al., ACS nano 13 (2), 2136-2142 (2019)
[5] D G Hopkinson et al, ACS nano, 13 (5), 5112-5123 (2019)
[6] N. Clark et al, Nano Letters, 18 (9), 5373-5381 , (2018)
[7] J Bekaert, et al, Nano Letters, 20 (5), 3808-3818 (2020)

We apply a multiscale modelling approach to study moiré superlattice in twisted homo- and heterobilayers of transition metal dichalcogenides (TMD), taking into account the interlayer hybridisation of the electronic orbital and lattice reconstruction due to stacking-dependent adhesion. First of all, we develop DFT-parametrized interpolation formulae for interlayer adhesion energies of MoSe₂, WSe₂, MoS₂, and WS₂ with both parallel and antiparallel orientation of their unit cells and arbitrary offset of the honeycomb lattices in the adjacent layers. Then, we combine those interpolation formulae with elasticity theory and analyze the bilayer lattice relaxation into mesoscale domain structures. We find that 3R and 2H stacking domains develop for, respectively, bilayers with parallel (P) and antiparallel (AP) orientation of the monolayer unit cells, separated by a network of dislocations, for twist angles θ <θ_P∼2.5 and θ <θ_AP∼1. Such lattice reconstruction has been verified by STEM imaging. We also show that the triangular domain structures of P-oriented homobilayers would manifest itself in local tunnelling characteristics of marginally twisted. For AP bilayer, we show that the deformation of the lattices around domain walls (which resemble twist dislocations oriented along the planes of in bulk 2H crystals) generate piezo-electric charges, reaching local density up to ±0.5x10¹²e/cm² at the junctions of the honeycomb domain wall network. Finally, we use DFT modelling of bandstructure of bilayers with various stacking configurations and interlayer distances to develop and parametrise interpolation formulae for the interlayer hoping between the layers for the relevant parts of the Brillouin zone, and, then, establish the electronic structure of the bilayeracross the moire supercell, taking into account the piezolectric potentials.

In the monolayer limit transition metal dichalcogenides (TMDs) such as MoS₂ and WS₂ have direct bandgaps, leading to significant interest in the use of these materials for opto-electronic applications. Their internal photoluminescence (PL) quantum efficiency (QE) is a sensitive measure of their quality. While some III-V thin films and II-VI quantum dot structures can achieve near-unity values of this metric, most semiconductor materials have QEs which are much smaller. We have discovered that treatment of sulfur-based TMDs with an organic superacid dissolved in a non-polar solvent suppresses non-radiative recombination at low pump powers, leading to near 100% QE. [1]. Similar, near-unity, QE values can be achieved via electrostatic gating by lowering the Fermi level to near mid-gap to suppress the formation of non-radiative trions [2].
Tuning of band parameters with strain is a common motif in semiconductor devices. Chemical vapor deposition of some monolayer TMDs on thermal expansion coefficient mismatched substrates can induce strains on the order of 1%, in spite of the weak van der Waals interaction [3]. Larger strains can be envisioned: Au grows epitaxially on MoS₂ with a predicted mismatch strain of 10% [4]. Further, the strain is predicted to be confined to the monolayer in contact with the Au, while other layers in the structure remain unstrained. This effect weakens the bonding between the top layer and the bulk [5], which enables exfoliation of large areas of monolayer material with defined patterns [6,7]. Further optimization of this nano-manufacturing process has improved the yield to nearly 50% and has led to the ability to form defined TMD heterostructures [8].

¹M. Amani, D.-H. Lien, D. Kiriya, J. Xiao, A. Azcatl, J. Noh, S.R. Madhvapathy, R. Addou, S. KC, M. Dubey, K. Cho, R.M. Wallace, S.-C. Lee, J.-H. He, J.W. Ager, X. Zhang, E. Yablonovitch, and A. Javey, Science 2015, 350, 1065.
²Lien, D.-H.; Uddin, S. Z.; Yeh, M.; Amani, M.; Kim, H.; Ager, J. W.; Yablonovitch, E.; Javey, A. Science 2019, 364, 468–471.
³Ahn, G. H.; Amani, M.; Rasool, H.; Lien, D.-H.; Mastandrea, J. P.; Ager III, J. W.; Dubey, M.; Chrzan, D. C.; Minor, A. M.; Javey, A. Nat. Commun. 2017, 8, 608.
⁴ Y. Zhou, D. Kiriya, E.E. Haller, J.W. Ager, A. Javey, and D.C. Chrzan, Phys. Rev. B 93, 054106 (2016).
⁵Sun, H.; Sirott, E. W.; Mastandrea, J.; Gramling, H. M.; Zhou, Y.; Poschmann, M.; Taylor, H. K.; Ager, J. W.; Chrzan, D. C. Phys. Rev. Mater. 2018, 2, 094004.
⁶S.B. Desai, S.R. Madhvapathy, M. Amani, D. Kiriya, M. Hettick, M. Tosun, Y. Zhou, M. Dubey, J.W. Ager, D. Chrzan, and A. Javey, Adv. Mater. 28, 4053 (2016).
⁷Gramling, H. M.; Towle, C. M.; Desai, S. B.; Sun, H.; Lewis, E. C.; Nguyen, V. D.; Ager, J. W.; Chrzan, D.; Yeatman, E. M.; Javey, A.; Taylor, H. ACS Appl. Electron. Mater. 2019, 1, 407–416.
⁸Nguyen, V.; Gramling, H; Towle C.; Li, W; Lien, D-H; Kim, H; Chrzan, D. C.; Javey, A; Xu, K; Ager, J; Taylor, H. J. Micro Nano-Manufacturing 2019 7 041006.

Interlayer interactions in stacked 2D crystals are one of key ingredients in exhibiting their qualitative distinct physical properties, differing from their single layer physics [1]. For example, a well-known 2D quantum spin Hall insulating transition metal dichalcogenide (TMDC) can show either topological Weyl metallic phase or trivial one depending on a minute stacking order difference [1,2]. Recent advances in fabricating stacked 2D crystals enable us to perform controlled studies on interesting electronic, magnetic and topological properties in low dimensional heterostructures. In this talk, I will first discuss interplay between intralayer correlation, interlayer interaction, and magnetism in layered materials [3-5]. To understand magnetic properties, I will introduce a new computationally efficient method to calculate various correlation effects in layered materials [3]. Another interesting examples revealing intriguing interlayer interactions will be discussed for graphene bilayer systems [6,7] and TMDCs [8]. For a graphene bilayer system with a rotation angle of 30 degrees, it is shown that a quasicrystalline order through a perfect incommensurate interlayer interaction shows localized 12 fold resonant states with fractal scaling [6]. For a bilayer MoS2 system, I will demonstrate a intrinsic coupling between strain and shear in the system that can alter its low energy Raman modes, thereby providing a way to obtain its almost all fundamental mechanical constants through light-matter interactions.
References
1) Y.-W. Son, Nature 565, 32 (2019).
2) H.-J. Kim, S.-H. Kang, I. Hamada, and Y.-W. Son, Phys. Rev. B 95, 180101 (R) (2017).
3) S. Lee and Y.-W. Son, arXiv.org:1911.05967 (2019).
4) S. Lee and Y.-W. Son, in preparation (2020).
5) E. Ko and Y.-W. Son, in preparation (2020).
6) S. J. Ahn et al, Science 361, 782 (2018).
7) P. Moon, M. Koshino, Y.-W. Son, Phys. Rev. B 99, 165430 (2019).
8) J. Lee et al., Nat. Comm. 8, 1370 (2017).

Scanning microwave microscopy (SMM, often called scanning microwave impedance microscopy, sMIM) is a scanning probe implementation that can directly probe free carriers with nanometer spatial resolution using tip-localized GHz fields. With the subsurface sensitivity inherent to near-field techniques and the ability to measure electrically isolated systems, SMM is well-suited for studying electronic inhomogeneities in 2D semiconducting materials and their devices. We begin by highlighting SMM measurements of photogenerated free carriers in transition metal dichalcogenide lateral heterostructures.

We then focus on the application of SMM to studying ambipolar field effect transistors of 2D films of the 1D van der Waals material tellurium (tellurene). Tellurene can be solution processed, exhibits high bipolar mobilities, and is stable under ambient conditions, which makes it an attractive alternative to conventional 2D semiconductors. We combine SMM with a differential implementation to image spatial variations in conductivity as well as the associated carrier type as a function of the applied backgate voltage in tellurene devices. We reveal spatial inhomogeneities in carrier type and conductivity and show that the measured backgate-dependent device conduction minimum in fact arises from the simultaneous coexistence of n-type and p-type regions in an overall minimum conductivity configuration. We further show that the spatial inhomogeneities in conductivity and carrier type in four-terminal devices results in differing transport characteristics between different terminals across the same device in well-defined backgate-controlled p-type and n-type conduction channels. While these results highlight the challenges facing the reproducible fabrication of 2D material devices, the possibilities of solution-based doping and deterministic weighting of multiple inputs into the same device hold future promise.

Liquid phase exfoliation has become an important top down production technique giving access to large quantities of nanosheets in colloidal dispersion. Importantly, this is a highly versatile technique that can be applied to numerous layered materials from graphite to transition metal dichalcogenides, h-BN, III-VI semiconductors, hydroxides etc. All materials can be exfoliated in a similar way using aqueous surfactant or suitable solvents as stabilisers. Nanosheets in dispersion are extremely polydisperse with broad lateral size and thickness distributions. To narrow size and thickness distributions, liquid cascade centrifugation has proven to be a powerful tool for efficient size selection yielding nanosheet dispersions with well-defined dimensions.
Building on these established methodologies of exfoliation and size selection, it became possible in recent years to adress some fundamental questions regarding the exfoliation process, such as: In which way is the yield, nanosheet size and layer number dependent on the liquid medium? What is a good descriptor for the exfoliation efficiency and what does it depend on? What is the role of defects? In this talk, these fundamental questions will be addressed and a picture of the current understanding will be drawn.

Two-dimensional magnetic van der Waals materials have attracted much interest recently. Magnetism in low dimensional systems is an interesting topic for the fundamental physics, and atomically thin magnetic materials are promising candidates for novel spintronic devices. Antiferromagnetic 2-dimensional materials are particularly interesting both for fundamental physics and also for antiferromagnetism-based spintronic devices. However, traditional research tools such as neutron scattering to probe antiferromagnetic ordering cannot be employed for atomically thin materials due to the small sample volume. Although magneto-optical Kerr effect measurements can be used to monitor the magnetic ordering in ferromagnetic materials, the lack of net magnetization precludes the use of the Kerr effect in probing antiferromagnetic ordering. Optical spectroscopy is becoming increasingly important for the study of antiferromagnetic 2-dimensional materials. Raman spectroscopy, for example, has been established as an invaluable tool to probe the magnetic transition in antiferromagnetic van der Waals materials as it has been found that the magnetic ordering can be reliably monitored by Raman spectroscopy [1]. Furthermore, recent spectroscopic studies revealed a novel coherent state in some of these materials stabilized by the antiferromagnetic ordering. In this presentation, I will review recent achievements in the study of antiferromagnetism in 2 dimensions using optical spectroscopy. FePS₃ exhibits an Ising-type antiferromagnetic ordering down to the monolayer limit, in good agreement with the Onsager solution for a 2-dimensional order-disorder transition [2]. The transition temperature remains almost independent of the thickness from bulk to the monolayer limit, indicating that the weak interlayer interaction has little effect on the antiferromagnetic ordering. On the other hand, NiPS₃, which shows an XXZ-type antiferromagnetic ordering in bulk, exhibits antiferromagnetic ordering down to 2 layers with a slight decrease in the transition temperature, but the magnetic ordering is suppressed in the monolayer limit [3]. Furthermore, an almost resolution-limited, sharp optical transition was observed in photoluminescence and optical absorption measurements on NiPS₃, which is interpreted as being due to a novel excitonic state coupled with the antiferromagnetic ordering [4]. Furthermore, a Heisenberg-type antiferromagnet MnPS₃ also exhibits ordering down to 2 layers [5] with a weak dependence of the Néel temperature on the thickness [6].



References
[1] Kim K, Lee J-U, Cheong H, Nanotechnology (2019); 30, 452001.
[2] Lee J-U, et al., Nano Letters (2016); 16, 7433.
[3] Kim K, et al., Nature Communications (2019); 10, 345.
[4] Kang S, et al., in press.
[5] Kim K, et al., 2D Materials (2019); 6, 041001.
[6] Lim S, et al., submitted.

The discovery of 2-dimensional (2D) materials, such as CrI3, that retain magnetic ordering at monolayer thickness has resulted in a surge of both pure and applied research in 2D magnetism. In this talk, I will detail our magneto-Raman spectroscopy study on multilayered CrI3, focusing on two additional features in the spectra that appear below the magnetic ordering temperature and were previously assigned to high frequency magnons. Instead, we conclude these modes are actually zone-folded phonons. We observe a striking evolution of the Raman spectra with increasing magnetic field applied perpendicular to the atomic layers in which clear, sudden changes in intensities of the modes and their polarization dependence are attributed to the interlayer ordering changing from antiferromagnetic to ferromagnetic at a critical magnetic field. Our work highlights the sensitivity of the Raman modes to weak interlayer spin ordering in CrI3.

Due to relativistic effects, heavy metals (Pb, Bi, etc.) demonstrate strong spin orbit coupling (SOC) effects, making them promising elemental precursors to broken spin degeneracy and non-trivial topology. Atomically thin heavy metal layers exhibit large Rashba effects, translating to large spin Hall angles suitable for spin pumping, memory, and other spintronic applications. Atomically thin lead (Pb) in particular may support superconductivity along with non-trivial topology, potentially serving as a route to exotic superconducting states. However, typical synthesis techniques do not provide a platform for scalable, stackable, and air-stable 2D metals, necessary for ex-situ device work. To answer this challenge, we demonstrate the synthesis of an air-stable 1ML-Pb with strong (~500meV) spin splitting within a graphene/silicon-carbide interface through confinement heteroepitaxy. We expect this material to exhibit a large spin Hall angle suitable for spintronic applications.

Electrical manipulation of spin in a magnetic material is a fundamental challenge for future spintronics. In this respect, recent discoveries of gate-tunable magnetism in ferromagnetic two-dimensional (2D) materials such as CrI₃ and Fe₃GeTe₂ demonstrate that they are an excellent platform for the study of interplay between charge and magnetic order. In this talk, we will discuss electrostatic control of Cr₂Ge₂Te₆ (CGT), a van der Waals ferromagnetic semiconductor, in an electric double-layer transistor (EDLT) geometry [1]. We show that degenerately electron-doped CGT exhibits peculiar magnetoresistance (MR) curves with pronounced hysteresis and abrupt switching features, which is characteristic of spontaneous magnetic order. Remarkably, the hysteresis persists up to 200 K, well above the Curie temperature of 61 K for pristine CGT, indicating enhancement of exchange energy by free carriers. Further, we find that the easy axis of the doping-induced ferromagnetic state lies in the 2D plane, in contrast to the out-of-plane easy axis of the pristine material. We will discuss the role of doping on the magnetic anisotropy and prospects for observing similar effects in other magnetic 2D materials.

[1] I. Verzhbitskiy et al. "Controlling the magnetic anisotropy in Cr₂Ge₂Te₆ by electrostatic gating" Nat. Electron. In Press (https://doi.org/10.1038/s41928-020-0427-7).

The 2D layered materials are the potential components in next-generation semiconductor devices and quantum devices. For the integration of 2D layered materials into such functional devices, it is necessary to develop a high-quality large-area growth process.
In this talk, I will introduce the MOCVD growth of 2D layered materials, including transition metal dichalcogenides, metal oxides, and metal halides. Furthermore, I will discuss the MOCVD grown 2D layered materials and their heterostructures that could potentially be utilized in a single-photon source for quantum devices.

Two-dimensional materials such as transition metal dichalcogenides (TMDs) are intensively investigated because of their potential applications in future electronics. So far mainly group six (Mo/W) TMDs have been investigated, which show thickness depend electronic and optical properties. Metal-to-semiconductor transitions, high mobilities, and high potential for various sensing applications, now raised interest to the group 10 (Pt/Pd) TMDs or Nobel Metal Dichalcogenides (NMDs).
In this presentation, the low temperature synthesis of various TMDs by thermally assisted conversion (TAC) is presented. [1] The composition and morphology of the resulting large-scale layers are investigated by a multitude of characterization techniques including Raman spectroscopy, [2] AFM and X-ray photoelectron spectroscopy (XPS). In particular, the low temperature TAC synthesis PtSe2 potentially allows back end of line (BEOL) integration compatible with silicon technology. The effects of growth on the underlying substrates or investigated by TOF-SIMS and transmission electron microscopy. Further, as pre-pattered structures can be grown by the TAC, which allows to fabricate electronic devices using standard micro-fabrication technology. [3] PtSe2 has shown potential for a range of novel electronics devices. Examples for high performance chemical sensors, [1] IR-photodetectors[4] and MEMS[5] devices based on PtSe2 will be presented.
References
[1] Yim et al. ACS Nano, 10 (10), 9550 2016.
[2] O’Brien et al., 2D Materials, 3, 021004, 2016.
[3] Yim, et al. NPJ 2D Materials and Applications 5 (2) 2018.
[4] Yim, et al. Nano Letters, 3 (18), 1794 2018
[5] Wagner et al. Nano Letters, 8 (6), 3738 2018

Understanding and control of surfaces and interfaces is necessary to exploit the potential of 2D metal chalcogenides in electronics and optical devices. The ubiquity and scalability of vapor-phase processing in particular provides a wealth of opportunities, from the growth of van der Waals heterostructures to the chemical tailoring of interfaces to optimize device performance, including devices made from "composite" 2D materials such as those printed from inks. This talk will describe research to leverage interfaces for devices via two routes: atomic layer deposition (ALD) of metal oxides to modify dichalcogenide properties and pulsed chemical vapor deposition of ferroelectric monochalcogenides. In prior work, tuning of ALD oxide stoichiometry was shown to tune the Fermi level of MoS₂. [1] More recently, in situ monitoring of MoS₂ and MoSe₂ transistor performance during deposition of an oxide dielectric enabled the disentangling of changes in carrier concentration and mobility during the alternating dosing of the metal-organic precursor and oxidant.[2] Intriguingly, we find a substantial enhancement in field effect mobility with sub-monolayer coverage. Variable temperature transport measurements indicate that charged impurity scattering is reduced by screening even prior to formation of a complete dielectric, pointing to the importance of seeding conditions in maximizing mobility. Vapor deposition processes also provide scalable routes to van der Waals heterostructures whose properties can be engineered by band-offsets and dimensionality. Building on prior work demonstrating low-temperature epitaxy of SnS on MoS₂[3], a range of processing conditions were explored to understand how SnS film morphology is influenced by substrates including amorphous SiO₂, mica, MoS₂, and hexagonal boron nitride (hBN). Control over film thickness and morphology is particularly important to harness the in-plane ferroelectricity of SnS for devices. Large area monolayer regions are readily achieved on MoS₂ substrates by targeting temperatures and purge times that enhance selective growth at edges. Due to the relative inertness of the vdW surface of hBN, modified surface preparation and pulsing schemes are necessary to favor lateral growth of SnS, and monolayers growth is more challenging. However, isolated islands of epitaxial SnS on hBN are relatively easy to achieve, providing a path towards highly scaled memristive devices based on ferroelectric switching.

[1] Chem. Mater. 2018, 30, 3628. [2] ACS Appl. Electron. Mater. 2020, 2, 1273. [3] ACS Appl. Mater. Interfaces 2019, 43, 40543.

This talk will present our latest progress on materials, transistor and memory devices based on 2D atomically-thin films. In particular, we explore a new substrate, namely, solid lithium-ion electrolytic substrates, for the direct growth and realization of high-performance TMD devices and amplifiers. Electrostatic gating with ionic liquids (ILs), leading to the accumulation of high surface charge carrier densities, has been often exploited in thin-film transistors. However, the intrinsic liquid nature of ILs inhibit it from constituting an ideal platform. In this regard, a lithium-ion solid electrolyte, often used for battery applications, offers similar advantages as ionic-liquids, with the added benefit of solid-state compatibility. With the solid electrolyte, we demonstrate its application in high-performance n-type MoS2 and p-type WSe2 transistors with sub-threshold values approaching the ideal limit of 60 mV/dec and complementary inverter amplifier gain of 34, the highest among comparable amplifiers with a 1 V power supply. These results establish lithium-ion substrates as a promising alternative to ionic liquids for the study of thin-film devices.

Two-dimensional (2D) materials offer the freedom to create novel condensed matter systems, with unique properties, by mechanically assembling different (or same) 2D materials layer-by-layer to form atomically sharp vertical or lateral heterostructures. The van der Waals (vdW) heterostructures with small lattice mismatch and a relatively small twist angle between the constituent layers, have shown to exhibit coexisting complex phases of matter including Mott insulating state, superconductivity, bound quasiparticles, and topological states. In addition, the extreme surface sensitivity of two-dimensional (2D) materials provides an unprecedented opportunity to engineer the physical properties of these materials using external perturbations. In this talk, I will discuss the utilization of angle-resolved photoemission spectroscopy (ARPES) with high spatial resolution to investigate the electronic band structure of 2D heterostructures and their devices. This can shed light on the intricate relationship between controlled external perturbations, substrate, and electronic properties of 2D materials [1, 2]. In particular, I will discuss our in-operando nanoARPES results that exhibit highly tunable many-body effects in graphene devices [3] and tunable van Hove singularities in twisted bilayer graphene [4].
References:
[1] Katoch et. al., Nature Physics 14, 355-359 (2018).
[2] Ulstrup, et. al., Science Advances, Vol. 6, no. 14, eaay6104, (2020).
[3] Muzzio, et. al., Physical review B Rapid Communications 101, 201409(R) (2020).
[4] Jones, et. al., arXiv:2006.00791 (2020). Accepted for publication in Advanced Materials.

Much of the current interest in atomically-thin transition metal dichalcogenide (TMD) semiconductors such as MoS₂ and WSe₂ is driven by the physics of coupled spin & valley degrees of freedom and the potential for new spin- and valley-based devices. This talk will discuss recent optical studies that probe the valley-related physics of electrons, holes, and excitons in monolayer TMD semiconductors.

In a first study [1,2], we measure the polarized magneto-absorption spectra of high-quality exfoliated TMD monolayers in very large magnetic fields to 91 tesla, which allows to follow the diamagnetic shifts of not only the 1s ground state of the neutral exciton but also its excited 2s, 3s, ..., ns Rydberg states. The energies and diamagnetic shifts provide a direct determination of the effective (reduced) exciton masses and also the dielectric properties of these monolayer semiconductors – fundamental material parameters that until now were available only from theoretical calculations. Moreover, we also measure other important material properties, including exciton binding energies, exciton radii, and free-particle bandgaps. These results provide essential and quantitative parameters for the rational design of opto-electronic van der Waals heterostructures incorporating 2D semiconductor monolayers.

In a second study, we introduce an entirely passive, noise-based approach for exploring intrinsic valley dynamics in monolayer TMD semiconductors [3]. Under conditions of strict thermal equilibrium, we use optical Faraday rotation to “listen” to the thermodynamic fluctuations of the valley polarization in a Fermi sea of resident electrons or holes, due to their spontaneous scattering between the K and K’. Frequency spectra of this valley noise reveal narrow Lorentzian lineshapes and therefore very long intrinsic valley relaxation, especially for holes (~microseconds). These results provide quantitative measurements of intrinsic valley dynamics, free from any external perturbation, pumping, or excitation.

*In collaboration with Mateusz Goryca, Jing Li, & Andreas Stier (NHMFL), Nathan Wilson & Xiaodong Xu (University of Washington), and Cedric Robert, Bernhard Urbaszek, & Xavier Marie (INSA-Toulouse)

[1] M. Goryca et al., Nature Communications 10, 4172 (2019).
[2] A. V. Stier et al., Phys. Rev. Lett. 120, 057405 (2018).
[3] M. Goryca, N. P. Wilson, P. Dey, X. Xu, & S. A. Crooker, Science Advances 5, eaau4899 (2019).

Polaritons, hybrid light-matter excitations, enable nanoscale control of light. Particularly large polariton field confinement and long lifetimes can be found in graphene and van der Waals materials, opening new avenues for low-loss photonic applications. Here, we present our recent progress in the discovery on engineering of these nanoscale polaritons in 2D materials [1,2]. First, plasmon polaritons arising from graphene and topological insulators were investigated in graphene/Bi₂Te₃, topological insulator Bi₂Te₃, graphene nanoribbon and 3D graphene using either spectroscopic or real space imaging techniques [3-6]. We show how the plasmonic coupling happens in two Dirac materials, how high-order plasmonic modes are observed in 3D graphene structure, how multiple plasmonic modes at sub-wavelength are achieved in graphene nanoribbon and how edge chirality controls the plasmonic shift.
We would also like to highlight our progress on the observation of anisotropic and ultra-low-loss polariton propagation along the surface of natural vdW material α-MoO₃ [7]. We visualized and verified phonon polaritons with elliptic and hyperbolic in-plane dispersion, which have been theoretically predicted but never experimentally observed in natural materials before. To better utilize the low-loss directional polaritons, we developed a chemical intercalation method to reversibly control and switch the phonon polaritons while maintaining their long lifetimes [8]. Combined with photolithography, spatially controlled polariton switching was achieved with various in-plane heterostructure configurations and patterns. Most recently, we show how the hyperbolic polaritons in α-MoO₃ thin slabs were delicately manipulated by controlling the interlayer twist angle [9]. We experimentally observed tunable topological transitions from open (hyperbolic) to closed (elliptical) dispersion contours in twisted α-MoO₃ bilayers at a photonic magic twist angle. The theoretical analysis showed these transitions of in-plane hyperbolic polaritons are controlled by a topological quantity. At the transitions, the photonic dispersion of the resulting topological polaritons flattens, exhibiting low-loss tunable polariton canalization and diffractionless propagation.
In summary, biaxial van der Waals semiconductors like α-MoO₃[7, 9] and V₂O₅[10] represent an emerging family of material supporting exotic polaritonic behaviors and these natural-born hyperbolic materials offer an unprecedented platform for controlling the flow of energy at the nanoscale.

The isolation of a growing number of two-dimensional (2D) materials has inspired worldwide efforts to study their fundamental physical properties and integrate distinct 2D materials into van der Waals (vdW) heterostructures.^1-2 While the few atoms thick structure of 2D semiconductors naturally makes their electronic structure quantum confined, it also has profound implications on their optical properties.^3-4 Further being 2D in nature it also presents opportunities for engineering them into 0D and 1D quantum confined structures.⁵ My talk will discuss our group’s more recent work on optical phenomena and photonic devices made from 2D semiconductors. I will present how 1D nanostructuring of excitonic 2D semiconductors into nanophotonic dielectric gratings can enable exploration of new regimes of light-matter confinement including formation of hybrid exciton-plasmon-polariton states.⁶ I will extend this discussion of light-matter interaction to excitonic/dielectric superlattices which are scalable over large areas. I will extend this concept to other excitonic materials and show dynamic tunability of their optical properties. If time permits, I will also discuss new approaches and methods for nanostructuring 2D semiconductors down to ~5 nm lateral dimensions with superior control of crystallinity and structure.⁷ I will conclude by giving a broad perspective on future prospects of 2D materials from fundamental science to applications.
References:
1. Jariwala, D.; Marks, T. J.; Hersam, M. C., Nature Materials 2017, 16, 170-181.
2. Jariwala, D.; et al. ACS Nano 2014, 8, 1102–1120.
3. Brar, V. W.; Sherrott, M. C.; Jariwala, D., Chemical Society Reviews 2018, 47, 6824-6844.
4. Jariwala, D.; et al. ACS Photonics 2017, 4, 2692-2970.
5. Stanford, M. G.; Rack, P. D.; Jariwala, D., npj 2D Materials and Applications 2018, 2, 20.
6. Zhang, H.;….Jariwala, D., et al. Hybrid Exciton-Plasmon-Polaritons in van der Waals Semiconductor Gratings. Nature Communications 2020, (in press).
7. Kumar, P….Stach, E. A.; Jariwala, D., et al. Npj 2D Materials and Applications 2020, https://doi.org/10.1038/s41699-020-0150-2.