Search

Deep learning (DL) is an emerging analysis tool across the sciences and engineering. Encouraged by the successes of DL in revealing quantitative trends in massive imaging data, we applied this approach to nanoscale deeply subdiffractional images of propagating polaritonic waves in complex materials. Utilizing the convolutional neural network (CNN), we developed a practical protocol for the rapid regression of images that quantifies the wavelength and the quality factor of polaritonic waves. Using simulated near-field images as training data, the CNN can be made to simultaneously extract polaritonic characteristics and material parameters in a time scale that is at least 3 orders of magnitude faster than common fitting/processing procedures. The CNN-based analysis was validated by examining the experimental near-field images of charge-transfer plasmon polaritons at graphene/α-RuCl

We present a detailed study of gaseous Br2 adsorption and charge transfer on graphene, combining in situ Raman spectroscopy and density functional theory (DFT). When graphene is encapsulated by hexagonal boron nitride (h-BN) layers on both sides, in a h-BN/graphene/h-BN sandwich structure, it is protected from doping by strongly oxidizing Br2. Graphene supported on only one side by h-BN shows strong hole doping by adsorbed Br2. Using Raman spectroscopy, we determine the graphene charge density as a function of pressure. DFT calculations reveal the variation in charge transfer per adsorbed molecule as a function of coverage. The molecular adsorption isotherm (coverage versus pressure) is obtained by combining Raman spectra with DFT calculations. The Fowler-Guggenheim isotherm fits better than the Langmuir isotherm. The fitting yields the adsorption equilibrium constant (∼0.31 Torr(-1)) and repulsive lateral interaction (∼20 meV) between adsorbed Br2 molecules. The Br2 molecule binding energy is ∼0.35 eV. We estimate that at monolayer coverage each Br2 molecule accepts 0.09 e- from single-layer graphene. If graphene is supported on SiO2 instead of h-BN, a threshold pressure is observed for diffusion of Br2 along the (somewhat rough) SiO2/graphene interface. At high pressure, graphene supported on SiO2 is doped by adsorbed Br2 on both sides.

Using Raman spectroscopy, we study the environmental sensitivity of mechanically exfoliated and electrically floating single-layer graphene transferred onto a hexagonal boron nitride (h-BN) substrate, in comparison with graphene deposited on a SiO(2) substrate. In order to understand and isolate the substrate effect on graphene electrical properties, we model and correct for Raman optical interference in the substrates. As-deposited and unannealed graphene shows a large I(2D)/I(G) ratio on both substrates, indicating extremely high quality, close to that of graphene suspended in vacuum. Thermal annealing strongly activates subsequent environmental sensitivity on the SiO(2) substrate; such activation is reduced but not eliminated on the h-BN substrate. In contrast, in a h-BN/graphene/h-BN sandwich structure, with graphene protected on both sides, graphene remains pristine despite thermal processing. Raman data provide a deeper understanding of the previously observed improved graphene electrical conductivity on h-BN substrates. In the sandwich structure, the graphene 2D Raman feature has a higher frequency and narrower line width than in pristine suspended graphene, implying that the local h-BN environment modestly yet measurably changes graphene electron and phonon dispersions.