Controllable defects in monolayer graphene induced by hydrogen and argon plasma.

Graphene has attracted wide attentions since its successfully exfoliation. Honeycombsp2carbon lattice and Dirac semi-metal band structure make graphene a promising material with excellent mechanical strength, thermal conductivity, and carrier mobility. However, the absence of intrinsic bandgap limits its application in semiconductor. Defects in graphene is supposed to modify its band structure and lead to an opened bandgap. Many methods have been demonstrated to introduce defects into graphene, such as chemical reaction, plasma, electron beam, and laser. However, the species of defects are mostly uncontrollable in most treatment processes. In this study, we report three kinds of defects can be controllably induced in graphene via hydrogen (H2) and argon (Ar) plasma. With different parameter and feeding gas, hydrogenated graphene, graphene nanomesh and graphene with vacancies can be well obtained. The defect density can be precisely controlled by tuning plasma power and irradiation time. Morphological, spectroscopic, and electrical characterizations are performed to systematically investigate the defect evolution. Graphene nanomesh and graphene with vacancies show obvious difference for roughness and coverage, whereas the morphology of hydrogenated graphene remains similar with that of as-prepared graphene. For hydrogenated graphene, an opened bandgap of ∼20 meV is detected. For graphene nanomesh and graphene with vacancies, the semiconductive on/off behaviors are observed. We believe this work can provide more details of plasma-induced defects and assist the application of graphene in semiconductor industry.

Stacking transfer of wafer-scale graphene-based van der Waals superlattices.

High-quality graphene-based van der Waals superlattices are crucial for investigating physical properties and developing functional devices. However, achieving homogeneous wafer-scale graphene-based superlattices with controlled twist angles is challenging. Here, we present a flat-to-flat transfer method for fabricating wafer-scale graphene and graphene-based superlattices. The aqueous solution between graphene and substrate is removed by a two-step spinning-assisted dehydration procedure with the optimal wetting angle. Proton-assisted treatment is further used to clean graphene surfaces and interfaces, which also decouples graphene and neutralizes the doping levels. Twist angles between different layers are accurately controlled by adjusting the macroscopic stacking angle through their wafer flats. Transferred films exhibit minimal defects, homogeneous morphology, and uniform electrical properties over wafer scale. Even at room temperature, robust quantum Hall effects are observed in graphene films with centimetre-scale linewidth. Our stacking transfer method can facilitate the fabrication of graphene-based van der Waals superlattices and accelerate functional device applications.

Stack growth of wafer-scale van der Waals superconductor heterostructures.

Two-dimensional (2D) van der Waals (vdW) heterostructures have attracted considerable attention in recent years1-5. The most widely used method of fabrication is to stack mechanically exfoliated micrometre-sized flakes6-18, but this process is not scalable for practical applications. Despite thousands of 2D materials being created, using various stacking combinations1-3,19-21, hardly any large 2D superconductors can be stacked intact into vdW heterostructures, greatly restricting the applications for such devices. Here we report a high-to-low temperature strategy for controllably growing stacks of multiple-layered vdW superconductor heterostructure (vdWSH) films at a wafer scale. The number of layers of 2D superconductors in the vdWSHs can be precisely controlled, and we have successfully grown 27 double-block, 15 triple-block, 5 four-block and 3 five-block vdWSH films (where one block represents one 2D material). Morphological, spectroscopic and atomic-scale structural analyses reveal the presence of parallel, clean and atomically sharp vdW interfaces on a large scale, with very little contamination between neighbouring layers. The intact vdW interfaces allow us to achieve proximity-induced superconductivity and superconducting Josephson junctions on a centimetre scale. Our process for making multiple-layered vdWSHs can easily be generalized to other situations involving 2D materials, potentially accelerating the design of next-generation functional devices and applications22-24.

Trapping Hydrogen Molecules between Perfect Graphene.

There is a lack of deep understanding of hydrogen intercalation into graphite due to many challenges faced during characterization of the systems. Therefore, a suitable route to trap isolated hydrogen molecules (H2) between the perfect graphite lattices needs to be found. Here we realize the formation of hydrogen bubbles in graphite with controllable density, size, and layer number. We find that the molecular H2 cannot be diffused between nor escape from the defect-free graphene lattices, and it remains stable in the pressurized bubbles up to 400 °C. The internal pressure of H2 inside the bubbles is strongly temperature dependent, and it decreases as the temperature rises. The proton permeation rate can be estimated at a specific plasma power. The producing method of H2 bubbles offers a useful way for storing hydrogen in layered materials, and these materials provide a prospective research platform for studying nontrivial quantum effects in confined H2.

Growth of wafer-scale chromium sulphide and selenide semiconductor films.

Two-dimensional (2D) transition metal chalcogenides have attracted enormous attention due to their stunning properties and great prospects for applications. Most of the reported 2D materials have layered structure, and non-layered transition metal chalcogenides are rare. Particularly, chromium chalcogenides are highly complexed in terms of structural phases. Researches on their representative chalcogenides, Cr2S3and Cr2Se3, are insufficient and most of them focus on individual crystal grains. In this study, large-scale Cr2S3and Cr2Se3films with controllable thickness are successfully grown, and their crystalline qualities are confirmed by multiple characterizations. Moreover, the thickness-dependent Raman vibrations are investigated systematically, presenting slight redshift with increasing thickness. The fundamental physical properties of grown Cr2S3and Cr2Se3films, including optical bandgap, activation energy and electrical properties, are measured with different thicknesses. The 1.9 nm thick Cr2S3and Cr2Se3films show narrow optical bandgap of 0.732 and 0.672 eV, respectively. The electrical properties of Cr2S3films demonstratep-type semiconductor behaviours, while the Cr2Se3films exhibit no gate response. This work can provide a feasible method for growing large-scale Cr2S3and Cr2Se3films, and reveal fundamental information of their physical properties, which is helpful for future applications.

Controllable growth of wafer-scale monolayer transition metal dichalcogenides ternary alloys with tunable band gap.

The tuning of band gap is very important for the application of two-dimensional (2D) materials in optoelectronic devices. Alloying of 2D transition metal dichalcogenides (TMDCs) is an important way to tune the wide band gap. In this study, we report a multi-step vapor deposition method to grow monolayer TMDC ternary alloy films with wafer scale, including Mo1-xWxS2, Mo1-xWxSe2and MoS2xSe2(1-x), which are accurately controllable in the elemental proportion (xis from 0 to 1). The band gap of the three 2D ternary alloy materials are continuously tuned for the whole range of metal and chalcogen compositions. The metal compositions are controlled by the as-deposited thickness. Raman, photoluminescence, elemental maps and TEM show the high spatial homogeneity in the compositions and optical properties across the whole wafer. The band gap can be continuously tuned from 1.86 to 1.99 eV for Mo1-xWxS2, 1.56 to 1.65 eV for Mo1-xWxSe2, 1.56 to 1.86 eV for MoS2xSe2(1-x). Electrical transport measurements indicate that Mo1-xWxS2and MoS2xSe2(1-x)monolayers shown-type semiconductor behaviors, and the carrier types of Mo1-xWxSe2can be tuned asn-type, bipolar andp-type. Moreover, this control process can be easily generalized to other 2D alloy films, even to quaternary or multi-element alloy materials. Our study presents a promising route for the preparation of large-scale homogeneous monolayer TMDC alloys and the application for future functional devices.

Tuning the morphology of 2D transition metal chalcogenides via oxidizing conditions.

Two-dimensional transition metal chalcogenides (TMCs) are emerging as an intriguing platform to realize nascent properties in condensed matter physics, materials science and device engineering. Controllable growing of TMCs becomes increasingly important, especially for the layer number, doping, and morphology. Here, we successfully tune the morphology of MoS2, MoSe2, WS2and WSe2, from homogenous films to individual single crystalline grains only via changing the oxidizing growth conditions. The oxidization degrees are determined by the oxygen that adsorbed on substrates and the oxygen concentrations in reaction gas together. We find the homogenous films are easily formed under the reductive conditions, triangular grains prefer the weak oxidizing conditions, and medium oxidizing conditions bring in dendritic grains with higher oxygen doping and inhomogenous photoluminescence intensities from edge to interior regions shown in the dendritic grains. These growth rules under different oxidizing conditions are easily generalized to other TMCs, which also show potential for growing specific TMCs with designed oxygen doping levels.

Semisupervised Feature Selection With Sparse Discriminative Least Squares Regression.

In big data time, selecting informative features has become an urgent need. However, due to the huge cost of obtaining enough labeled data for supervised tasks, researchers have turned their attention to semisupervised learning, which exploits both labeled and unlabeled data. In this article, we propose a sparse discriminative semisupervised feature selection (SDSSFS) method. In this method, the ϵ -dragging technique for the supervised task is extended to the semisupervised task, which is used to enlarge the distance between classes in order to obtain a discriminative solution. The flexible l2,p norm is implicitly used as regularization in the new model. Therefore, we can obtain a more sparse solution by setting smaller p . An iterative method is proposed to simultaneously learn the regression coefficients and ϵ -dragging matrix and predicting the unknown class labels. Experimental results on ten real-world datasets show the superiority of our proposed method.

Proton-assisted growth of ultra-flat graphene films.

Graphene films grown by chemical vapour deposition have unusual physical and chemical properties that offer promise for applications such as flexible electronics and high-frequency transistors1-10. However, wrinkles invariably form during growth because of the strong coupling to the substrate, and these limit the large-scale homogeneity of the film1-4,11,12. Here we develop a proton-assisted method of chemical vapour deposition to grow ultra-flat graphene films that are wrinkle-free. Our method of proton penetration13-17 and recombination to form hydrogen can also reduce the wrinkles formed during traditional chemical vapour deposition of graphene. Some of the wrinkles disappear entirely, owing to the decoupling of van der Waals interactions and possibly an increase in distance from the growth surface. The electronic band structure of the as-grown graphene films shows a V-shaped Dirac cone and a linear dispersion relation within the atomic plane or across an atomic step, confirming the decoupling from the substrate. The ultra-flat nature of the graphene films ensures that their surfaces are easy to clean after a wet transfer process. A robust quantum Hall effect appears even at room temperature in a device with a linewidth of 100 micrometres. Graphene films grown by proton-assisted chemical vapour deposition should largely retain their intrinsic performance, and our method should be easily generalizable to other nanomaterials for strain and doping engineering.

Local Adaptive Projection Framework for Feature Selection of Labeled and Unlabeled Data.

Most feature selection methods first compute a similarity matrix by assigning a fixed value to pairs of objects in the whole data or to pairs of objects in a class or by computing the similarity between two objects from the original data. The similarity matrix is fixed as a constant in the subsequent feature selection process. However, the similarities computed from the original data may be unreliable, because they are affected by noise features. Moreover, the local structure within classes cannot be recovered if the similarities between the pairs of objects in a class are equal. In this paper, we propose a novel local adaptive projection (LAP) framework. Instead of computing fixed similarities before performing feature selection, LAP simultaneously learns an adaptive similarity matrix and a projection matrix with an iterative method. In each iteration, is computed from the projected distance with the learned and W is computed with the learned . Therefore, LAP can learn better projection matrix by weakening the effect of noise features with the adaptive similarity matrix. A supervised feature selection with LAP (SLAP) method and an unsupervised feature selection with LAP (ULAP) method are proposed. Experimental results on eight data sets show the superiority of SLAP compared with seven supervised feature selection methods and the superiority of ULAP compared with five unsupervised feature selection methods.

Enhancing the Strength of Graphene by a Denser Grain Boundary.

From a device application point of view, the extreme mechanical strength of graphene is highly desirable. However, the unavoidable polycrystalline nature of graphene films produced by chemical vapor deposition (CVD) leads to significant fluctuations in mechanical properties. Although the effects of atomic defects or grain boundaries (GBs) on mechanical strength have been widely studied and some modifications have been applied to enhance the stiffness of graphene, the problems of fragility as well as significantly reduced breaking strength arise. Here we report a systematic study on the effect of elastic modulus and breaking strength of CVD-derived graphene films with a controlled density and distribution of GBs. We find that graphene films become much stronger by hugely increasing the density of GBs without triple junctions (TJs) formed inside, in analogy to the two-dimensional (2D) plum pudding structures. The comprehensive performance with a 2D Young's modulus of 436 N/m (∼1.3 TPa) and 2D breaking strength of 43 N/m (∼128 GPa) can be achieved with the average grain size of 20 nm. Moreover, the existence of TJs will slightly reduce the strength in these GB structures. Due to defect types, the graphene films will show various tearing behaviors after indentation. All these mechanical studies of GBs provide a guideline to obtain the optimal performance of 2D materials through GB structure engineering.