Symposium QN01—2D Layered Materials Beyond Graphene---Theory, Discovery and Design

The rapid rise of novel 2D materials presents the exciting opportunity for materials science to explore an entirely new class of materials. This comes at the time when sophisticated materials informatics approaches provide the predictive capability to enable the computational discovery, characterization, and design of 2D materials and provide the needed input and guidance to experimental studies. I will present our data-mining, evolutionary algorithm, and machine-learning approaches to identify novel 1D and 2D materials with low formation energies and show how unexpected structures emerge when a material is reduced to sub-nanometers in thickness. These materials informatics approaches identify low-dimensional materials that can be exfoliated from bulk crystal structures or by CVD and MBE synthesis. Among the low-dimensional materials, we predict a variety of magnetically ordered structures from ferromagnetic and antiferromagnetic order, to Berezinsky-Kosterlitz-Thouless phases, and half-metals. These new low-dimensional materials provide the opportunity to investigate the interplay of magnetic order and reduced dimensionality and may provide materials suitable for optoelectronic and spintronic applications. The structures and calculated properties for all 2D materials are available in the MaterialsWeb database at https://materialsweb.org.

Two-dimensional materials are increasingly popular for applications in catalysis due to useful properties like high surface area, differing surface reactivity from bulk counterparts, and their ability to be formed into nanostructures. Additionally, computational screening allows for the discovery of novel two-dimensional phases of bulk structures. These ideas motivate our first-principles based study of the structure and electronic properties of novel 2D phases of several binary semiconductor materials which have been identified as promising CO2 reduction photocatalysts in the bulk [1]. Using high throughput screening, we identify stable monolayer structures of parent bulk photocatalysts via van der Waals corrected functionals. From there, we use HSE06 to identify structures with suitable band gaps for visible light absorption. Finally, we qualitatively evaluate the compounds' catalytic efficiency by computing the adsorption energy of reaction intermediates on each respective surface.


[1] A. Singh, J. Montoya, J. Gregoire, K. Persson; In prep.


S.B.T. was supported by the Dept. of Energy (DOE) Computational Science Graduate Fellowship. Computational work was supported by the Materials Project (Grant No. EDCBEE) Predictive Modeling Center through the U.S. (DOE), Office of Basic Energy Sciences, Materials Sciences and Engineering Division, under Contract No. DE-AC02-05-CH11231: Materials Project program KC23MP.

In this talk we will discuss the construction of a comprehensive electronic structure database including electronic, magnetic, and mechanical properties of single-layer two dimensional (2D) compounds. These 2D compounds are classified by their plane groups based on 2D symmetry operations and full band structures of these 2D materials are evaluated using PBE and HSE06 hybrid functional. These electronic structure data provides a fertile ground for the future discovery of 2D compounds for solar energy conversion, electronics, optoelectronics and other applications. We will discuss the general trends in the electronic structures of binary 2D compounds and the interplay between orbital hybridization and band edge energies. As a benchmark of using this database for 2D materials discovery and design, I will present the discovery of novel 2D materials with different functionalities. These functional compounds include light absorbers for photocatalysis, a family of 2D compounds exhibiting exotic mechanical behavior such as negative Poisson's ratio, and 2D materials that correlate magnetic and topological properties.

We have performed an extensive high-throughput screening of known inorganic materials, in order to identify those that could be exfoliated into novel two-dimensional monolayers and multilayers [1]. The screening protocol first identifies bulk materials that appear layered according to a simple and robust chemical definition of bonding, determining then for all of these the binding energies of the respective monolayers, and their electronic state (metallic vs insulating), magnetic configuration (ferro-,ferri- or antiferro-magnetic), and phonon dispersions (to evaluate mechanically stability). Such protocol identifies a portfolio of close to 2,000 inorganic materials that appear either easily or potentially exfoliable, to be investigated further for promising properties. First focus has been on the determination of the effective masses and mobilities (from the full solution of the Boltzmann transport equation) for electronic applications; of topological invariants; of superconductivity and charge-density waves; and of photocatalytic parameters for water splitting. Thanks to the use of the AiiDA (http://aiida.net) materials' informatics platform, all the high-throughput calculations can be performed and streamlined in fully searchable and reproducible ways, they are stored in a database with their full provenance tree of all parent and children calculations, and can be shared with the community at large in the form of raw or curated data via the Materials Cloud (http://www.materialscloud.org) dissemination portal.

[1] Nicolas Mounet, Marco Gibertini, Philippe Schwaller, Davide Campi, Andrius Merkys, Antimo Marrazzo, Thibault Sohier, Ivano Eligio Castelli, Andrea Cepellotti, Giovanni Pizzi and Nicola Marzari, Two-dimensional materials from high-throughput computational exfoliation of experimentally known compounds, Nature Nanotechnology 13, 246–252 (2018).

Two-dimensional materials and weakly bonded layered materials exhibit potentially advantageous properties as thin electronic materials, but research has been largely focused on only a couple of dozen types. We have utilized data mining approaches to elucidate over 1000 2D materials and several hundred 3D materials consisting of van der Waals bonded 1D subcomponents, or molecular wires. I will provide a guided tour of the spectrum of properties of these materials that are of interest for electronic applications. We find that hundreds of these 2D materials have the potential to exhibit observable piezoelectric effects, representing a new class of piezoelectrics. A further class of layered materials consists of naturally occurring vertical hetero structures, i.e., bulk crystals that consist of stacks of chemically dissimilar van der Waals bonded layers like a 2-D super lattice.

Chemical vapor deposition (CVD) is a powerful technique for synthesizing monolayer materials such as transition metal dichalcogenides. It has advantages over exfoliation techniques, including higher purity and the ability to control the chemistry of the products. However, controllable and reproducible synthesis of 2D materials using CVD is a challenge because of the complex growth process and its sensitivity to subtle changes in growth conditions, making it difficult to extend conclusions obtained in one CVD chamber to another. Here, we developed a multiscale model linking CVD control parameters to the morphology, size, and distribution of synthesized 2D materials, which then used for creating a database of growth condition-morphology relation. Its capabilities are experimentally validated via the systematic growth of MoS₂. In particular, we coupled the reactor-scale governing heat and mass transport equations with the mesoscale phase-field equations for the growth morphology considering the variation of edge energies with the precursor concentration within the growth chamber. The predicted spatial distributions of 2D islands are statistically analyzed, and experiments are then performed to validate the predicted island morphology and distributions. It is shown that the model can be employed to predict and control the morphology and characteristics of synthesized 2D materials.

Growing interest in the potential applications of two-dimensional (2D) materials has fueled advancement in the identification of new 2D systems with exotic properties. Increasingly, the bottleneck in this field is the synthesis of these materials. Although theoretical calculations have been able to predict a myriad of promising 2D materials, only a few dozen have been experimentally realized since the initial discovery of graphene. Here, we adapt the state-of-the-art positive and unlabeled (PU) machine learning framework to predict which theoretically proposed 2D materials have the highest likelihood of being successfully synthesized. Using elemental information and data from high-throughput density functional theory calculations, we apply the PU learning method to the MXene family of 2D transition metal carbides, carbonitrides and nitrides, and their layered precursor MAX phases, and identify 18 new MXene compounds that are highly promising candidates for synthesis. By considering both the MXenes and their precursors, we further propose 20 new synthesizable MAX phases that can be chemically exfoliated to produce new MXenes.

The stacking of individual layers of two-dimensional materials can be experimentally controlled with remarkable precision on the order of a tenth of a degree. The relative orientation of successive layers introduces variations in the electronic properties that can be controlled by the twist angle. We use simple theoretical models and accurate electronic structure calculations to predict that the electronic density in stacked 2D layers can vary in real space in a manner similar to the band-structure in momentum-space, creating moire super-lattices. A direct consequence of the patterns is the localization of electronic states. We demonstrate this effect in graphene, a semi-metal, and MoSe₂, a representative material of the transition metal dichalcogenide family of semiconductors.
This effect can be useful in the design of localized electronic modes with specific geometries for experimental or technological applications, including superconductivity, as has been recently reported for 2D twisted bilayer graphene.

It has become possible in recent years to fabricate and manipulate two-dimensional nanomaterials in the laboratory that are as thin as one to few atomic layers. The reduced dimensionality gives rise to unique physical and chemical properties that differ from those of traditional bulk materials, and intriguing physical properties have been found in these few-layer systems. Computational studies have played a central role in understanding and predicting these novel properties. In this talk, I will focus on a few representative systems, including graphene systems and monolayers of transition metal dichalcogenides that exhibit properties ranging from charge density waves [1] to the quantum spin-Hall effect [2]. The hybrid system of boron nitride and graphene (h-BNC) at low BN doping serves as an ideal platform for band-gap engineering and valleytronic applications. The calculations find a linear dependence of the band gap on the BN concentration at low doping, arising from an induced effective on-site energy difference at the two C sublattices as they are substituted by B and N dopants alternately. In addition, our calculations show that the Moiré patterns in van der Waals heterostructures will modify the local band gap, interlayer interaction, and structural parameters [3-5]. I will also discuss one-dimensional topological insulators manifested in graphene nanoribbons, in which localized spin states may exist at the end or near the junctions. A symmetry protected topological classification is formulated for any type of termination with -quantized Berry phase when summed over all occupied bands.
[1] P. Chen et al., Nature Communications 8, 516 (2017).
[2] P. Chen et al., Nature Communications 9, 2003 (2018).
[3] Q. Zhang et al., Nature Communications 7, 13843 (2016).
[4] C. Zhang et al., Science Advances 3, e16001459 (2017).
[5] Q. Zhang et al., ACS Nano 12, 9355 (2018).

MXenes, two dimensional (2D) nanosheets of transition metal carbides, have garnered significant interest due to their intriguing physicochemical properties associated with the unique 2 dimensionally confined chemical structures. These new classes of 2D materials have already shown promising electrochemical energy storage characteristics in both supercapacitor and Li-ion battery anode applications and endowed fascinating prospects for a variety of emerging applications in electronics, optics, energy conversion and storage¹. Recently, MXenes exhibited impressive optical properties e.g., tunable broadband absorption and intense surface plasmon excitations, underlining their potential for photonic and plasmonic devices^2,3. However, theoretical and experimental investigations in this context remain sparse and mainly limited to the most studied Ti₃C₂T_x MXene.
Herein, we demonstrate molybdenum carbide MXene (Mo₂CT_x) photodetectors for broadband visible light photodetection. The Mo₂CT_x MXene thin films deposited on paper substrates exhibit broad photoresponse in the range of 400-800 nm with high responsivity (9 A W^-1), detectivity (~5x10¹¹ Jones) and reliable photoswitching characteristics at a wavelength of 660 nm. Photodetectors made of four other MXenes have been also fabricated. We found that the photocurrent generation in Mo₂CT_x is principally controlled by surface plasmon-induced hot electrons. The existence and the distribution of a variety of surface plasmon modes over individual Mo₂CT_x nanosheets are visualized by the combination of scanning transmission electron microscopy (STEM) and ultra-high resolution electron energy loss spectroscopy (EELS). In addition, Mo₂CT_x thin film devices are shown to be resistant to environmental factors, illustrating their stable and durable photodetection operation. Moreover, the arrays of Mo₂CT_x thin film photodetectors are shown to be readily bendable under multiple and repeated deformations. The explicit ability to detect and excite individual surface plasmons, provide a viable platform for various MXene-based optoelectronic applications.

References.
1. B. Anasori, M. R. Lukatskaya, Y. Gogotsi, 2D metal carbides and nitrides (MXenes) for energy storage. Nat. Rev. Mater. 2, 16098 (2017).
2. 11. Y. Dong, S. Chertopalov, K. Maleski, B. Anasori, L. Hu, S. Bhattacharya, A. M. Rao, Y. Gogotsi, V. N. Mochalin, R. Podila, Saturable Absorption in 2D Ti₃C₂ MXene Thin Films for Passive Photonic Diodes. Adv. Mater. 30, 1705714 (2018).
12. J. K. El-Demellawi, S. Lopatin, J. Yin, O. F. Mohammed, H. N. Alshareef, Tunable Multipolar Surface Plasmons in 2D Ti₃C₂T_x MXene Flakes. ACS Nano 12, 8485-8493 (2018).

The band gap is of paramount importance to almost all of the electronic and optical properties of semiconducting materials. By controlling the size of the band gap and their electronic structure, we can control both the transport properties and the optical interactions of such materials. The ability to energetically control electrons in solid-state devices is pivotal in the fields of sensing, renewable energy [1], information processing and communications technology [2]. Here we present the rules of 2D band gap engineering in 2D heterostructures. Our insights offer the potential of engineering not just the band gap, but the electronic dispersion itself; making it far more versatile than strain engineering of band gaps.

The field of 2D semiconductors has been of growing research interest in recent years, but 2D heterostructures have only recently become experimentally viable [3]. We have performed a large scale first principles study of many transition metal dichcogenides and other 2D semiconductors using density functional theory. These 2D layered heterostructures demonstrate weak inter-layer interactions. Due to this, the band structures of individual layers of a heterostructure have a high fidelity to their isolated counterparts. We show that a layered heterostructure will, therefore, have a dispersion which consists of an overlay of its components' band structures. We determine that the electronic dispersions of 2D layered heterostructures can be tailored by layer composition, and demonstrate a wide range of potentially attainable tunable band structures.

1. Phys. Rev. B 92, 075439
2. Nat. Comm, 8, 15251
3. ACS Nano, 8, 9550

We present the analysis of electronic band structure of InSe and GaSe films, from the stoichiometric mono- to N-layer films which determines optical properties of these 2D materials [1,2]. This study is based on the ab initio DFT and related multi-orbital tight-binding model analysis of the electronic band structure and wave functions in N-layer InSe and GaSe films, and it is compared to the results of ARPES and luminescence spectroscopy of this material [3]. We show [1,2] that the band gap in InSe (and GaSe) strongly depend on the number of layers, with a strong (more than twice) reduction from the monolayer to crystals with N>6. We find that the conduction-band-edge electron mass in few-layer InSe is quite light (comparable to Si), which suggests opportunities for high-mobility devices and the development of nanocircuits. In contrast, the valence band in mono-, bi- and trilayer InSe is flat, opening possibilities for strongly correlated hole gases in p-doped films. Using the band structure and wave functions, we analyse optical transitions in thin films of InSe and GaSe, identify their polarisation and compare the results of modelling to the measurements performed on hBN-encapsulated atomically thin InSe and GaSe crystals. Then, we show that the same materials have potential for optical functionality in the far-infrared range, for which one can exploit the intersubband transitions in InSe and GaSe few-layer films [4].
[1] S. Magorrian , V. Zolyomi, V. Fal’ko V, Phys. Rev. B 94, 245431 (2016); S. Magorrian, V. Zolyomi, V. Fal'ko, Phys. Rev. B 96, 195428 (2017); V. Zolyomi, N. Drummond, V. Fal'ko, Phys Rev B 89, 205416 (2014); V. Zolyomi, N. Drummond, V. Fal'ko, Phys. Rev. B 87, 195403 (2013)
[2] D. Bandurin, Nature Nanotechnology 12, 223–227 (2017); G. Mudd, et al, Scientific Reports. 6, 39619 (2016); D. Terry et al, 2D Materials 5, 041009 (2018)
[3] Z. Ben Aziza, et al, Phys. Rev. B 98, 115405 (2018)
[4] S. Magorrian, et al, Phys. Rev. B 97, 165304 (2018)

The deliberate tailoring of semiconducting 2d transition metal chalcogenides for optoelectronic applications – via an understanding of their emergent electronic structure and photophysics in terms of their chemical composition and atomic arrangement – presents significant opportunities, due to their strong light-matter interactions, reduced dimensionality, and chemical diversity and heterogeneity. Here, I will describe the use first-principles computational methods – including density functional theory, ab initio many-body perturbation theory, and materials databases – to predict and understand the electronic structure, chemical bonding, and excited state phenomena in these systems, leading to new intuition about this important class of materials. I will review our recent calculations of point defects, and in their impact on electronic structure and valley polarization, in 2d transition metal dichalogenides; I will also discuss ongoing work in which we interface these systems with magnetic substrates to break valley degeneracy. Additionally, time permitting, I will discuss the opportunity for large nonlinear optical responses in monolayer transition metal monochalogenides; and for photocatalytic reduction of CO₂. In each case, I will emphasize implications for experiments and new principles for the design of functional layer materials. This work supported by the Department of Energy and the Department of Defense; computational resources provided by NERSC.

Excitonic effects dominate the optical properties of materials, especially of low-dimensional materials such as transition metal dichalcogenides, e.g. monolayer molybdenum disulfide [1]. When considering the excitonic effect, the band structure changes remarkably which can even result in a change of the material’s bandgap type (e.g. from indirect to direct) [1]. However, while primary bright excitonic transitions are well studied, a systematic approach to dark transitions is still lacking.
In this research, we study the excitonic effect in transition metal dichalcogenide monolayers starting with MoS₂ using density functional theory, quasiparticle methods (GW0) and the Bethe–Salpeter equation for calculating electronic excitations. In order to show the bandgap and exciton landscape, we include bright and momentum-resolved dark excitons in a novel approach of plotting the excitonic band structure. This way of visualizing the band structure allows us to easily tell if the material has a direct band gap after considering excitonic effects or not. Further, it enables the classification of dark excitons into spin-forbidden and/or momentum-forbidden transitions while clearly displaying the respective exciton binding energies. We believe that our approach will allow for exciting results, especially for the application to materials like ReSe₂ for which the nature of the band gap is still debated.

[1] E. Malic et al. (2018), Physical Review Materials, vol. 2, no. 1, DOI: 10.1103/PhysRevMaterials.2.014002

A direct band gap [1], remarkable light-matter coupling [2] as well as strong spin-orbit and Coulomb interaction establish two-dimensional (2D) crystals of transition metal dichalcogenides (TMDCs) as an emerging material class for fundamental studies as well as novel technological concepts. Valley selective optical excitation allows for optoelectronic applications based on the momentum of excitons [3]. In addition to lattice imperfections and disorder, scattering by phonons is a significant mechanism for valley depolarization and decoherence in TMDs at elevated temperatures preventing high-temperature valley polarization required for realistic applications [4]. Thus, a detailed knowledge about strength and nature of the interaction of excitons with phonons is vital. We directly access exciton-phonon coupling in charge tunable single layer MoS₂ devices by polarization resolved Raman spectroscopy. We observe a strong defect mediated coupling between the long-range oscillating electric field induced by the longitudinal optical (LO) phonon in the dipolar medium and the exciton. We find that this so-called Fröhlich exciton LO-phonon interaction [5] is suppressed by doping. This suppression correlates with a distinct increase of the degree of valley polarization of up to 60% even at elevated temperatures of 220 K [6]. Our result demonstrates a promising strategy to increase the degree of valley polarization towards room temperature valleytronic applications.
We acknowledge financial support by the DFG via excellence cluster “Nanosystems Initiative Munich” and DFG projects WU 637/4- 1.
[1] K. F. Mak et al., Phys. Rev. Lett. 105,136805 (2010).
[2] U. Wurstbauer et al., J. Phys. D. Appl. Phys. 50, 173001 (2017).
[3] K. F. Mak et al., Nat. Nanotechnol. 7, 494 (2012).
[4] H. Zeng et al., Nat. Nanotechnol. 7, 490 (2012).
[5] H. Fröhlich, Adv. Phys. 3, 325 (1954).
[6] B. Miller et al. submitted (2018).

Recent years have observed a plethora of strong dipole type polaritonic excitations in 2D materials owing to the reduced screening. These polaritons can be sustained as electromagnetic modes at the interface between a positive and negative permittivity material. In the case of the plasmon-polaritons (e.g. in semi-metallic graphene), the negative permittivity is provided by the coherent oscillations of the free carriers. For exciton-polaritons (e.g. in semiconducting transition metal dichalcogenides, TMD) and phonon-polaritons (e.g. in diatomic hexagonal boron nitride, hBN), it is associated with their resonant optical absorption, resulting from a highly dispersive permittivity. These optical resonances can also result in a negative permittivity, albeit over a narrow spectral window.

In this talk, I will discuss our recent efforts in understanding plasmons behavior in 2D materials, such as graphene and its bilayer, black phosphorus, and transition metal dichalcogenides, and using them to control the flow of light both in the far- and near-field. The general constitutive materials response of 2D materials, in conjunction with metasurface approaches, can potentially enable arbitrary control of phase, amplitude, polarization and directionality of light.

Development of 2D transition metal dichalcogenides (TMDs) for photonic applications advanced greatly in recent years, for example, in providing improvements in optical nonlinearities, such as two-photon absorption (TPA) cross-sections. Interestingly, recently, even larger TPA cross-sections were measured for 2D organic-inorganic perovskites (OIPs). The 2D OIP structure emerges when Goldschmidt’s tolerance factor, which depends primarily on the ionic radius of the organic cation in this case, is larger than 1. 2D OIPs exhibit a large increase in the exciton binding energy due to quantum and dielectric confinement, and provide spectral tunability. Here we report on 2D OIPs with organic chromophore incorporation, in comparison to previously investigated cations, aiming to design for intensity enhancements. Density functional theory calculations of 2D OIP structures will be described, and quantification of the geometrical distortion by the bond length quadratic elongation, bond angle variance, and perovskite layer distortion. Electronic structure calculations, including spin-orbit coupling, demonstrate variations in the band alignments of the organic and inorganic parts, leading to varying confinement of the electrons and holes. The interplay between hybridization and band alignment type, as well as the calculated absorption spectra, as dependent on the chromophore type, will be discussed. Finally, in the context of tunability in the optical properties of 2D photonic nanostructures, the optical response of a defective monolayer TMD (WSe₂) will be reported. Upon introduction of single or double Se vacancies, or a single W vacancy within monolayer WSe₂, the computational results demonstrate emergence of deeper defect excitons in comparison to shallower values observed for edges. Defect excitons are shown to red-shift significantly for large rotational defects that pattern the monolayer.

This talk will discuss the synthesis of large-area, high-quality monolayers of nitrogen-, silicon- and boron-doped graphene graphene sheets on Cu foils using ambient-pressure chemical vapor deposition (AP-CVD) [1-3]. Scanning tunneling microscopy (STM) and spectroscopy (STS) reveal that the defects in the doped graphene samples arrange in different geometrical configurations exhibiting different electronic and magnetic properties. Different substitutional nitrogen sites could be controlled by controlling the CVD temperature, flowrate and the dureation of the doping precursor insider the reactor. Interestingly, these doped layers could be used as efficient molecular sensors.
We will show that B-doped graphene (BG) exhibits unique sensing when detecting toxic gases such as NO₂ and NH₃ [3]. The detection limit for BG can reach as low as 95 ppt for NO₂ gas and 60 ppb for NH₃ gas. BG has enhanced sensitivity values of 27 and 105 times better than graphene for NO₂ and NH₃ detection [3]. This is attributed to the presence of B atoms within the graphene lattice that results in a high affinity to both donor and acceptor molecules, leading to stronger interactions between the molecules and graphene.
Finally we will discuss the use of graphene for enhancing Raman signals for specific molecules, also known as graphene enhanced Raman scattering (GERS). NG and Si-doped graphene (SiG) could significantly enhance the Raman signal of fluorescent molecules: Rhodamine B (RhB), crystal violet (CRV) and methylamine blue (MB) [1]. SiG exhibits an enhancement factor 10-40 times higher than that of PG (see Fig. 1) [1]. We will also show that by using NG as a substrate, Raman signals of dye molecules can be detected at a record low concentration of 10^-11 M [4]. This is very close to single molecular detection. However, this enhanced Raman sensing requires the Fermi energy of the substrate to match the LUMO level of the probe molecule, which allows an effective charge-transfer excitation [4].
References
[1] R. Lv, M. Terrones, et al. Adv. Mater., 2014, 26, 7593-7599.
[2] R. Lv, M. Terrones, et al. Sci. Rep., 2012, 2, 586.
[3] R. Lv, M. Terrones, et al. Proc. Natl. Acad. Sci., 2015, 112, 14527-14532
[4] S. M. Feng, M. Terrones, Sci. Adv., 2016, 2, e1600322.

The recent discovery of 2D magnets provides new ways to study 2D magnetism by harnessing the unique features of atomically-thin materials, such as electrical control for magnetoelectronics and van der Waals (vdW) engineering for novel interface phenomena. In this talk, I will describe our recent magneto-optical spectroscopy experiments on vdW magnets. I will discuss the layered antiferromagnetic properties of CrI3 with electrical control, giant tunneling magnetoresistance through spin filtering effect in vdW magnetic tunnel junctions, magnetic dynamics in zigzag antiferromagnets, and progress on exfoliable 2D magnets with stable magnetic order near room temperature.

Light matter interactions are strongly enhanced in two dimensional semiconductors, such as monolayer transition metal chalcogenides, as a result of extremely strong quantum confinement and many-body interactions. In these systems, optical transitions are dominated by excitons with giant binding energies, giving rise to resonantly enhanced absorption, reflection, emission as well as nonlinear multiphoton effects. Many applications would benefit from stacked multi-quantum-well-like structures of these materials, similar to the chemically synthesized 2D layered perovskite systems. However, most 2D systems still require laborious serial stacking of the layers to create such structures. In this work, we report strongly-bound excitonic optical properties of a bulk 2D-layered hybrid metal chalcogenolate, silver benzeneselenolate, also called mithrene, which is a three dimensional multi-quantum-well material synthesized via facile self-assembly [1]. In mithrene, silver atoms are bonded with selenium atoms to form a semiconducting monolayer lattice, where the layers are electronically isolated from each other by organic ligand barrier layers, resulting in a multilayered structure with minimal interlayer coupling. For 2D semiconductors, photoluminescence excitation (PLE) spectroscopy has been shown to be a powerful tool for revealing their exciton fine structures and quantifying exciton binding energies [2]. From PLE and two-photon PLE studies of mithrene, we identify the ground state exciton with a binding energy as large as ~0.4 eV and a series of exciton excited states including dark excitons. In addition, we perform theoretical calculations based on density functional theories, showing that mithrene is a direct gap semiconductor with the band edge carriers confined within the atomically thin silver selenide layers. It’s also shown experimentally that mithrene has highly efficient photoluminescence and ultrafast exciton dynamics which are promising for constructing optoelectronic devices. Furthermore, we demonstrate how structural modifications can be employed to engineer the optical response in various derived mithrene materials. The benzene molecules forming the quantum well barrier layers can be changed to biphenyl and methoxybenzene molecules, respectively, while the silver selenide semiconducting layers are kept unchanged. In this way, we found that the band edge excitons are minimally affected while the higher energy optical transitions can be tuned. As such, we anticipate the emergence of a whole family of easily synthesized mithrene-inspired layered excitonic materials in the form of multi quantum wells for both fundamental research and practical applications.

[1] B.Trang, J.N. Hohman, et al., J. Am. Chem. Soc., 140 (42), 13892 (2018)
[2] K.Yao, N.J.Borys, P.J.Schuck, et al., Phys. Rev. Lett., 119, 087401 (2017)

In this talk, we use theory and computation to investigate the decoherence properties of spin qubits in 2D materials and we propose strategies to search for ideal host materials for spin qubits.

Deep-level defects in wide-bandgap semiconductors exhibit localized electronic energy levels with paramagnetic spin states, and hence they constitute promising platforms for quantum bits (qubits). Defect-based solid-state spin qubits have been widely studied, including the nitrogen-vacancy (NV) center in diamond, vacancies in SiC, donor spins in silicon, and rare-earth ions, targeting quantum computing and sensing applications. The spin coherence time T₂, usually measured by Hahn-echo experiments, is a critical property for qubit applications. In quantum computers, in order to achieve quantum error correction, at least 10⁴ quantum gate operations need to be performed within T₂, yielding a desired coherence time larger than 100 μs. The sensitivity of quantum sensors, which is inversely proportional to T₂, benefits as well from qubits with long coherence times.

The decoherence of spin qubits can arise from different sources, including the magnetic noise from high-density paramagnetic defects and dangling bonds on the material surface. In highly purified bulk solids, the phase decoherence is mainly caused by the entanglement between the electron spin qubit and the nuclear spin bath of the host material. Thus, reducing the concentration of nuclear spins, e.g. through isotopic purification, is an effective technique to enhance T₂. For example, the ensemble T₂ of the NV center in diamond has been extended from 0.63 ms to 1.8 ms by reducing the concentration of ¹³C nuclear spins from 1.1% to 0.3%.

The nuclear spin density can also be reduced by decreasing the dimensionality of the nuclear spin bath from three dimension to two dimension or even lower dimensions. In addition, two-dimensional (2D) layered materials exhibit a variety of promising properties for the realization of hosts of defect-based spin qubits. 2D layered materials can be easily transferred and integrated into hybrid quantum systems. In 2D systems, point defects can be created and precisely located using a scanning transmission electron microscope or an annular dark field microscopy. Furthermore, compared with thin films of three-dimensional (3D) solids, 2D layers are free from unsaturated dangling bonds at the surface that are detrimental to spin coherence. However, the nuclear spin environments in many 2D materials are much more complex than, e.g. in diamond. Currently, h-BN and MoS₂ are among the most studied 2D material host candidates, and in h-BN all nuclei have large nuclear-spin moments.

In this talk, we investigate the coherence of spin qubits in 2D materials. We first explore the effect of dimensionality on spin decoherence and discuss the interplay of dimensionality, crystal geometry and nuclear spin concentration in determining T₂. We then predict the coherence time of spin qubits in 2D materials such as MoS₂ and h-BN, and show that isotopic purification can be much more effective in 2D than in 3D, leading to a spin coherence time of more than 30 ms in isotopically pure monolayer MoS₂. However achieving low nuclear spin concentration is not the only design rule to obtain optimal spin qubit hosts: the important roles of gyromagnetic ratios, homo-nuclear distance and nuclear magnetic moments are also discussed. Overall we predict that single-layer MoS₂ layers and delta-doped diamond slabs can be promising hosts for spin qubits, while the T₂ of h-BN appears to be much more challenging to engineer, due to the large gyromagnetic ratio of the boron nuclei.

This work is supported by NSF-MRSEC.

Since the demonstration of intrinsic ferromagnetism in atomically thin crystals, there has been a lot of interest in two-dimensional (2D) magnets. Among them, CrI3 presents an intriguing case. While bulk CrI3 is a ferromagnet, it becomes a layered antiferromagnet when thinned down to a few atomic layers. A number of interesting phenomena associated with this layered antiferromagnetism have been observed, including giant tunneling magnetoresistance and electrostatic doping control of magnetism. In this talk, I will present our study of the stacking dependent magnetism of bilayer CrI₃ using first-principles calculations. We show that the stacking order defines the magnetic ground state. By changing the interlayer stacking order one can tune the interlayer exchange interaction between antiferromagnetic and ferromagnetic. We propose that the predicted stacking-dependent magnetism can be probed via linear magnetoelectric effect. Our results not only gives a possible explanation for the observed antiferromagnetism in bilayer CrI₃ but also have direct implications in heterostructures made of two-dimensional magnets.

Magnetic order in two-dimensional (2D) materials is intimately coupled to magnetic anisotropy (MA) since the Mermin-Wagner theorem implies that rotational symmetry cannot be spontaneously broken at finite temperatures in 2D. Large MA thus comprises a key ingredient in the search for magnetic 2D materials that retains the magnetic order above room temperature. Magnetic interactions are typically modeled in terms of Heisenberg models and the temperature dependence on magnetic properties can be obtained with the Random Phase Approximation (RPA), which treats magnon interactions at the mean-field level. In the present work we show that large MA gives rise to strong magnon-magnon interactions that leads to a drastic failure of the RPA. We then demonstrate that classical Monte Carlo (MC) simulations correctly describe the critical temperatures in the large MA limit and agree with RPA when the MA becomes small. A fit of the MC results leads to a simple expression for the critical temperatures as a function of MA and exchange coupling constants, which significantly simplifies the theoretical search for new 2D magnetic materials with high critical temperatures. The expression is tested on a monolayer of CrI3, which were recently observed to exhibit ferromagnetic order below 45 K and we find excellent agreement with the experimental value.

We apply the developed methodology to perform a computational screening for magnetic 2D materials with large critical temperatures based on the computational 2D database C2DB. We identify several stable materials, which are predicted to exhibit magnetic order above room temperature.

Symmetry, interaction and topological effects greatly influence the quantum properties of atomically thin two-dimensional (2D) materials. Defects and environmental screening can play equally important roles in determining their behaviors. In this talk I present several interesting phenomena related to point defects and substrate screening on the properties of 2D materials. 1) We study, using ab initio GW-BSE theory, the effect of point-defect chalcogen vacancies on the optical properties of monolayer transition metal dichalcogenides (TMDs). The chalcogen vacancies introduce unoccupied in-gap states and occupied resonant quasiparticle defect states. These defect states give rise to strongly bound defect excitons and hybridize with excitons of the pristine system, reducing the valley-selective circular dichroism. The results suggest a pathway to tune spin-valley polarization and other optical properties through defect engineering. 2) A fascinating property of monolayer semiconductors is that the electronic bandgap ceases to be an intrinsic physical parameter and can be engineered through the surrounding dielectric environment. We have exploited such a dielectric-dependent electronic bandgap and demonstrated the formation of a lateral heterojunction within a homogeneous MoS₂ monolayer through substrate engineering. The capability to tailor the local band structure and electrical transport in 2D semiconductors through dielectric engineering can open exciting new ways to design 2D nanoelectronic and optoelectronic devices. 3) Plasmons in atomically thin 2D materials are distinct from those of the bulk and traditional 2D electron gas systems. We discover that the plasmon dispersion in real quasi-2D metals is qualitatively different, being virtually dispersionless for wavevectors of typical experimental interest. This stems from a lack of continuous translational symmetry, making dispersionless plasmons a universal phenomenon in quasi-2D metals. Moreover, our ab initio calculations show that plasmons on monolayer metallic TMDs are long lived and localizable in wave packets (within ~15 nm) which does not significantly disperse over practical measurement time. Their slow propagation opens the possibilities of tracking plasmon wave packets in real time and of localizing long-lived, boson-like excitations in extended, atomically thin materials.

This work was supported by the U. S. Department of Energy, National Science Foundation, and Office of Naval Research. I would like to acknowledge collaborations with members of the Louie group.

In this work we investigate the production of two-dimensional molybdenum disulfide, MoS₂, through two nonthermal based routes of synthesis: plasma enhanced atomic layer deposition (PEALD) and plasma enhanced chemical vapor deposition (PECVD). MoS₂ is a class of van der Waals two-dimensional materials whose applications include transistors, memristors, hydrogen evolution reactions, photodetectors, lithium ion batteries, water purification, and dye sensitized solar cells. However, many production methods for two-dimensional molybdenum disulfide suffer from either high temperature, long time scales, dangerous chemicals (hydrogen sulfide) or low yield. Here, we describe two processes capable of producing large area films of MoS₂ utilizing molybdenum pentachloride and elemental sulfur as precursors. Both
PECVD and PEALD can produce polycrystalline material at low temperatures (275-315 ^oC). In the case of PECVD grown films, time scales for nucleation and large area growth occurs on the order of one to two minutes to produce few layer molybdenum disulfide. For PEALD, due to the steric factor of the molybdenum pentachloride precursor, only 2 cycles are required for a complete polycrystalline two-layer MoS₂ film with Angstrom level surface roughness. Film quality and structure are confirmed by Raman spectroscopy, transmission electron microscopy, and atomic force microscopy.
This work was supported primarily by the U.S. National Science Foundation through the University of Minnesota MRSEC under Award Number DMR-1420013. Parts of this work were carried out in the Characterization Facility, University of Minnesota, which receives partial support from NSF through the MRSEC program.

MXenes have attracted significant attention since their discovery because their elegant synthesis yields nearly 100% monolayer samples. MXenes are prepared from a layered MAX (metal aluminum carbide) phase by selectively etching layers of aluminum (or Si, Ga) atoms using HF, leaving the 2D metal carbide lattice intact. Building on the synthetic concepts of MXenes, here we report the first synthesis of 2D materials made by selectively etching halides from layered phases. First, we synthesized layered metal carbide iodides by high temperature reactions of the metal, carbon, and the metal iodide. By tuning the temperature, pressure, and stoichiometry, we produced powders consisting of thin platy crystals about 1 µm thick and 1-10 µm. Then we explored routes to etch the iodide layers selectively, including solution-based methods using lithium naphthalide as the etchant and a gas-phase method using potassium vapor at 200 °C. Like the oxidative etchings used to produce MXenes, we find that these reductive etchings produce materials with hollow, accordion-like morphologies, which can be delaminated further by sonication to few-layer 2D materials. We performed X-ray diffraction to reveal that the reaction of layered metal carbide iodides with potassium vapor extracts iodide and forms potassium iodide, which can be dissolved and removed. Interestingly, unlike traditional MXenes, the surface of the reductively etched metal carbides do not have hydroxyl or fluoride ions bound to the surface. In fact, our DFT calculations show that the reductively etched 2D materials, under appropriate conditions, can remain unfunctionalized at the surface, offering unique properties that could be useful in a number of applications in energy storage, catalysis, and electronics. Motivated by our initial promising findings, further experimental characterization of the 2D materials and their surfaces are underway.

Recent results have identified two-dimensional (2D) tellurene as a potential van der Waals (vdW) material for thermoelectric and optoelectronics applications owing to their pseudo-one-dimensional (anisotropic) behavior and structure. While hydrothermal synthesis is suitable for yielding potentially crystalline materials, it is well known for its incompatibility with manufacturing and scaling. This talk will introduce novel synthesis of tellurene sheets onto a variety of surfaces of vdW-bonded 2D layered crystals. Our systematic tellurene growth studies on six different vdW surfaces show that GaS and GaSe surfaces enable highly crystalline and self-oriented tellurene sheets. Detailed cross-sectional transmission electron microscopy (TEM), angle-resolved Raman spectroscopy, scanning electron microscopy, and energy-dispersive spectroscopy measurements provide fundamental insights into the growth dynamics and characteristics of tellurene. First-principles calculations and cross-sectional TEM measurements suggest that tellurene chains are well-oriented along the GaSe armchair lattice direction owing to the much-reduced total energy of the system and a stronger degree of coupling across adjacent layers. Synthesized tellurene sheets exhibit remarkable structural anisotropy, as evidenced by angle-resolved Raman measurements, and the overall results open a clear pathway for layer-by-layer growth of tellurene sheets with controlled crystalline anisotropy using conventional synthesis techniques.

Electrides have gained considerable attention in recent literature thanks to demonstrated applications as superconductors, catalysts, and solid-state dopants. However, current electride materials are largely limited by the tradeoff between desired electronic properties and the preventive reactivity of these materials. This is caused by the distinctive feature of electrides: bare-electrons that exist at isolated lattice sites. Further, reactivity increases when bare electrons form interconnected pathways, but these 'multidimensional' electrides also provide desirable properties like heightened electrical conductivity. Thus, this tradeoff must be addressed by discovery of electrides specialized for given applications (i.e. where stability is vital) and/or stabilization of highly desired multidimensional electrides. Both strategies are employed here. High-throughput screening is used to address the scarcity of stable multidimensional electrides and propose a new series of robust, low-work function compounds. This screening is driven by derivatives of known halide materials, where electrochemical dehalogenation can produce electride forms, and the synthetic approach is experimentally demonstrated with previously reported two-dimensional (layered) electrides. By avoiding high-temperature synthesis, kinetically stable and oxidation-resistant electrides can be realized. Secondly, stabilization of known and newly reported electrides are computationally examined and experimentally applied via novel charge-transfer stabilization at the electride-substrate interface. By reducing the reactivity of electrides, previously metastable forms can be synthesized as electride thin-films, where their increased surface area is idealistic for the aforementioned applications and further enables new uses. Together, the discovery of multidimensional electrides and charge-transfer stabilization expands on reported multidimensional electrides and their corresponding applications as low work-function materials.

In recent years, researchers have tried to develop various methods to synthesize high-quality large-scale graphene using chemical vapor deposition (CVD) processes. However, the performance of CVD graphene is often degraded by defects, cracks and polymer residues that occur during transfer processes, which causes the similar problems in the transfer of other 2D materials. The PMMA coating is still a very effective method for the transfer of 2D materials, but the wet process to remove PMMA using acetone is problematic for device integration with soluble layers. Here we report a novel transfer method based on the use of pressure sensitive adhesive (PSA) that can be simply spin-coated on the sample, and completely removed by peeling-off after transfer, which is expected to allow the use of soluble layers for various 2D device fabrication processes. We confirmed that the sheet resistance of graphene has not been altered after transfer thanks to the good adhesion between graphene and PSA layers. In addition, the transparent graphene FETs transferred by PSA shows high charge carrier mobility compared to the FETs transferred by PMMA or thermal release tapes (TRT), implying that the graphene surface is ultra-clean without residues after the removal of PSA. We believe that the current method is applicable to various flexible and wearable applications that have been limited by wet transfer conditions for more practical and larger scale device fabrications of 2D materials.

Two-dimensional (2D) materials can be single layer because the bonds between each layer are made up of Van der Waals bonds while interlayer atoms bind together as covalent bonds. They can be separated into atomic-level single layers, which are very flexible and transparent, and are suitable for future electronics because of their various electrical and physical properties.
Among 2D materials, MoS₂ has drawn much attraction and chemical vapor deposition (CVD) is most popular method for synthesizing large size 2D materials. However, this process is difficult to control the temperature accurately and causes problems of the non-uniformity depending on the distance (sulfur position), and the internal contamination of furnace.
In this study, a novel chemical vapor deposition (CVD)-free and solution-processed synthetic method were developed and successfully synthesized large size and uniform MoS₂ thin films. The MoS₂ precursor solution was formulated by sulfur-dissolving method to obtain uniform coating property. MoS₂ thin film was prepared by simple spin-coating and one-step annealing method without any CVD process (CVD-free). The solution-synthesized MoS₂ thin films were characterized to confirm 2H-MoS₂ structure. The various atomic layers of synthesized MoS₂ could be controlled with the precursor concentrations, such as two-layer for 0.0070 M, three-layer for 0.0125 M, and five-layer for 0.025 M of MoS₂ precursor solution, which were confirmed by STEM.

A phosphorus isomer named green phosphorus was recently predicted by theory with a similar interlayer interaction compared to that of black phosphorus, which indicates that individual layers can be mechanically exfoliated to form two-dimensional (2D) layers called green phosphorene. This 2D monolayer showed high stability and was predicted to have a direct band gap up to 2.4 eV. First-principles density functional theory calculations were used to investigate the mechanical properties and strain effect on electronic band structure of the 2D green phosphorene along two perpendicular in-plane directions, namely armchair and zigzag. It was found that the 2D nano-sheet can sustain a tensile strain in the armchair direction up to 35% which is larger than that of graphene and black phosphorene, suggesting that green phosphorene has more puckered structure compared to black phosphorene. The results also showed that the Young’s modulus and Poisson’s ratio in the zigzag direction are four times larger than those in the armchair direction, confirming the anisotropy of the material. Furthermore, uniaxial strain showed an ability to trigger the direct-indirect bandgap transition in the material, and the critical strains for the bandgap transition are revealed.
Furthermore, one-dimensional (1D) green phosphorene nanoribbons obtained from its 2D monolayer were also studied along both armchair and zigzag directions with ribbon widths up to 57 Å. Effects of ribbon width (size) and strain on the mechanical and electronic properties of the ribbons were studied. The Young’s modulus, effect of quantum confinement on the band gap, and effect of strain on the electronic band structures of the ribbons were investigated. The green phosphorene ribbons were found to exhibit prominent anisotropic properties, with the Young’s modulus in the range of 10-35 GPa for the armchair green phosphorene nanoribbons (AGPNR) and 160-170 GPa for the zigzag green phosphorene nanoribbons (ZGPNR), which are the same order of magnitude as those of its 2D sheet. The work function was found to be between 5 eV ~ 5.7 eV for the range of ribbon widths studied. Both size and strain trigger direct-indirect band gap transitions in the ribbons and their transition mechanisms were discussed.

MXene materials, 2D transitional metal carbides, have shown some promising properties in battery, supercapacitor and energy applications. These 2D layers have been exfoliated from ternary carbides named MAX phases. However, recently , a new group of 2D materials called MBene have been in interest since they have a similar structure and exfoliation mechanism to MXene. In this study, we discuss the possibility of forming 2D Nickel boride layers via selectively etching Ni₂ZnB MAB phase. Single phase Ni₂ZnB MAB phase has been synthesized for the first time. The zinc interlayer was then etched out, via immersing the MAB phase in diluted HCl. The selectively etched particles were investigated via scanning electron microscopy and powder X-ray diffraction. It was observed for the first time that ternary Ni₂ZnB phase can be exfoliated via selectively etching the zinc interlayer. Demonstrating the zinc interlayer etching opens a door to the synthesis of a large number of new 2D borides from ternary MAB phases, especially those containing Zn.

We demonstrate that intercalation of zero-valent atoms can be used to chemically change the color of molybdenum trioxide (MoO3) from transparent white to dark blue. Chemochromism is achieved through dilute creation of a zero-valent atomic species in solution (such as V, Mo, Sn, and Co) which is then intercalated into the van der Waals gap. Chemochromism of MoO3. is shown to be reversible, robust, and has remarkably high absorptivity. We show that this 2D layered oxide shows promise in color-changing smart devices.

Transition metal dichalcogenides (TMDs) are rapidly becoming of interest when transitioning from bulk phase to two dimensional structures. Not only do they boast high carrier mobilities, but first-principles calculations and experimental results have demonstrated that methods like doping and alloying can tailor critical properties such as the band gap. Now, the focus on first-principles calculations on TMDs has shifted from predicting performance and properties, to understanding and controlling TMD growth and processing. Our experimental collaborators have successfully growth MoS₂ via atomic layer deposition (ALD) but have little insight into how the hydroxyl concentration on the substrate affects growth kinetics. We use density functional theory (DFT) to model the initial growth mechanics between the gas precursors and substrate and predict the optical hydroxyl surface concentration for ALD growth. By thermally controlling hydroxyl groups on our alumina surface we characterize our substrates with x-ray photoelectron spectroscopy (XPS) to determine experimental hydroxyl concentration. We propose a novel approach of combining DFT calculations and ALD growth measurements for finding optimal hydroxyl concentration on alumina substrates.

CO₂ emissions have increased substantially in recent years, to a point that is detrimental to the environment and needs to be addressed. Methods to capture CO₂ and convert the greenhouse gas into its fundamental components are underway but are currently energy-intensive. We aim to approach this existing problem with novel materials, such as doped-2D layered materials and metal-organic frameworks on metal supports, and understand the interfacial interactions using microscopy, infrared and photoemission spectroscopy at near-ambient pressures.

Introduction: MoS₂ nanotube (MoS₂NT) that composed of semiconducting layered material has potential applications to the sensors, transistors, catalysts, etc. due to the unique 1D geometry and the edge state. Several methods have been used to synthesize MoS₂NT such as physical vapor transport, thermal decomposition of the precursor and hydrothermal synthesis. However, there were some disadvantages in these methods, for example, long time reaction and necessity of toxic gas (H₂S). Therefore, it is important to establish the economical and environmentally friendly way for the synthesis. We found that MoS₂NT can be synthesized by chemical vapor deposition (CVD) using SiO₂/Si substrate dispersed with FeO nanoparticles^(1,2). It can be considered that FeO nanoparticles work as catalysts as seen in the growth of carbon NT. The characterization and the growth mechanism of the MoS₂NT are focused in this study.
Method: MoS₂NTs were synthesized using a home-built CVD apparatus⁽³⁾ with two-flow system. Several kinds of metal or metal oxide nanoparticles (Fe, FeO, Fe₂O₃, CuO, Cu₂O, NiO) were used to investigate their catalytic activity. SiO₂/Si wafers dispersed with the nanoparticles were used as substrates. The temperatures and flow rates of the source materials (MoO₃ and sulfur) were separately optimized for the growth of the MoS₂NTs. The MoS₂NTs were characterized by transmission electron microscopy, energy dispersive spectrometry and Raman spectrometry.
Results and Discussion: Among the prepared nanoparticles, only FeO is active for the synthesis of MoS₂NT. Nothing or leaf-shape MoS₂ formed when the other nanoparticles were used. This is probably because FeO nanoparticles are reactive due to the O vacancy (non-stoichiometry) and Fe-Mo-S system forms layered material. We also found that the shape of the FeO nanoparticles affect the yield of the MoS₂NTs. The spherical and truncated octahedral FeO nanoparticles can be synthesized by optimizing the synthetic conditions⁽⁴⁾. It was found that the truncated octahedral FeO nanoparticles have greater catalytic activity than the spherical one. This is probably because the MoO₃ and sulfur vapor easily deposit on the sharp edges. Furthermore, the size of FeO nanoparticles also affects the growth of MoS₂NT. SEM images clearly revealed that the 30-nm truncated octahedral FeO nanoparticles show the greatest catalytic activity among the prepared FeO. The NTs on the SiO₂/Si substrate were identified as MoS₂ by Raman spectrometry because the two distinctive peaks, A_1g and E₂, were detected. EDS spectra that showed the existence of Mo and S is another evidence for the formation of MoS₂. The growth mechanism was concluded as VS mechanism by TEM observation because there was no FeO nanoparticles on the tip of the MoS₂NT. The TEM observation also clarified that the MoS₂NT has multilayer structure. The chirality of MoS₂NT was armchair that was determined by electron diffraction pattern.
References
(1) M. Weng, T. Yanase et al. CrystEngComm 19, 3915 (2017).
(2) M. Weng, T. Yanase et al. Jpn. J. Appl. Phys. 57 030304 (2018).
(3) T. Yanase et al. Cryst. Growth Des. 16 4467 (2016).
(4) Y. Hou et al. Angew. Chem. Int. Ed. 119, 6445 (2007).

Transition metal carbides (MXenes) have emerged as a promising family of 2D layered materials, and are considered as potential candidates for a myriad of applications including plasmonic sensing due to their outstanding electronic, optical, mechanical, and thermal properties. Herein, we report a flexible and reusable substrate made from a hybrid material consisting of 2D titanium carbide MXene (Ti2C) and silver nanoparticles, fabricated by spray coating for highly sensitive surface-enhanced Raman scattering (SERS). The 2D Ti2C layers were delaminated from their bulk Ti2AlC MAX phase by using a mixture of hydrochloric acid and potassium fluoride as the etchant solution and silver nanoparticles were synthesized by thermal reduction of silver trifluoroacetate in isoamyl ether in the presence of oleic acid. The as-prepared hybrid materials were characterized by UV-visible spectroscopy, Raman spectroscopy, X-ray photoelectron spectroscopy, atomic force microscopy, scanning electron microscopy, and X-ray diffractometry. The plasmonic properties of 2D Ti2C/silver nanoparticle hybrid material was evaluated with several organic dyes, which showed significantly different enhancement factors for different dyes, indicating a possibility of producing dye-selective SERS substrate. The unequal SERS enhancement factor and the mechanism behind MXene-based SERS detection will also be discussed.

In this talk, I present recent theoretical discoveries of several fascinating manifestations of topological effects in the electronic and optical properties of atomically thin one-dimensional (1D) and two-dimensional (2D) materials. First, we find that symmetry-protected topological phases exist in graphene nanoribbons (GNRs) [1]. Semiconducting GNRs of different width, edge shape, and terminating unit cells can belong to different electronic topological classes, characterized by a Z₂ invariant. Junctions between segments of topologically distinct GNRs are predicted to support robust in-gap topological junction states which can be used for band engineering. Experimental realizations of these predictions have been achieved [2]. Second, we show that the conventional optical selection rules for excitons must be replaced in 2D by a novel simpler formula, owing to a topological characteristic inherent to the photoexcitation of excitons in 2D [3]. The new selection rule is dictated by a winding number of the interband optical transition matrix elements (a heretofore unrecognized topological invariant). This appealingly simple and general new rule is applied to elucidate the optical spectra of gapped graphene systems [4]. Finally, if time permits, I will discuss some very recent work on the effects of electron-hole interactions on shift currents in non-centrosymmetric 2D crystals (the so-called bulk photovoltaic effect), in which we show excitonic effects lead to an enormous enhancement and, more interestingly, gives rise to DC conduction with sub-bandgap-frequency excitations [5].
This work was supported by the U. S. Department of Energy, National Science Foundation, and Office of Naval Research. I would like to acknowledge collaborations with members of the Louie group.
References:
[1] T. Cao, F.-Z. Zhao, and S. G. Louie, Phys. Rev. Lett. 119, 076401 (2017).
[2] D. J. Rizzo, G. Veber, T. Cao, C. Bronner, T. Chen, F.-Z. Zhao, H. Rodriguez, S. G. Louie, M. F. Crommie, F. R. Fischer, Nature 560, 204 (2018)
[3] T. Cao, M. Wu, and S. G. Louie, Phys. Rev. Lett. 120, 087402 (2018).
[4] L. Ju, L. Wang, T. Cao, T. Taniguchi, K. Watanabe, S. G. Louie, F. Rana, J.-W. Park, J. Hone, F. Wang, and P. L. McEuen, Science 358, 907 (2017).
[5] Y.-H. Chan, D. Y. Qiu, F. H. da Jornada, and S. G. Louie, submitted (2018).

JARVIS (Joint Automated Repository for Various Integrated Simulations) is a NIST-developed, unique integrated framework to accelerate material design using classical force-fields (FF), density functional theory (DFT) and machine learning (ML). The JARVIS-DFT hosts data for more than 30000 bulk and 800 monolayer materials. We identified more than 1500 2D materials using lattice parameter criteria and carried out exfoliation energy calculations. We charted improved lattice parameters and formation energies using van der Waals functional for all the 30000 materials in the database and also calculated elastic tensors for more than 12000 bulk and 250 monolayer materials. Using the above data, we established relations between exfoliation energies and elastic constants. To alleviate bandgap underestimation in conventional DFT and to improve frequency dependent dielectric function predictions, we re-evaluated band gaps and dielectric functions using meta-GGA based approaches for more than 10000 materials. We used a spectroscopic limited maximum efficiency approach to identify potential photovoltaic materials. Using spin-orbit spillage criteria, we identified more than 1500 potential topological materials including topological insulators, Weyl and Dirac semimetals, and topological crystalline insulators. The database is publicly available at https://jarvis.nist.gov/.

Nonlinear optical responses provide basis for ultrafast probing of material's intrinsic symmetry. Here we present first-principles theory of quantum nonlinear ferroic optical Hall effect (QNFOHE), a Hall like photocurrent originated from the second order current response in (multi)-ferroics. The interplay of crystalline, permutation, gauge, time reversal symmetries and inherent causality governs the symmetry of QNFOHE. We elucidate QNFOHE in a class of 2D multiferroics using first-principles calculations and group theoretical analysis. Our results suggest QNFOHE-based optical technique as a route for ultrafast characterization of multiferroic orders and domain evolution in multiferroic materials. These microscopic understandings of QNFOHE from first-principles theory, together with very recent discoveries of 2D ferroics/multiferroics, will open up a variety of new avenues for nonlinear optoelectronics. This research was supported by the National Science Foundation under award number NSF DMR-1753054. References: [1] H Wang and X Qian, Nano Letters 17, 5027-5034 (2017). [2] H Wang and X Qian. 2D Materials 4, 015042 (2017). [3] H Wang and X Qian. Quantum Nonlinear Ferroic Optical Hall Effect, submitted (2018).

Two-dimensional (2D) topological insulators (TIs) have been recognized as a new class of quantum state of matter. They are distinguished from normal 2D insulators with their nontrivial band-structure topology identified by the Z2 number as protected by time-reversal symmetry (TRS). 2D TIs have intriguing spin-velocity locked conducting edge states and insulating properties in the bulk. In the edge states, the electrons with opposite spins propagate in opposite directions and the backscattering is fully prohibited when the TRS is conserved. This leads to a quantized dissipationless “two-lane highway” for charge and spin transportation and promises potential applications. Up to now, only very few 2D systems have been discovered to possess this property. The lack of suitable material obstructs further study and application. In this talk, I will introduce by using first-principles calculations we have proposed that the layer compounds ZrTe5 and HfTe5, the transition metal dichalcogenide (TMD) haeckelites with square-octagonal lattice, the functionalized MXenes with oxygen M2CO2 (M=W, Mo, and Cr), are 2D TIs with quite large band gap. Importantly, ZrTe5 and HfTe5 are very easily to be synthesized by exfoliating. This will pave the way to tremendous applications of 2D TIs, such as “ideal” conducting wire and the realization of topological superconductivity and Majorana modes for quantum computing.

References:
[1] Hongming Weng,^† Xi Dai and Zhong Fang, MRS Bulletin 39, 849 (2014)
[2] Hongming Weng, Xi Dai and Zhong Fang, Phys. Rev. X 4, 011002 (2014)
[3] S. M. Nie, Zhida Song, Hongming Weng,^† and Zhong Fang,^* Phys. Rev. B 91, 235434 (2015)
[4] Hongming Weng et al., Phys. Rev. B 92, 075436 (2015)
[5] Y. Liang et al., Phys. Rev. B 96, 195414 (2017)

There has been tremendous interest in recent years to study two-dimensional (2D) atomic layers which form building blocks of many bulk layered materials. This was started by the spectacular discovery of graphene. This talk will focus on the materials science of the emerging field of 2D atomic layers of various compositions. Several aspects that include synthesis, characterization of defects, doping and manipulation will be explored with the objective of achieving functional structures with atomic layer building blocks. The concept of nanoscale engineering and the goal of creating new artificially stacked van der Waals solids and 3D constructs will be discussed through a number of examples including graphene and other 2D layer compositions. The talk will explore the emerging landscape of 2D materials systems that include hybrid compositions and multi-component 2D alloys. Some of the anticipated applications of these materials will also be discussed.

Search for a new two dimensional (2D) systems become very intense in the last years and was initiated by exfoliation of graphene and synthesis of silicone and germanene. Recently a new group of elemental 2D materials has been discovered in 15th group of periodic table and include phosphorene, α- and β-bismuthene and recently synthetized β-antimonene. It was shown that both α- and β-bismuthene support topologically protected edge states while β-antimonene is topologically trivial. It is therefore highly interesting to fabricate heterostructures basing on combination of these two elements.
Crystallographic structure of β-bismuthene and β-antimonene is graphene-like i.e. hexagonal but with every second atom lifted out of surface plane. In contrast α phase is black phosphorus-like i.e. characterized by two paired layers with rectangular symmetry.
Our recent LEEM/PEEM experiments allowed us for the first time to observe growth of α-bismuthene and its transformation to β phase. Surprisingly, we discovered that growth of Bi islands is characterized by their anomalous diffusion with islands jump length distribution described by truncated Levy statistics. Moreover, for the first time we were able to investigate electronic structure of single 2 µm wide α-bismuthene island using µARPES. These experiments allowed us to understand both crystallographic and electronic structure of α- and β-bismuthene. In consequence we decided to use α-bismuthene islands as a support for antimony films in order to engineer β-antimonene on α-bismuthene heterostructure. Surprisingly we fabricated two Van der Waals heterostructure systems: α- and β-antimonene grown on top of α-bismuthene deposited on MoS₂ [1]. It is worth pointing out here that α-antimonene was never synthetized before. Crystallographic structure of both Van der Waals heterostructres is investigated using STM and is further confirmed by moiré pattern simulations. Their electronic structure is probed experimentally using STS and theoretically using DFT. In particular DFT results suggest that α-Sb is 2D topological insulator while β-Sb is topologically trivial semiconductor. We note that the α-phases of Bi and Sb have a black phosphorus-like structure which has not previously been employed in van der Waals heterostructures, and in which the absence of surface dangling bonds means the structure is resistant to oxidation. Together with the semiconducting nature of the MoS₂ substrate, and potential to fabricate similar structures on a range of other substrates (especially those with rectangular symmetry), many new possibilities open up for the engineering of spintronic / topological devices.
This work is supported by the Polish National Science Centre under project DEC-2015/17/B/ST3/02362.

[1] Maerkl, T.; Kowalczyk, P. J.; Ster, M. L.; Mahajan, I. V.; Pirie, H.; Ahmed, Z.; Bian, G.; Wang, X.; Chiang, T.-C.; Brown, S. A.; 2D Metrials 5, 011002 (2017)

Quasibinary alloying among pairs of 2-dimensional (2D) transition metal dichalcogenides (TMDCs) is an attractive method for tuning properties for applications such as optoelectronics and catalysis. Of the many possible combinations of TMDCs, the small subset of semiconducting alloys has garnered widespread attention. Outside this limited subset, the synthesizability of alloys remains largely unknown. In order to the guide synthesis of such alloys, we present ab initio calculations of equilibrium phase diagrams with regions of stability for 27 TMDC alloys: M(1-x)M'xX2 and MX(2-2x)X'(2x) (M,M'= V, Nb, Ta, Mo, or W; X,X'= S or Se) and two heterostructural alloys, Mo(1-x)WxTe2 and WS(2-2x)Te(2x). We predict four new alloys that are miscible at all temperatures: Nb(1-x)TaxS2, Nb(1-x)TaxSe2, VS(2-2x) Se2x, and TaS2(1-x)Se2x. For the rest of the alloys, we present a qualitative analysis of synthesizability based on miscibility temperatures. We relate TMDC size mismatch with a mixing asymmetry that indicates the synthesizability of low concentration alloys. Our results open new compositional spaces that can be used to design and synthesize TMDC alloys for various applications. Acknowledgments: This work was funded by NSF DMREF-1729787.

The interplay of magnetism and topology is a key research subject in condensed matter physics and material science, which offers great opportunities to explore emerging new physics, like the quantum anomalous Hall (QAH) effect, axion electrodynamics and Majorana fermions. However, these exotic physical effects have rarely been realized in experiment, due to the lacking of suitable working materials. In this talk, we will review recent research progresses of magnetic and topological materials. In particular, we will present our recent works of finding intrinsic magnetic topological insulators in van der Waals layered MnBi2Te4-family materials [1,2]. The materials intrinsically show two-dimensional (2D) ferromagnetism in the single layer and three-dimensional (3D) A-type antiferromagnetism in the bulk, which could serve as a next-generation material platform for the state-of-art research. Remarkably, we predict extremely rich topological quantum effects with outstanding features in an experimentally available material MnBi2Te4, including a 3D antiferromagnetic topological insulator with the long-sought topological axion states, the type-II magnetic Weyl semimetal (WSM) with simply one pair of Weyl points, and the high-temperature intrinsic QAH effect. These striking predictions, if proved experimentally, could profoundly transform future research and technology of topological quantum physics.
Reference:
[1] J. Li, et al., arXiv:1808.08608.
[2] Y. Gong, et al., arXiv:1809.07926.

2D materials have unique structural mechanics [Nano Letters 15 (2015) 1302] that makes their defects and phase transformations dramatically different from those of 3D materials. I will focus on light-structure and electron-structure interactions in this talk, with a range of behaviors such as ferroelasticity [Nature Comm 7 (2016) 10843], ferroelectricity and electronic topology change [Science 346 (2014) 1344] in one to few-layer transition metal dichalcogenides. Design of ultrafast responsive 2D materials is emphasized.

Structure prediction complemented by density functional theory (DFT) calculation indicates that TeSe₂ is the most stable among the various Te_(1-x)Se_x compounds. Different from the case of bulk Te, the material can equally adopt three different crystal structures: H_γT, M_H, and M_βα phases, which are similar to 1T-TiSe₂, trigonal Te, and orthorhombic Te, respectively. These phases can be transformed from one to another by uniaxial tensile and shear stresses; they can be even transformed to their chiral mirror images. Band structure calculations including spin-orbit coupling (SOC) show that all three phases are semiconductors. The band gap (= 0.43 eV) increases with density, being the largest (= 1.86 eV) in the M_βα phase with the highest density. The H_γT phase exhibits hidden spin texture because of centrosymmetry. The other two phases display chiral spin texture due to the lack of the symmetry, in that two spin components of frontier bands can split by more than 100 meV in opposite directions. In addition, the layered TeSe₂ possess ferroelectricity. It can be used for optoelectronic applications.

Ion and electron irradiation suits particularly well for engineering the properties of nanostrucutres, see Refs. [1-2] for an overview. In particular, focused ion beams can be used for patterning two-dimensional (2D) materials, but the optimization of irradiation parameters requires full microscopic understanding of defect production mechanisms. In contrast to freestanding 2D systems, the details of damage creation in supported 2D materials are not fully understood, whereas the majority of experiments have been carried out for 2D targets deposited on substrates. Here, we suggest a universal and computationally efficient scheme to model the irradiation of supported 2D materials, which combines analytical potential molecular dynamics with Monte Carlo simulations and makes it possible to independently assess the contributions to the damage from backscattered ions and atoms sputtered from the substrate. Using the scheme, we study [3] the defect production in graphene and MoS2 sheets, which are the two most important and wide-spread 2D materials, deposited on a SiO2 substrate. For helium and neon ions with a wide range of initial ion energies including those used in a commercial helium ion microscope (HIM), we demonstrate that depending on the ion energy and mass, the defect production in 2D systems can be dominated by backscattered ions and sputtered substrate atoms rather than by the direct ion impacts and that the amount of damage in 2D materials heavily depends on whether a substrate is present or not. We also study the factors which limit the spatial resolution of the patterning process. Our results, which agree well with the available experimental data, provide not only insights into defect production but also quantitative information, which can be used for the minimization of damage during imaging in HIM or optimization of the patterning process.
1. A. V. Krasheninnikov, and F. Banhart, Nature Materials, 6 (2007) 723.
2. A. V. Krasheninnikov, and K. Nordlund, J. Appl. Phys. (Applied Physics Reviews), 107 (2010) 071301.
3. S. Kretschmer, M. Maslov, S. Ghaderzadeh, M. Ghorbani-Asl, G. Hlawacek, and A.V. Krasheninnikov, ACS Applied Materials & Interfaces 10 (2018) 30827.

We report theoretical/experimental investigations of modifications of electronic and structural properties of 2D materials induced by compression. The experiments are performed with a combination of scanning probe techniques, and we employ theoretical techniques involving first-principles, molecular dynamics and analytical calculations.

We first address the effects of out-of-plane (c-axis oriented) compression of few-layer 2D materials [1,2]. The experiments indicate a metal-to-insulator electronic modification with increasing pressure in few-layer graphene [1], and the opposite behavior in hexagonal boron nitride (h-BN) [2]. First-principles modelling of both materials in the presence of water molecules indicates that this phenomenology is associated to the formation of lamellar sp³ structures under pressure [1,2].

We also address the effects of in-plane uniaxial compressive stress on the top layers of a multilayer 2D material supported on a substrate. We predict, through molecular dynamics simulations, that a critical stress of this type induces spontaneous symmetry breaking that evolves into planar folds oriented perpendicularly to the surface, which we name “standing folds” [3]. We observe such standing folds in h-BN, graphene, and 2D talc [4,5] , and we propose that the dynamical production of these folds under shear is also responsible for transient thickening of sheared 2D materials [6,7]. The folds have widths of few layers of the 2D material, and heights of several tens of nanometers, up to more than 100 nm [4,5], essentially behaving as suspended few-layers of the 2D material. This allowed us to detect the onset of the size-dependent flexural hardening of h-BN, graphene, and 2D talc through a combination of theory and experiments on the standing folds [5].

[1] A. P. M.Barboza et al., Adv. Mater. 23, 3014 (2011).
[2] A. P. M. Barboza et al., ACS Nano 12, 5866 (2018).
[3] A. L. Lima et al., Nanotechnology 26, 045707 (2015).
[4] C. K. Oliveira et al., Nano Research 8, 1680 (2015).
[5] H. Chacham et al., Nanotechnology 29, 095704 (2018).
[6] A. P. M. Barboza et al., Nano Lett. 12, 2313 (2012).
[7] A. B. Alencar et al., 2D Mater. 2, 015004 (2015).

Hexagonal boron nitride (h-BN) also known as “white graphene” is a prominent member of the 2D materials family, which has gained a significant attention for its remarkable properties in the last decade. Its superior thermal conductivity (1700-2000W mK^-1), high mechanical strength (15.7 Nm^-1), large band gap energy (6 eV), atomically smooth surface, ultra-high flexibility, and transparency make h-BN a suitable candidate for numerous applications such as flexible and transparent electronics.

Owing to its binary element nature, the growth of high quality and large area CVD h-BN with controlled morphology is still a challenging task. CVD grown single crystalline h-BN domains exhibit various morphologies such as triangles, truncated triangles, hexagons, and other irregular shapes. Oftentimes edges of the triangular domains are nitrogen terminated while the edges of hexagonal domains are alternatively nitrogen and boron terminated¹. Similarly, other shapes (polygon, circular and elliptical etc.) have their specific termination at the edges⁵. The morphology of h-BN domains has been found to depend on various parameters including surface roughness¹, growth temperature², water assisted argon to hydrogen ratio³, position of the substrate⁴, hydrogen passivation⁵, substrate orientation⁶ and the annealing time of the substrate⁷.

In this work, we investigated the impact of precursor mass on the modification of the shape of h-BN domains. Synthesis of h-BN has been done by CVD process on polycrystalline copper foils using Ammonia Borane (AB) as a precursor. It is observed that the mass of the AB plays a key role in shape-influencing parameters of h-BN domains. With the increase in the weight of AB, h-BN growth is dominated by compact hexagonal domains whilst a decrease in the weight (<30 mg) leads to triangular and other irregular domains. We also observe that diluted hydrogen leads to the appearance of unusual elliptical and circular shapes.

Characterization of h-BN samples has been done by XPS, UV-Vis spectroscopy, Optical microscopy RAMAN, ToF-SIMS, and SEM. XPS data confirms that the binding energy of boron and nitrogen is around 190 eV and 398 eV, respectively, which are very close to the values reported in the literature. As grown h-BN on the top surface of the foil (directly exposed to precursor flow) and on the bottom surface (confined with the quartz support) shows variable nitrogen to boron ratio. We calculated N:B ratio ranging from 0.7 to 1.11 in our samples, which clearly signifies nitrogen-rich or boron-rich h-BN regions. UV-Vis spectroscopy confirms a very sharp absorption peak near 202 nm corresponding to a bandgap of 6.13 eV.

The evolution of these morphologies is attributed to the alteration of N:B ratio evolved by the mass of the AB, which has a significant impact on the decomposition process. N:B ratio can be tuned by the decomposition monitoring of ammonia borane in order to get better control over the shape of synthesized h-BN domains. Furthermore, optimal AB decomposition conditions will be investigated by the AB mass, AB heating profile, background pressure and concentration of carrier gasses to get direct access to N:B ratio.

In summary, the shape of the h-BN domains can be used as a direct indicator of nitrogen-rich or boron-rich growth. Certainly, variation in the shape of domains will influence the intrinsic and functional properties of grown h-BN layers. These results envisage that depending on the shape of individual h-BN domains, CVD synthesis will lead to either nitrogen-rich or boron-rich in-plane regions of the 2D h-BN.
References
Roland Y.T. et al. Nano Lett. 14, 839-846 (2014).
Xiufeng S. et al. 2D Mater. 3, 035007 (2016).
Lifeng et al. Mater. Chem. Front. DOI: 10.1039/c7qm00100b.
Yijing S. et al. Chem. Mater 27, 8041-8047 (2015).
Jingzhao Z. et al. Nanoscale DOI: 10.1039/C8NR04732D.
Yanxin J. et al. ACS Nano 11, 12057-12066 (2017).
Qinke W. et al. Scientific Reports 5:16159 (2015).

Semiconducting two-dimensional materials such as transition metal dichalcogenides are promising for advanced atomically thin electronic and optoelectronic devices. Searching for new two-dimensional optoelectronic semiconductors, we have identified unusual functional properties in two novel classes of layered materials. One class includes p-elements such as Ge, Sn, and Pb. Although these group-IV elements have four valence electrons, they frequently occur in the 2+ charge state in ionic compounds such as monochalcogenides and monoxides. Our atomistic calculations based on density functional and many-body perturbation theory revealed that materials such as SnSe, GeSe, and PbO exhibit unusual properties including anisotropic spin transport, strong visible light absorbance, and strong exciton binding. Another class consists of hydrogen-passivated group-V nitrides such as AlN, GaN, and InN. Unlike hexagonal BN, these heavier nitrides favor hydrogen passivation and crystallize in a buckled wurtzite-like crystal structure that induces a strong electrical polarization perpendicular to the atomic planes. As a result of the polarization, electrons and holes are spatially separated and their energies strongly reduced by the polarization field. Thus, electrical polarization plays an essential role in controlling the exciton binding energy and radiative lifetime, and can be exploited in multilayered structures to tune the luminescence energy from the visible to the deep UV. This work was supported by the NSF ECCS-CDS&E program (1607796). Computational resources were provided by the DOE NERSC facility (DE-AC02-05CH11231).

One of the most exciting properties of two dimensional materials is their sensitivity to external tuning of the electronic properties, for example via electric field or strain. Recently discovered analogues of phosphorene, group-IV monochalcogenides (MX with M = Ge, Sn and X = S, Se, Te), display several interesting phenomena related to the in-plane strain, such as giant piezoelectricity and multiferroicity, which combines ferroelastic and ferroelectric properties. Here, using calculations from first principles, we predict for the first time giant spin Hall effect (SHE) in these materials which suggests their high potential for 2D spintronics. We reveal that the spin Hall conductivity is tunable via combinations of external strain and doping. In some configurations, the strain-induced semiconductor to metal transition enables the logic functionality of switching on/off spin currents which indicates a new route for the design of multi-tunable spintronics devices.

Acknowledgements. The members of the AFLOW Consortium acknowledge support by DOD-ONR (N00014-13-1-0635, N00014-11-1-0136, N00014-15-1-2863). MBN and FTC acknowledge partial support from Clarkson Aerospace Corporation. The calculations were performed at the Computing Center at the University of North Texas and the Texas Advanced Computing Center at the University of Texas, Austin.

There has been increasing research interest in two-dimensional (2D) materials with honeycomb structure due to their potential applications in electronic devices. Among various 2D materials, group III-V materials such as BN with hexagonal (Hex) structure have been paid much attention. However, this is available mainly in the form of small flakes, and successful fabrication of BN thin films with large area is limited. Recently, Tsipas et al. have successfully fabricated graphite-like AlN by molecular beam epitaxy (MBE) on Ag(111) substrate. [1] They have also reported that ultrathin (sub-monolayer to 12 monolayers) AlN adopts Hex structure with larger lattice constant compared to bulk AlN with three-dimensional wurtzite (WZ) structure. Two-dimensional GaN can also be formed by migration-enhanced encapsulated growth technique using epitaxial graphene. [2] In spite of the experimental findings, there is a still open question how much layers the Hex structure is stabilized for group III-V materials. From theoretical viewpoints, we have previously investigated the stability of group-III nitrides with Hex structure by mean of density functional theory (DFT) calculations, and revealed that five-fold coordinated Hex structure is more stable than WZ structure for the films up to 15 bilayers due to interlayer electrostatic interactions. [3] In this study, we extend our approach to various group III-V materials including binary and ternary alloys. Our DFT calculations for various III-V materials (BP, AlP, GaP, InP, GaAs, AlAs, and etc.) demonstrate that they forms five-fold coordinated Hex structure and its stability depends on their ionicity: The number of layers up to which Hex structure has a lower cohesive energy than WZ structure, L_h, becomes large for III-V materials with larger ionicity. [4] Furthermore, the calculations using empirical bond-order potentials with the aid of DFT calculations [5] for B_xAl_1-xN and B_xGa_1-xN alloys reveal that the Hex structure can be stabilized depending on the number of film thickness. The calculated L_h is also found to be dependent on boron composition x, and the miscibility of these alloys with Hex structure is lower than those with WZ structure due to the contribution of relaxation of atoms depending on the crystal structure. These calculated results thus suggest that various 2D ultrathin films can be formed in group III-V materials. Effects of substrates such as Ag(111) and graphene on the stability of 2D ultrathin films are also examined.
References
[1] P. Tsipas et al., Appl. Phys. Lett. 103, 251605 (2013).
[2] Z. Y. A. Balushi et al., Nature Materials 15, 1166 (2016).
[3] Y. Tsuboi, T. Akiyama, K. Nakamura, and T. Ito, Phys. Status Solidi 255, 1700446 (2018).
[4] T. Akiyama, Y. Tsuboi, K. Nakamura, and T. Ito, submitted.
[5] Y. Hasegawa, T. Akiyama, A. -M. Pradipto, K. Nakamura, and T. Ito, J. Cryst. Growth 504, 13 (2018).

We have recently shown that NaSn₂As₂ exhibits opposite conduction polarities along in-plane and cross-plane directions, defined as “goniopolarity”. On the lowest level of theory, we can show that this novel phenomenon originates from a special topology of the Fermi surface, which is essentially determined by the nature of the bonding states in this layered crystal. In this paper, we present an improved fundamental understanding of goniopolarity based on a novel formulation of goniopolarity within the tight-binding model. The tight-binding matrix elements are calculated from GW calculations based on Density Functional Theory (DFT) via maximally localized Wannier functions. Considering a minimum-basis set with sp³ orbitals for both Sn and As, , the 64 hopping integrals are evaluated. Within our model, we provide a new tight-binding based formulation for both Seebeck and Hall coefficients for NaSn₂As₂ and can show that critical ratios of hopping and on-site matrix elements exist that pose limits for goniopolarity to appear in materials. Based on that, additional candidate materials for goniopolarity can be proposed, and the design space for goniopolar materials in general will be defined.

I will present first-principles calculations and predictions of novel electronic, optical, thermal, and polarization properties of a new family of two-dimensional materials, black phosphorus and its group IV-VI isoelectronic materials. For black phosphorus, we predicted a wide range of its fundamental characters, such as quasiparticle band gaps, excitons, and anisotropic thermal conductance. We also proposed to engineer various properties of black phosphorus, such as strain-tunable anisotropic electrical conductance and pressure-induced topological nodal-line semi-metallic states. Beyond black phosphorus, we found that the group IV-VI isoelectronic materials (monolayer GeS, GeSe, SnS, and SnSe) may exhibit unusual polarization properties. Their in-plane piezoelectric coefficients are about two orders of magnitude larger than those of monolayer transition metal dichalcogenides and conventional bulk piezoelectric materials. Moreover, these monolayer group IV monochalcogenides are essentially spontaneously polarized and exhibit a robust in-plane ferroelectric effect above room temperature. Many of these predictions have been confirmed by recent experiments, and they also provide guidance for future applications.

Two-dimensional (2D) materials have shown great potential for many applications such as electrocatalysis and electronics. However, their developments are impeded by a number of significant challenges, many of which root in the 'problems of electrons'. On the physics side, the electrons/holes in current 2D semiconductors do not move fast enough under electric field (i.e. low mobility), impeding their use in high-performance electronics and optoelectronics. On the chemistry side, the current 2D electrocatalysts still suffer from relatively low activity/selectivity, hindering their use in efficient energy conversion. These challenges urge for an improved understanding of the electrons in 2D materials, and using these insights to design/discover 2D host materials with better performance.
Here I will discuss our recent progress in understanding the chemistry and physics of electrons in 2D materials for electronic and energy applications. Specifically, I will show (1) what limits the intrinsic carrier mobility of two dimensional metal dichalcogenides, and how to find high-mobility 2D semiconductors [1]; (2) the significant impact of charge on electrochemical reactions of 2D materials, which is often neglected in modeling their electrochemistry, and the implications for energy applications [2].
[1] D. Kim, J. Shi, Y. Liu, JACS, 2018, 140, 9127
[2] L. Cheng, Y. Liu, JACS, 2018 DOI: 10.1021/jacs.8b07871

We investigate topological structure of complex energy surfaces of non-equilibrium magnetic systems. We consider one- and two-dimensional spin-transfer torque-driven models described by non-Hermitian Hamiltonians. The spectrum of the Hamiltonian admits branch point singularities that appear as chiral exceptional points where both eigenvalues and eigenstates coalesce. Encircling individual exceptional points in parameter space has non-trivial physical consequences resulting in non-reciprocal spin wave transfer through the system. This effect can be used in a spintronic spin filter device.

Ubiquitous applications of rechargeable batteries, particularly in portable electronics and electric vehicles, have stimulated a vigorous search for high-capacity electrode materials. In particular, Lithium-Oxygen batteries have attracted substantial attention due to their superior energy-density; yet several technical challenges must be overcome in order to enable their successful commercialization. It is well-known that the precipitation of the reaction products from the oxygen reduction reactions (ORR) during the discharge process increases the overpotential and in turn leads to reduced cyclability. MXenes are a new class of two-dimensional inorganic compounds. The versatile chemistry and the high electrical conductivity of these materials make them a prevalent candidate for catalysis. In this study, free-standing 2D titanium nitride MXene is proposed as an efficient cathode-catalyst for non-aqueous Li-O₂ batteries. By performing comprehensive first-principles simulations employing density-functional theory and reactive molecular dynamics, we discovered the enhanced catalytic activity for ORR and oxygen evolution reactions (OER) through surface functionalization of MXene monolayers.
Atomic-scale analysis was performed to identify the role of surface functionalization for controlling nucleation of Li₂O₂ and electrochemical reaction steps of ORR/OER on MXene surfaces. The free-energy pathways for adsorption of the intermediate and final products of ORR are determined, demonstrating that synergistic effects of vacancies and functionalized groups accelerate ORR by up to a two orders of magnitude compared to commonly used graphene-based cathode electrodes. These results contribute towards the design of functionalized MXene for modulating catalytic properties of ORR/OER in cathodes of Li-air batteries.

Thermal transport in nanoscale two-dimensional (2D) materials is of great scientific interest and has practical implications for energy related applications like thermal management of energy devices, composite battery materials and on-board thermoelectric power generation for sensors. The abilities to manipulate thermal transport in 2D materials is thus highly desirable for future nano energy technologies. In this work, we identify a general rule for controlling the thermal transport in 2D pentagonal materials through bond saturation. We use first-principles calculations to investigate the phonon properties of a series of pentagonal materials, including penta-graphene (PG), hydrogenated PG (h-PG) and fluorinated PG (f-PG), and find that the bond saturation of the carbon atoms through functionalization can reduce the bond anharmonicity and thus increase the phonon lifetime. We can follow this rule to predict very high thermal conductivity of other pentagonal structures with saturated bonds, including penta-CN₂ (1027 W/mK) and two three-dimensional counterparts of PG called T12-carbon (819 W/mK) and AA T12-carbon (1049 W/mK). Moreover, similar trend of bond saturation-induced thermal conductivity enhancement can be found in other 2D pentagonal materials, such as penta-SiC₂ and penta-SiN₂. The results from this work unveil a general bonding-thermal transport relation for 2D materials, which can provide important guidance for designing novel materials with desirable thermal transport properties for energy applications.

A Raman peak of a material is characterized by a temperature-dependent width. Accurate determination of the widths of the Raman peaks and their behaviour with temperature is of fundamental and practical importance to the characterization of a material. We employ a semi ab-inito theory of anharmonic interactions [1] to obtain accurate results for the linewidths of the Raman active phonon modes in pristine bulk and monolayer transition metal dichalcogenides MoS₂, WS₂ and MoTe₂. For each Raman mode we extract the contribution towards its linewidth originating from three-phonon coalescence and decay processes. In addition, the relative contributions of the Klemens channel [2], Ridley channel [3], Vallèe-Bogani channel [4], and Barman-Srivastava channel [5] to the overall anharmonic decay contribution are established. An explanation is proposed for the difference in the linewidth results (i) in low and high frequency modes for a given system, (ii) across the TMDs, and (iii) between bulk and monolayer systems. Temperature-dependent measurements of the linewidths for the high frequency A and E modes for a few-layer MoS₂ [6] and for the monolayer MoS₂ [7] are explained through combining the anharmonic, isotopic point defect, and background inhomogeneity contributions.

[1] G. P. Srivastava and I. O. Thomas, Phys. Rev. B 98, 035430 (2018).
[2] P. G. Klemens, Phys. Rev. B 148, 845 (1966).
[3] B. K. Ridley and R. Gupta, Phys. Rev. B 43, 4939 (1991).
[4] F. Vallèe and F. Bogani, Phys. Rev. B 43, 12049 (1991).
[5] S. Barman and G. P. Srivastava, Phys. Rev. B 69, 235208 (2004).
[6] S. Sahoo et al, J. Phys. Chem. 117, 9042 (2013).
[7] N. A. Lanzillo et al, Appl. Phys. Lett. 103, 093102 (2013).

Physical behaviors of 2D materials are ultimately governed by complex interactions of electrons and ions, ranging from ultrafast process at atomistic scale to slow dynamics at mesoscale. Here we present our recent development of first-principles theory based machine learning force fields for 2D materials simulations. We will show one example on 2D ferroelastic-ferroelectric GeSe, including temperature induced phase transition and strain-induced domain switching. Interestingly, three distince stages of domain switching were observed, involving from primary domain nucleation and growth, to transient local deformation induced growth, and to secondary domain nucleation and growth. The generated force fields are able to capture essential physics both qualitatively and quantitatively such as energetic, structural, and dynamic properties as well as complex phase transition, opening up avenues for understanding and predicting complex physical behaviors of 2D materials.

Low-frequency (LF) Raman spectroscopy has emerged as a very useful tool for characterization of the number of layers and stacking configurations in two-dimensional materials based on simple linear chain (LCM) and interlayer bond polarizability models. Here, we show unusual effects of strong interlayer coupling on low-frequency Raman scattering in PdSe₂crystals with different number of layers. We found that the measured frequencies of the breathing modes cannot be described by a conventional LCM that treats each layer as a single rigid object. We show that strong deviations from layer rigidity can occur for the LF breathing vibrations of PdSe₂, which accounts for the observed disagreement with the conventional LCM. The layer non-rigidity and strong interlayer coupling could also explain the unusual strong intensities of the LF breathing modes that are comparable with those of the high-frequency Raman modes. These strong intensities allowed us to use a set of the measured LF Raman lines as unique fingerprints for a precise assignment of the layer numbers from 2 to 19. The assignment of the layer numbers was further confirmed using second harmonic generation that appeared only in the non-centrosymmetric even-layer PdSe₂crystals. This work demonstrates a simple and fast approach for the determination of the number of layers in 2D materials with strong interlayer coupling and non-rigid interlayer vibrations.
Synthesis science was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences (BES), Materials Sciences and Engineering Division. Characterization and computational science at CNMS were supported by the Scientific User Facilities Division, BES.

Giant nonlinear optical responses were recently observed in 2D materials. Understanding their microscopic origin is important for both fundamentals of nonlinear optics and development of technological applications. Here I will report our recent first-principles theoretical development towards unveiling the electronic structure origin of their nonlinear optical responses, including (a) second harmonic generation [1] and nonlinear photocurrent [2] in 2D noncentrosymmetric materials and (b) nonlinear ferroelectric photocurrent in topological 2D materials [3]. I will also discuss microscopic relations between nonlinear responses and topological quantities such as shift vector, Berry curvature dipole, etc. Our findings suggest the possibility to achieve deep understanding of nonlinear optical responses in 2D materials using firstprinciples approaches. This research was supported by the National Science Foundation under award number NSF DMR-1753054. References: [1] H Wang and X Qian. Nano Letters 17, 50275034 (2017). [2] H Wang and X Qian. Quantum Nonlinear Ferroic Optical Hall Effect. submitted (2018). [3] H Wang and X Qian. Nonlinear Ferroelectric Photocurrent in Topological Materials. in preparation (2018).

With rapid advances in the synthesis and characterization of two-dimensional (2D) π-conjugated organic covalent frameworks (COFs), the search for 2D COF-based materials that can display –like graphene– a Dirac cone at the Fermi level, is attracting much interest. Radical-carrying carbon-based hexagonal COFs, due to their similarity to graphene, are natural choice. While there are recent theoretical reports ^1-3 of Dirac-cone electronic structures in such 2D COFs, the chemical, structural, and electronic stabilities of these systems have not been fully addressed.

Here, we discuss a series of 2D COFs consisting of three radical-carrying building blocks and various linkers, and investigate their relative stabilities in different electronic configurations and geometrical structures, including closed-shell semimetal and semiconducting electronic configurations, and open-shell ferromagnetic and antiferromagnetic configurations. Density functional theory (DFT) calculations are carried out at the global-hybrid PBE0 level to optimize the geometric structures and determine the electronic band structures and to estimate the Hubbard on-site repulsion parameter U and inter-site electronic coupling t. In contrast to the conclusions of earlier studies ^1-3 (which either did not consider spin-polarized configurations or attributed the presence of magnetically ordered phases to a symmetry lowering of the COF lattice geometry from hexagonal to trigonal), we find that the most stable phase of these 2D COFs is in fact an antiferromagnetic (AFM) Mott insulating phase with the same lattice symmetry as in the semimetal (SM) phase.

To rationalize the appearance of the AFM phase, the systems can be cast as “effective graphene lattices” ⁴ and described in the context of the standard Hubbard model. Importantly, earlier Quantum Monte Carlo (QMC) studies of a half-filled honeycomb lattice, based on the Hubbard model, have revealed a rich phase diagram spanning from nonmagnetic semimetal to spin-liquid and to insulating antiferromagnet (AFM), as a function of the U/t ratio.⁴ The QMC investigations ⁴ indicate that the insulating AFM phase is the most stable when U/t ≥ 4.3, while the nonmagnetic semimetal phase appears when U/t ≤ 3.5 (as is the case in graphene), and a disordered spin-liquid phase is predicted in the intermediate U/t range (3.5 ≤ U/t ≤ 4.3). As a function of the nature of the building blocks and linkers, the U/t ratios in our 2D COFs evolve between ca. 16 and 6 (with U estimated to be in the 2.2-1.3 eV range and t, in the 0.1- 0.2 eV), which places all our 2D COFs on the AFM side of the phase diagram. Interestingly, 2D COFs with a U/t ratio approaching 6, appear close to the boundary between the antiferromagnetic and spin-liquid phases. Further molecular design, aimed at increasing the electronic coupling t and decreasing the on-site repulsion U, can thus be expected to push the U/t ratio onto the spin-liquid phase and even the semimetal phase.

1. Adjizian, J.-J.; Briddon, P.; Humbert, B.; Duvail, J.-L.; Wagner, P.; Adda, C.; Ewels, C., Nat. Commun. 2014, 5, 5842.
2. Alcón, I.; Viñes, F.; Moreira, I. d. P. R.; Bromley, S. T., Nat. Commun. 2017, 8 (1), 1957.
3. Jing, Y.; Heine, T., J. Am. Chem. Soc. 2018. DOI: 10.1021/jacs.8b09900
4. Meng, Z. Y.; Lang, T. C.; Wessel, S.; Assaad, F. F.; Muramatsu, A., Nature 2010, 464, 847.

Molybdenum disulfide (MoS₂) is a promising two-dimensional (2D) semiconductor. Similar to graphite, MoS₂ has a layered structure comprising weak van der Waals bonding between layers, and strong covalent bonding within layers. The weak secondary bonding allows for isolation of these 2D materials to a single layer, like graphene. While bulk MoS₂ is an indirect band gap semiconductor with a band gap of ~1.3 eV, monolayer MoS₂ exhibits a direct band gap of ~1.8 eV, making it an attractive candidate for replacing Si in electronic devices. Atomic layer deposition (ALD) has been used previously to grow MoS₂ films using molybdenum chloride and carbonyl precursors. However, these precursors are solids at room temperature and require high temperature vapor transport. Recently, MoS₂ ALD using molybdenum hexafluoride (MoF₆), a high vapor pressure liquid at room temperature, and hydrogen sulfide (H₂S) has been demonstrated. While the nucleation of MoS₂ during chemical vapor deposition (CVD) is understood, the nucleation of MoS₂ ALD using MoF₆ and H₂S on dielectric surfaces has yet to be explored. Unlike films grown by high temperature CVD, ALD MoS₂ is amorphous as deposited and must be annealed to crystallize the film. In this study, we utilized in situ quartz crystal microbalance and Fourier transform infrared spectroscopy measurements to investigate the first few cycles of MoS₂ ALD on Al₂O₃, HfO₂, and MgO surfaces prepared in situ by ALD. The MoS₂ nucleation, and in particular the effect of the first MoF₆ exposure, was found to depend strongly on the underlying metal oxide, the OH surface concentration, and the growth temperature. Following these in situ studies, we prepared ultrathin ALD MoS₂ films and subsequently annealed them in H₂ to crystallize the films for device measurements. Our work will provide a solid understanding of the interactions of MoF₆ and H₂S with dielectric surfaces, which is critical for the implementation of ALD MoS₂ in microelectronics.

Transition metal dichalcogenide (TMD) thin films are promising candidate materials in the continuing development of nanoscale devices. These materials exhibit a direct band-gap when thinned down to a single layerand also provide properties that include an improved on-off ratio and photonic application which set them apart from bulk silicon. Methods to produce wafer-scale TMD films include powder chemical vapor deposition (CVD), liquid-phase and mechanical exfoliation, metal-organic CVD, and atomic layer deposition (ALD). These methods suffer from issues with particulate contamination, pyrophoric precursors, and costly precursors along with the lack of reproducibility and layer thickness controllability. Here we demonstrate the growth of homogeneous wafer-scale mono-, bi-, and tri-layer molybdenum disulfide (MoS₂) using solid inorganic and liquid organic precursors in a high-vacuum environment. These samples are grown on amorphous 30-300 nm SiO₂/Si substrates and fused silica windows and without any powder or metal-organic precursors. Growth proceeds by the decomposition of carbon disulfide at a heated molybdenum filament, which yields volatile MoS_x precursors that aggregate on a heated wafer. We have developed a procedure to control our thin film growth using a standard video camera from the top of the chamber to monitor the hue (0-255) of the film from the reflection of the filament onto the top of the SiO₂/Si. This allows controlled integer number layer thickness of our films. The continuous and homogeneous single-layer film of MoS₂ is deposited at wafer-scale with a total growth time of approximately an hour. Optical and electrical characterization indicates performance comparable to or better than MoS₂ film grown by other wafer-scale growth techniques. Our method provides a scalable process to deposit thin TMD films in a high-vacuum environment using a colorimetric hue control.

Novel bi-phase synthesis method is introduced to synthesize highly crystalline, perfectly layered, and industrially scalable vdW Metal-Organic Frameworks (MOFs) growth technique. Usually, 2D MOFs synthesis is carried out involving combination of ditopic ligands and metals with four (or six) coordination. However, the formation of hydrogen bonds between layers is a main challenge for separating individual flakes with large lateral dimension and product yield. Our innovative process shows that replacing water molecules in the reaction with pyridine eliminates hydrogen bond formation at metal cluster sites. This prohibits tight coupling across adjacent MOF layers and sustains controllable 2D vdW MOF-2 growth. Owing to its large lateral size, vdW nature, and high crystallinity, it is possible to perform Kelvin probe force microscope on MOF-2 crystals with varied thickness for the first time. Furthermore, the vdW nature of MOF-2 displays unique anisotropic vibration behaviors. Results not only establish its fundamental properties and layer-dependent responses but also show striking differences from other 2D inorganic materials.

The desire to exploit the unique properties of 2D materials requires the synthesis of large-scale high-quality 2D materials. While epitaxial growth of 2D materials on single-crystal substrates can generate large-scale production of 2D materials without compromising the quality, it is unclear whether the 2D materials can exist as structurally and electronically distinct layers anchored by van der Waals (vdW) forces, as opposed to more strongly bound adlayers. The strength of these interactions, from weak van der Waals to strong covalent bonds, are linked to the local atomic structure of the interface. The resulting 2D materials exhibit different properties depending on the strength of the interactions with the underlying substrate.

We used x-ray standing waves (XSW) generated by Bragg diffraction from a sapphire single crystal substrate to excite X-ray fluorescence (XRF) from a 2D epitaxial layer of WSe₂. These measurements, performed at rotating anode and synchrotron x-ray sources, were used to find the positions of Se and W atoms relative to the substrate lattice with subangstrom resolution.

These results combined with DFT calculations could give distinct insights into WSe₂ and other transition metal dichalcogenides integrated epitaxially on oxide substrates. Overall, the work herein reveals how distinct X-ray surface science techniques can be used to understand questions about 2D material structure and interfaces.

Acknowledgements: NU Center for Hierarchical Materials Design (NIST 70NANB14H01), NU-MRSEC (DMR-1121262) and Argonne National Laboratory (DOE DE-AC02-06CH11357). The Pennsylvania State University Two-Dimensional Crystal Consortium – Materials Innovation Platform (2DCC-MIP) supported by NSF cooperative agreement DMR-1539916.

Two-dimensional transition metal dichalcogenides (TMDCs) have attracted interest for their compelling nanoscale new properties and numerous potential applications including fast optoelectronic devices, ultrathin photovoltaics, and high-performance catalysts. Large-scale growth of uniform TMDC materials is essential for investigating their physics and for their integration into device. However, the wafer scale deposition of TMDCs on arbitrary non-selective substrates is still beyond the current state-of-the-art. In this article, we report a method to synthesize layered TMDCs (MoS₂ and WS₂) at the wafer-scale by sulfurization of transition metal ions (Mo⁵⁺ and W⁶⁺) in a gelatin template (metallo-hydrogel). This process is adaptable to versatile substrates, including amorphous silicon oxide, high-temperature quartz, and silicon. Although the products are nominally few layer materials, we observe direct band photoluminescent (~1.8 eV), consistent with single- or decoupled multilayer MoS₂. This is attributed to the presence of surfactants included in the template that serve to isolate the TMDC layers electronically. Finally, the solution-based deposition enables contact printing of TMDC channels useable for device applications including thin film transistors with printed silver contacts using the same process.

van der Waals (vdW) heterostructures display a tremendous potential as building blocks for novel nanoscale and quantum technologies. These stacks of various two-dimensional (2D) materials can exhibit a wide variety of tunable physical properties. Nonetheless, their integration in mainstream technologies is hindered by the lack of large-scale synthesis methods. In fact, most vdW heterostructures are produced by mechanical transfer and yield micrometer scale heterostructures. Developing large-scale epitaxial growth processes for vdW heterostructures is therefore critical. Even though some vdW heterostructures have been produced by epitaxy, the control over their growth is remains limited and no global understanding of the vdW epitaxy mechanisms currently exist. In fact, their growth has never been observed in real-time. To address this problem, we present here an in situ low-energy electron microscopy (LEEM) study of the vdW epitaxy of antimonene-graphene vdW heterostructures on germanium substrates. Graphene was grown by Chemical Vapor Deposition (CVD) on Ge(100), Ge(110) and Ge(111). Then antimonene, a newly discovered two-dimensional material with several intriguing thickness dependent electronic and topological properties, is grown by solid-source molecular beam epitaxy (MBE) under direct LEEM observation. Nucleation of antimonene mainly occurs on small 3D nuclei which form on defects on the graphene surface. The antimonene then grows laterally to form few micrometers large flakes. The antimonene islands rarely coalesce due to competition for the precursor Sb₄ species. The growth morphology is strongly influenced by the Sb₄ deposition rate. In fact, high deposition rates (>10 nm/minute) lead to the growth of large flat islands with 3 branches, whereas lower deposition rates lead to triangular atoll-like islands with thick bands along the edges. The transition from 2D to atoll-like growth is explained based on the interplay between the growth rate, surface diffusion and edges orientation. The understanding of the vdW growth developed in this study will allow for a better control of the large-scale growth of vdW heterostructures with highly tunable physical properties.