By computing the double-resonant Raman scattering cross section completely from first principles and including the electron-electron interaction at the GW level, we unravel the dominant contributions for the double-resonant 2D mode in bilayer graphene. We show that, in contrast to previous works, the so-called inner processes are dominant and that the 2D-mode line shape is described by three dominant resonances around the K point. We show that the splitting of the transversal optical (TO) phonon branch in the Γ-K direction, as large as 12 cm(-1) in the GW approximation, is of great importance for a thorough description of the 2D-mode line shape. Finally, we present a method to extract the TO phonon splitting and the splitting of the electronic bands from experimental data.
The methods to calculate electron-phonon coupling from first principles and their present role in the simulation of thermoelectric properties are reviewed. The various kinds of electron-phonon scattering processes are exposed in a pedagogical way, as well as the different experiments by which they manifest themselves. The distinction between the deformation potential theory and the ab initio theoretical framework to compute electron-phonon scattering rates is explained, and the results obtained during the last several years are reviewed. The link between these calculated scattering rates and the experimentally measured quantities, such as the broadening of spectral lines or the determination of relaxation times, is clarified. The key role of data produced ab initio, as imput parameters in the simulation of thermoelectric properties of materials, is illustrated. Finally, challenges and open problems in the calculation of the electron-phonon coupling without adjustable parameters are critically examined.
The charge carrier dynamics of photoexcited bismuth generates a Drude response that evolves over time. Our data show that the plasma frequency of bismuth displays an initial increase and a subsequent decay. We have performed ab initio calculations on bulk bismuth within the density functional theory and show that this peculiar behavior is due to local extrema in the valence and conduction bands. It follows that most of the carriers first accumulate in these extrema and reach the Fermi level only 0.6 ps after the photoexcitation.
Motivated by recent experimental studies on single molecular magnets grafted on graphene and single-walled carbon nanotubes, we investigate the structural, electronic, and magnetic properties of an iron-based magnetic molecule grafted on a graphene sheet using ab initio calculations. In particular, the induced charge transfer and magnetization are described in terms of the coupling between the molecule and the graphene orbitals. This interaction and its effects on graphene electronic properties are determined and discussed in view of the potential utilization of graphene in spintronics.
Submitted for the MAR11 Meeting of The American Physical Society Ab initio calculation of double-resonant Raman spectra for bilayer graphene PAOLA GAVA, LSI Ecole Polytechnique, MATTEO CALANDRA, MICHELE LAZZERI, FRANCESCO MAURI, IMPMC Universite Paris 6 — The discovery that the application of an external electric field induces a band gap opening in bilayer graphene attracted a lot of interest on this system, due to important applications in nanoelectronics [1]. Raman spectroscopy is one of the most important experimental techniques for the characterisation of carbon based materials, providing informations on carriers concentration [2], disorder [3], number of layers on multi-layers graphene systems [4], and phonon properties. Most of the theoretical studies on multi-layers graphene are performed using a Tight Binding (TB) model, and full calculation of Raman matrix elements to obtain frequencies, intensities and linewidths of Raman bands has not been performed up to now. The developpement of a fully ab initio theoretical tool to compute Raman spectra is therefore higly desirable and particularly relevant for systems where a simple TB parametrization of the electronic structure and of the electron-phonon interaction is not available. In this talk I will discuss a recently developped methodology to compute fully ab initio double-resonant Raman spectra and I will present results for bilayer graphene. [1] Ohta et al, Science 313, 951 (2006), [2] Malard et al, PRL 101, 257410 (2008), [3] Lucchese et al, Carbon 48, 1592 (2010), [4] Ferrari et al, PRL 97, 187401 (2006) Paola Gava LSI Ecole Polytechnique Date submitted: 22 Nov 2010 Electronic form version 1.4
We present Kelvin probe force microscopy measurements of single-and few-layer graphene resting on SiO2 substrates. We compare the layer thickness dependency of the measured surface potential with ab initio density functional theory calculations of the work function for substrate-doped graphene. The ab initio calculations show that the work function of single-and bilayer graphene is mainly given by a variation of the Fermi energy with respect to the Dirac point energy as a function of doping, and that electrostatic interlayer screening only becomes relevant for thicker multilayer graphene. From the Raman G-line shift and the comparison of the Kelvin probe data with the ab initio calculations, we independently find an interlayer screening length in the order of four to five layers. Furthermore, we describe in-plane variations of the work function, which can be attributed to partial screening of charge impurities in the substrate, and result in a nonuniform charge density in single-layer graphene.
We present first-principles calculations of the effects of magnetism on the ballisticconductance of a model Pd nanocontact, made of a short Pd monatomicstretched chain placed between two Pd leads, simulated by semi-infinite (100) slabs.The stretching makes the suspended Pd chain generally ferromagnetic. The spin-resolved ballisticconductance, calculated according to the Landauer-Büttiker formula isfound to be 0.85G0 for the spin-up and 1.15G0 for the spin-down electrons(G0 = 2e2/h is the conductance quantum). The total conductance ~2G0 islower, but still relatively close to that of the nonmagnetic Pd nanocontact with the same geometry,calculated to be 2.3G0. To illustrate how magnetism and conductance depend onstructural details, we change the three atom chain docking from the top to a hollowsurface site, where at the same stress the Pd contact is nonmagnetic and the conductancedecreases to 1.8G0. Overall we find these calculated ballistic conductance valuesof very similar magnitude to the first histogram peak in the experimental data obtained for Pdat low temperature in mechanically controllable break junctions. We conclude that the15% conductance changes caused by the onset or the demise of local magnetism,similar in magnitude to geometry-related conductance changes, are probably too small tobe used as a diagnostic for the presence or absence of nanocontact magnetism.
The Raman shift, broadening, and relative Raman intensities of bilayer graphene are computed as functions of the electron concentration. We include dynamic effects for the phonon frequencies and we consider the gap induced in the band structure of bilayer graphene by an external electric field. We show that from the analysis of the Raman spectra of gated bilayer graphene it is possible to quantitatively identify the amount of charges coming from the atmosphere and from the substrate. These findings suggest that Raman spectroscopy of bilayer graphene can be used to characterize the electrostatic environment of few-layers graphene.
The Raman shift, broadening, and relative Raman intensities of bilayer graphene are computed as functions of the electron concentration. We include dynamic effects for the phonon frequencies and we consider the gap induced in the band structure of bilayer graphene by an external electric field. We show that from the analysis of the Raman spectra of gated bilayer graphene it is possible to quantitatively identify the amount of charges coming from the atmosphere and from the substrate. These findings suggest that Raman spectroscopy of bilayer graphene can be used to characterize the electrostatic environment of few-layers graphene.
The electronic properties of doped bilayer graphene in presence of bottom and top gates have been studied and characterized by means of density-functional theory (DFT) calculations. Varying independently the bottom and top gates it is possible to control separately the total doping charge on the sample and the average external electric field acting on the bilayer. We show that, at fixed doping level, the band gap at the K point in the Brillouin zone depends linearly on the average electric field, whereas the corresponding proportionality coefficient has a nonmonotonic dependence on doping. We find that the DFT-calculated band gap at K, for small doping levels, is roughly half of the band gap obtained with standard tight-binding (TB) approach. We show that this discrepancy arises from an underestimate, in the TB model, of the screening of the system to the external electric field. In particular, on the basis of our DFT results we observe that, when bilayer graphene is in presence of an external electric field, both interlayer and intralayer screenings occur. Only the interlayer screening is included in TB calculations, while both screenings are fundamental for the description of the band-gap opening. We finally provide a general scheme to obtain the full band structure of gated bilayer graphene for an arbitrary value of the external electric field and of doping.
The activated adsorption of ethylene on atomic-oxygen-covered Ag(100) surface and on an open-type step edge thereon was studied using density-functional-theory. On perfect Ag(100), such adsorption results in the formation of an oxametallacycle (OMC), with an activation energy slightly larger than 0.3 eV. We find that this activation energy is only weakly dependent on the coverage of on-surface oxygen (for Theta <= 1/2 ML), whereas the OMC-surface interaction is substantially reduced at high oxygen coverage. Three types of OMCs have been identified on the (100) surface, which display similar. stability, and the transformation between them is facile with activation energies below 0.1 eV. We find that the presence of subsurface oxygen reduces the activation energy for ONIC formation and substantially increases the OMC-surface interaction. The reactivity of the step edge toward the OMC formation strongly depends on the local coverage of oxygen. Our calculations indicate that the relative stability of the OMC intermediate in ethylene epoxidation reaction is strongly affected by the coverage and configuration of chemisorbed oxygen.
The adsorption of molecular and atomic chlorine on perfect Ag(111) surface has been studied and characterized by means of extensive density-functional-theory calculations. For the molecular adsorption, we find that the dissociation of Cl-2 proceeds with an almost vanishing barrier. As for the adsorption of atomic Cl, on-surface, subsurface, and substitutional adsorptions are considered as a function of the coverage. At coverage lower than 1/2 ML, the on-surface adsorption displays the most exothermic chemisorption energies, whereas the mixed on-surface+subsurface and on-surface+substitutional adsorption modes become competitive with pure on-surface adsorption at about 1/2 ML of coverage and at higher coverages even preferred. The analysis of the adsorption free energy as a function of chlorine chemical potential reveals that the on-surface (root 3x root 3)R30 degrees adsorption phase is thermodynamically the most stable over a very broad range of Cl chemical potential. The mixed adsorption modes become thermodynamically more stable at high coverage for values of the Cl chemical potential that are substantially larger than those needed to form silver chloride. This finding seems to indicate that the formation of mixed adsorption phases, if they would ever occur, cannot be due to thermodynamic equilibrium but can only result from kinetic effects. We also find that the presence of open surface steps does not stabilize the subsurface Cl adsorption at low coverage. However due to the stronger Cl-surface interaction near steps, the mixed on-surface+subsurface adsorption on Ag(210) at high coverage becomes thermodynamically the most stable phase at Cl chemical potential close to that needed for the formation of bulk AgCl.
The selectivity of a catalyst in ethylene epoxidation reaction was addressed using quantum mechanical computer simulations. We found that the catalyst's selectivity in the reaction of oxametallacycle to form ethylene epoxide (EO) rather than the competing acetaldehyde (Ac) is determined in part by the differential bonding affinity of the catalyst toward the O and C atoms of the oxametallacycle. This interplay between O– and C–metal bond strength is due to the different structures of the two transition states. Based on this finding, we introduce a new indicator that determines the difference between the EO and Ac activation energies in the oxametallacycle reaction remarkably well for a number of different materials.
A simple direct mechanism for methane-to-methanol conversion has been investigated by first principles on a series of oxygen-precovered transition-metal surfaces. Energy barriers and reaction paths have been determined for three competing elementary processes by the nudged elastic band algorithm. Indicators of reactivity toward each elementary step have been identified, providing significant insight into a rational search for a suitable catalyst. The effect of chemical environment, local geometry, strain, and coadsorption have been addressed, and general guidelines have been identified. On the basis of this analysis, we suggest that upon suitable conditions O-dosed Ag surfaces could display considerable reactivity toward direct methane-to-methanol conversion.