RESUMEN
We build new material descriptors to predict the band gap and the work function of 2D materials by tree-based machine-learning models. The descriptor's construction is based on vectorizing property matrices and on empirical property function, leading to mixing features that require low-resource computations. Combined with database-based features, the mixing features significantly improve the training and prediction of the models. We find R[Formula: see text] greater than 0.9 and mean absolute errors (MAE) smaller than 0.23 eV both for the training and prediction. The highest R[Formula: see text] of 0.95, 0.98 and the smallest MAE of 0.16 eV and 0.10 eV were obtained by using extreme gradient boosting for the bandgap and work-function predictions, respectively. These metrics were greatly improved as compared to those of database features-based predictions. We also find that the hybrid features slightly reduce the overfitting despite a small scale of the dataset. The relevance of the descriptor-based method was assessed by predicting and comparing the electronic properties of several 2D materials belonging to new classes (oxides, nitrides, carbides) with those of conventional computations. Our work provides a guideline to efficiently engineer descriptors by using vectorized property matrices and hybrid features for predicting 2D materials properties via ensemble models.
RESUMEN
The search for high-quality transition metal dichalcogenides mono- and multi-layers grown on large areas is still a very active field of investigation. Here, we use molecular beam epitaxy to grow WSe2 on 15 × 15 mm large mica in the van der Waals regime. By screening one-step growth conditions, we find that very high temperature (>900 °C) and very low deposition rate (<0.15 Å min-1) are necessary to obtain high quality WSe2 films. The domain size can be as large as 1 µm and the in-plane rotational misorientation of 1.25°. The WSe2 monolayer is also robust against air exposure, can be easily transferred over 1 cm2 on SiN/SiO2 and exhibits strong photoluminescence signal. Moreover, by combining grazing incidence x-ray diffraction and transmission electron microscopy, we could detect the presence of few misoriented grains. A two-dimensional model based on atomic coincidences between the WSe2 and mica crystals allows us to explain the formation of these misoriented grains and gives insight to achieve highly crystalline WSe2.
RESUMEN
The Hall effect can be extended by inducing a temperature gradient in lieu of electric field that is known as the Nernst (-Ettingshausen) effect. The recently discovered spin Nernst effect in heavy metals continues to enrich the picture of Nernst effect-related phenomena. However, the collection would not be complete without mentioning the valley degree of freedom benchmarked by the valley Hall effect. Here we show the experimental evidence of its missing counterpart, the valley Nernst effect. Using millimeter-sized WSe[Formula: see text] mono-multi-layers and the ferromagnetic resonance-spin pumping technique, we are able to apply a temperature gradient by off-centering the sample in the radio frequency cavity and address a single valley through spin-valley coupling. The combination of a temperature gradient and the valley polarization leads to the valley Nernst effect in WSe[Formula: see text] that we detect electrically at room temperature. The valley Nernst coefficient is in good agreement with the predicted value.
RESUMEN
In this work, we study growth and migration of atomic defects in MoSe2 on graphene using multiple advanced transmission electron microscopy techniques to explore defect behavior in vdW heterostructures. A MoSe2/graphene vdW heterostructure is prepared by a direct growth of both monolayers, thereby attaining an ideal vdW interface between the monolayers. We investigate the intrinsic defects (inversion domains and grain boundaries) in synthesized MoSe2, their evolution amid growth processing steps, and their influence on the formation and movement of extrinsic defects. Electron diffraction identifies a preferential interlayer orientation of 2° between MoSe2 and graphene, which is caused by the presence of intrinsic IBD defects. Extrinsic defects (point and line defects) are generated by in situ electron irradiation in the MoSe2 layer. Our results shed light on how to independently modify the MoSe2 atomic structure in vdW heterostructures for potential utilization in device processing.
RESUMEN
Van der Waals heterojunctions composed of graphene and transition metal dichalcogenides have gain much attention because of the possibility to control and tailor band structure, promising applications in two-dimensional optoelectronics and electronics. In this report, we characterized the van der Waals heterojunction MoSe2/few-layer graphene with a high-quality interface using cutting-edge surface techniques scaling from atomic to microscopic range. These surface analyses gave us a complete picture of the atomic structure and electronic properties of the heterojunction. In particular, we found two important results: the commensurability between the MoSe2 and few-layer graphene lattices and a band-gap opening in the few-layer graphene. The band gap is as large as 250 meV, and we ascribed it to an interface charge transfer that results in an electronic depletion in the few-layer graphene. This conclusion is well supported by electron spectroscopy data and density functional theory calculations. The commensurability between the MoSe2 and graphene lattices as well as the band-gap opening clearly show that the interlayer interaction goes beyond the simple van der Waals interaction. Hence, stacking two-dimensional materials in van der Waals heterojunctions enables us to tailor the atomic and electronic properties of individual layers. It also permits the introduction of a band gap in few-layer graphene by interface charge transfer.
RESUMEN
Heteroepitaxial growth of Ge nanowires was carried out on Si(111) substrates by MBE. Au seeds were used as precursor for the VLS growth of the nanowires. Even if the Au droplets do not act as catalyst for the dissociation of gas, they are local preferential areas where the energetic barrier of Ge nucleation is lowered compare to the remaining non activated surface. Two sets of Au seeds were used as precursors for the VLS process. The first set have an average diameter of 125 nm and the second of 25 nm. In-situ RHEED monitoring showed a Au wetting layer between these seeds before the nanowires growth as well as at the end of the Ge nanowires growth. It means that the wetting layer acted as a surfactant from the Si(111) surface to the Ge grown layer between the nanowires. Analysis of SEM images brought the fact that the diffusion of gold from the droplets on the surface and the sidewalls of the nanowires via the Ostwald ripening is a key parameter of the growth of the nanowires.