Search

The effect of electronic Coulomb interaction on the phase formation of fcc Ce lattice is investigated by full ab initio calculations without adjustable Coulomb $U$ and $J$ parameters using the Gutzwiller wavefunction-based correlation matrix renormalization theory (CMRT). Its total energy and pressure as a function of volume agree reasonably well with existing $\text{DFT}+\text{Gutzwiller}$ calculations and experiments, indicating correct capture of electronic correlation and screening effects within the CMRT formalism. A stable phase is found in line with the experimental $\ensuremath{\alpha}$-Ce phase, and a lurking phase is identified supposedly linked with the experimental $\ensuremath{\gamma}$-Ce phase. A criterion based on the local $4f$ electron charge fluctuation is introduced to confirm the distinct electronic correlation natures of both phases.

Electron-phonon interactions are fundamentally important physical processes responsible for many key discoveries in condensed matter physics and material sciences. Herein, by exploiting the scattering-type scanning near-field optical microscope (s-SNOM) excited with a femtosecond infrared (IR) laser, we explored the strong coupling between IR phonons in few-layer graphene (FLG) with ultrahot electrons, which are heated up by the intense laser field enhanced by the s-SNOM tip. More specifically, we found that the intensity of the phonon resonance can be tuned systematically by varying the laser power that controls the electron temperature. Furthermore, the high spatial resolution of s-SNOM allows us to map the local phonon characteristics at sharp boundaries and nanostructures. Our findings offer insights into the intriguing physics behind the electron-phonon interactions in nonequilibrium conditions and open a pathway for manipulating phonons with optical means.

We studied the electronic structure of ${\mathrm{PtPb}}_{4}$ using laser angle-resolved photoemission spectroscopy (ARPES) and density functional theory (DFT) calculations. This material is closely related to ${\mathrm{PtSn}}_{4}$, which exhibits exotic topological properties such as Dirac node arcs. The Fermi surface (FS) of ${\mathrm{PtPb}}_{4}$ consists of two electron pockets at the center of the Brillouin zone (BZ) and several hole pockets around the zone boundaries. Our ARPES data reveal significant Rashba splitting at the $\mathrm{\ensuremath{\Gamma}}$ point, in agreement with DFT calculations. The presence of Rashba splitting may render this material of potential interest for spintronic applications.

Manipulating the electronic properties of graphene has been a subject of great interest since it can aid material design to extend the applications of graphene to many different areas. In this paper, we systematically investigate the effect of lead (Pb) intercalation on the structural and electronic properties of epitaxial graphene on the SiC(0001) substrate. We show that the band structure of Pb-intercalated few-layer graphene can be effectively tuned through changing intercalation conditions, such as coverage, location of Pb, and the initial number of graphene layers. Lead intercalation at the interface between the buffer layer (BL) and the SiC substrate decouples the BL from the substrate and transforms the BL into a $p$-doped graphene layer. We also show that Pb atoms tend to donate electrons to neighboring layers, leading to an $n$-doping graphene layer and a small gap in the Dirac cone under a sufficiently high Pb coverage. This paper provides useful guidance for manipulating the electronic properties of graphene layers on the SiC substrate.

We analyze several methods of obtaining the accurate relativistic total energy (TE) variations using traditional perturbation theories (PTs) and proposed coupling constant integration (CCI) methods. For this purpose, we perform benchmark calculations within the density functional theory taking the spin-orbit coupling (SOC) and its derivative as a perturbation. The TE change due to SOC addition obtained from both PTs and CCI is shown to reach the accuracy of fully self-consistent TE calculations. Similar accuracy is also obtained even for the magnetocrystalline anisotropy energy (MAE). The real advantage of the proposed methods is to use PTs and CCI methods in those electronic structure methods where accurate total energies currently cannot be obtained with required accuracy. Correspondingly, we demonstrate the applicability of suggested methods for calculations of MAE in different magnetic materials using a dynamic mean-field method. All suggested PTs and CCI methods also provide convenient site, orbital, and spin decompositions of the TE variation, creating a powerful way to analyze microscopic physics in strongly correlated materials.

We propose a scheme of decomposition of the total relativistic energy in solids to intra- and interatomic contributions. The method is based on a site variation of such fundamental constant as the speed of light. As a practical illustration of the method, we tested such decomposition in the case of a spin-orbit interaction variation for the decomposition of the magnetic anisotropy energy (MAE) in CoPt. We further studied the ${\ensuremath{\alpha}}^{\ensuremath{''}}\text{\ensuremath{-}}{\mathrm{Fe}}_{16}{\mathrm{N}}_{2}$ magnet doped by Bi, Sb, Co, and Pt atoms. It was found that the addition of Pt atoms can enhance the MAE by as much as five times while Bi and Sb substitutions double the total MAE. Using the proposed technique, we demonstrate the spatial distribution of these enhancements. Our studies also suggest that Sb, Pt, and Co substitutions could be synthesized by experiments.

The correlation between structure and glass formability in glassy systems is a long-standing puzzle. To solve this puzzle, many descriptors based on the short-range order (SRO) have been proposed. Here we show that the SRO, however, offers little help in explaining the glass formability and stability; instead it is the formation of medium-range order that stabilizes the glass against crystallization by suppressing the atomic rearrangement and compositional change. Our results provide a perspective for understanding the correlation between structure and stability in glasses.

We introduce Gutzwiller conjugate gradient minimization (GCGM) theory, an ab initio quantum many-body theory for computing the ground-state properties of infinite systems. GCGM uses the Gutzwiller wave function but does not use the commonly adopted Gutzwiller approximation (GA), which is a major source of inaccuracy. Instead, the theory uses an approximation that is based on the occupation probability of the on-site configurations, rather than approximations that decouple the site-site correlations as used in the GA. We test the theory in the one-dimensional and two-dimensional Hubbard models at various electron densities and find that GCGM reproduces energies and double occupancies in reasonable agreement with benchmark data at a very small computational cost.

The development of long-range ferromagnetic (FM) order in dilute magnetic topological insulators can induce dissipationless electronic surface transport via the quantum anomalous Hall effect. We measure the magnetic excitations in a prototypical magnetic topological crystalline insulator, ${\mathrm{Sn}}_{0.95}{\mathrm{Mn}}_{0.05}\mathrm{Te}$, using inelastic neutron scattering. Neutron diffraction and magnetization data indicate that our ${\mathrm{Sn}}_{0.95}{\mathrm{Mn}}_{0.05}\mathrm{Te}$ sample has no FM long-range order above a temperature of 2 K. However, we observe slow, collective FM fluctuations $(l70\phantom{\rule{0.28em}{0ex}}\ensuremath{\mu}\mathrm{eV})$, indicating proximity to FM order. We also find a series of sharp peaks originating from local excitations of antiferromagnetically (AF) coupled and isolated Mn-Mn dimers with ${J}_{\mathrm{AF}}=460\phantom{\rule{4pt}{0ex}}\ensuremath{\mu}\mathrm{eV}$. The simultaneous presence of collective and localized components in the magnetic spectra highlight different roles for substituted Mn ions, with competition between FM order and the formation of AF-coupled Mn-Mn dimers.

Interatomic potentials based on neural-network machine learning (ML) approach to address the long-standing challenge of accuracy versus efficiency in molecular-dynamics simulations have recently attracted a great deal of interest. Here, utilizing Pd-Si system as a prototype, we extend the development of neural-network ML potentials to compounds exhibiting various types of bonding characteristics. The ML potential is trained by fitting to the energies and forces of both liquid and crystal structures first-principles calculations based on density-functional theory (DFT). We show that the generated ML potential captures the structural features and motifs in $\mathrm{P}{\mathrm{d}}_{82}\mathrm{S}{\mathrm{i}}_{18}$ and $\mathrm{P}{\mathrm{d}}_{75}\mathrm{S}{\mathrm{i}}_{25}$ liquids more accurately than the existing interatomic potential based on embedded-atom method (EAM). The ML potential also describes the solid-liquid interface of these systems very well. Moreover, while the existing EAM potential fails to describe the relative energies of various crystalline structures and predict wrong ground-state structures at $\mathrm{P}{\mathrm{d}}_{3}\mathrm{Si}$ and $\mathrm{P}{\mathrm{d}}_{9}\mathrm{S}{\mathrm{i}}_{2}$ composition, the developed ML potential predicts correctly the ground-state structures from genetic algorithm search. The efficient ML potential with DFT accuracy from our study will provide a promising scheme for accurate atomistic simulations of structures and dynamics of complex Pd-Si system.

Superconductors with the A15 structure are prototypical type-II s-wave superconductors which have generated considerable interest in early superconducting material history. However, the topological nature of the electronic structure remains unnoticed so far. Here, using first-principles band structure calculations based on density functional theory, we show that the A15 superconductors (Ta$_3$Sb, Ta$_3$Sn, and Ta$_3$Pb) have nontrivial band topology in the bulk electronic band structures, leading to the formation of topological surface states near the Fermi energy. Due to the bulk superconductivity, the proximity effect in the topological surface states would induce topological superconductivity even without heterostructure of a topological insulator and an s-wave superconductor. Our results indicate that the A15 superconductors are promising candidates for the realization of the topological superconductivity and the Majorana fermion., 11 pages, 4 figures

The onset of ferromagnetism in cobalt-tin alloys is investigated experimentally and theoretically. The $\mathrm{C}{\mathrm{o}}_{1+x}\mathrm{Sn}$ alloys were prepared by rapid quenching from the melt and form a modified hexagonal NiAs-type crystal structure for $0.45\ensuremath{\le}x\ensuremath{\le}1$. The magnetic behavior is described analytically and by density-functional theory using supercells and the coherent-potential approximation. The excess of Co concentration $x$, which enters the interstitial $2d$ sites in the hypothetical NiAs-ordered parent alloy CoSn, yields a Griffiths-like phase and, above a quantum critical point $({x}_{c}\ensuremath{\approx}0.65)$, a quantum phase transition to ferromagnetic order. Quantum critical exponents are determined on the paramagnetic and ferromagnetic sides of the transition and related to the nature of the magnetism in itinerant systems with different types of chemical disorder.

We report terahertz (THz) electrodynamics of a moderately clean A15 superconductor (SC) following ultrafast excitation to manipulate quasiparticle (QP) transport. In the Martensitic normal state, we observe a photo enhancement in the THz conductivity using optical pulses, while the opposite is observed for the THz pump. This demonstrates wavelength-selective nonthermal control of conductivity distinct from sample heating. The photo enhancement persists up to an additional critical temperature, above the SC one, from a competing electronic order. In the SC state, the fluence dependence of pair-breaking kinetics together with an analytic model provides an implication for a ``one photon to one Cooper pair'' nonresonant energy transfer during the 35-fs laser pulse; i.e., the fitted photon energy $\ensuremath{\hbar}{\ensuremath{\omega}}_{}$ absorption to create QPs set by $2{\mathrm{\ensuremath{\Delta}}}_{SC}/\ensuremath{\hbar}\ensuremath{\omega}=0.33%$. This is more than one order of magnitude smaller than in previously studied BCS SCs, which we attribute to strong electron-phonon coupling and possible influence of phonon condensation.

In this paper, we present a coupled experimental/theoretical investigation of pressure effect on the ferromagnetism of LaCrGe3 and LaCrSb3 compounds. The magnetic, electronic, elastic and mechanical properties of LaCrGe3 and LaCrSb3 at ambient condition are studied by first-principles density functional theory calculations. The pressure dependences of the magnetic properties of LaCrGe3 and LaCrSb3 are also investigated. The ferromagnetism in LaCrGe3 is rather fragile with a ferro- to paramagnetic transition at a relatively small pressure (around 7 GPa from our calculations, and 2 GPa in experiments). The key parameter controlling the magnetic properties of LaCrGe3 is found to be the proximity of the Cr DOS to the Fermi surface, a proximity that is strongly correlated to the distance between Cr atoms along the c-axis, suggesting that there would be a simple way to suppress magnetism in systems with one dimensional arrangement of magnetic atoms. By contrast, the ferromagnetism in LaCrSb3 is not fragile. Our calculation results are consistent with our experimental results and demonstrate the feasibility of using first-principles calculations to aid experimental explorations in screening for materials with fragile magnetism.

The multimode Jahn-Teller (JT) effect in a bulk system of a neutral nitrogen-vacancy ($\mathrm{N}{\mathrm{V}}^{0}$) center in diamond is investigated via first-principles density-functional-theory calculations and the intrinsic distortion path (IDP) method. The adiabatic potential energy surface of the electronic ground state of the $\mathrm{N}{\mathrm{V}}^{0}$ center is calculated based on the local spin-density approximation. Our calculations confirm the presence of the dynamic Jahn-Teller effect in the ground $^{2}E$ state of the $\mathrm{N}{\mathrm{V}}^{0}$ center. Within the harmonic approximation, the IDP method provides the reactive path of JT distortion from unstable high-symmetry geometry to stable low-symmetry energy minimum geometry, and it describes the active normal modes participating in the distortion. We find that there is more than one vibrational mode contributing to the distortion, and their contributions change along the IDP. Several vibrational modes with large contributions to JT distortion, especially those modes close to 44 meV, are clearly observed as the phonon sideband in photoluminescence spectra in a series of experiments, indicating that the dynamic Jahn-Teller effect plays an important role in the optical transition of the $\mathrm{N}{\mathrm{V}}^{0}$ center.

A series of iron arsenides (e.g., ${\text{CaRbFe}}_{4}{\text{As}}_{4}, {\text{SrCsFe}}_{4}{\text{As}}_{4}$) have been discovered recently, and have provoked a rise in superconductor searches in a different phase, known as the 1144 phase. For the presence of various chemical substitutions, it is believed that more 1144 compounds remain to be discovered. In this work, we perform general model analysis as well as scenario calculation on a basis of density functional theory to investigate phase stability in a variety of compounds. We predict that the 1144-type phase could be stabilized in ${\text{EuKFe}}_{4}{\text{As}}_{4}, {\text{EuRbFe}}_{4}{\text{As}}_{4}, {\text{EuCsFe}}_{4}{\text{As}}_{4}, {\text{CaCsFe}}_{4}{\text{P}}_{4}, {\text{SrCsFe}}_{4}{\text{P}}_{4}, {\text{BaCsFe}}_{4}{\text{P}}_{4}, {\text{InCaFe}}_{4}{\text{As}}_{4}, {\text{InSrFe}}_{4}{\text{As}}_{4}$, etc. Remarkably, it involves rare earths, trivalence elements (e.g., indium) and iron phosphides, which greatly expands the range of its existence and suggests a promising prospect for experimental synthesis. In addition, we find that the formation of many random doping compounds (e.g., ${\text{Ba}}_{0.5}{\text{Cs}}_{0.5}{\text{Fe}}_{2}{\text{As}}_{2}, {\text{Ba}}_{0.5}{\text{Rb}}_{0.5}{\text{Fe}}_{2}{\text{As}}_{2}$) is driven by entropy and could be annealed to a 1144-type phase. Eventually, we plot a phase diagram about two structural factors $\mathrm{\ensuremath{\Delta}}a$ and $\mathrm{\ensuremath{\Delta}}c$, giving a bird's-eye view of stability of various 1144 compounds.

Developing accurate and computationally efficient methods to calculate the electronic structure and total energy of correlated-electron materials has been a very challenging task in condensed matter physics and materials science. Recently, we have developed a correlation matrix renormalization (CMR) method which does not assume any empirical Coulomb interaction $U$ parameters and does not have double counting problems in the ground-state total energy calculation. The CMR method has been demonstrated to be accurate in describing both the bonding and bond breaking behaviors of molecules. In this study, we extend the CMR method to the treatment of electron correlations in periodic solid systems. Using a linear hydrogen chain as a benchmark system, we show that the results from the CMR method compare very well with those obtained recently by accurate quantum Monte Carlo (QMC) calculations. We also study the equation of states of three-dimensional crystalline phases of atomic hydrogen. We show that the results from the CMR method agree much better with the available QMC data in comparison with those from density functional theory and Hartree-Fock calculations.

Intriguing topological phases may appear in both insulating and semimetallic states. Topological insulators exhibit topologically nontrivial band inversion, while topological Dirac/Weyl semimetals show ``relativistic'' linear band crossings. Here, we report an unusual topological state of ${\mathrm{Pt}}_{3}\mathrm{Sn}$, where the two topological features appear simultaneously. Based on first-principles calculations, we show that ${\mathrm{Pt}}_{3}\mathrm{Sn}$ is a three-dimensional weak topological semimetal with topologically nontrivial band inversion between the valence and conduction bands, where the band structure also possesses type-II Dirac points at the boundary of two electron pockets. The formation of the Dirac points can be understood in terms of the representations of relevant symmetry groups and the compatibility relations. The topological surface states appear in accordance with the nontrivial bulk band topology. The unique coexistence of the two distinct topological features in ${\mathrm{Pt}}_{3}\mathrm{Sn}$ enlarges the material scope in topological physics, and is potentially useful for spintronics.

Co3Si was recently reported to exhibit remarkable magnetic properties in the nanoparticle form [Appl. Phys. Lett. 108, 152406 (2016)], yet better understanding of this material is to be promoted. Here we report a study on the crystal structures of Co3Si using adaptive genetic algorithm, and discuss its electronic and magnetic properties from first-principles calculations. Several competing phases of Co3Si have been revealed from our calculations. We show that the hexagonal Co3Si structure reported in experiments has lower energy in non-magnetic state than ferromagnetic state at zero temperature. The ferromagnetic state of the hexagonal structure is dynamically unstable with imaginary phonon modes and transforms to a new orthorhombic structure, which is confirmed by our structure searches to have the lowest energy for both Co3Si and Co3Ge. Magnetic properties of the experimental hexagonal structure and the lowest-energy structures obtained from our structure searches are investigated in detail.

Structures and magnetic properties of Fe16-xCoxN2 are studied using adaptive genetic algorithm and first-principles calculations. We show that substituting Fe by Co in Fe16N2 with Co/Fe ratio smaller than 1 can greatly improve the magnetic anisotropy of the material. The magnetocrystalline anisotropy energy from first-principles calculations reaches 3.18 MJ/m3 (245.6 {\mu}eV per metal atom) for Fe12Co4N2, much larger than that of Fe16N2 and is one of the largest among the reported rare-earth free magnets. From our systematic crystal structure searches, we show that there is a structure transition from tetragonal Fe16N2 to cubic Co16N2 in Fe16-xCoxN2 as the Co concentration increases, which can be well explained by electron counting analysis. Different magnetic properties between the Fe-rich (x 8) Fe16-xCoxN2 is closely related to the structural transition.

We employ a real-space pseudopotential method to compute the structural energies of a prototypical metal-semiconductor interface. Specifically, we examine a Pb(111) film overlaid on a Si(111) substrate as a function of the metal thickness. For each layer of Pb, we fully relax the atomic coordinates and determine the lowest-energy structure. Owing to the lattice mismatch between the Pb and Si crystal structures, we consider a large supercell containing up to 1505 atoms for the largest system. Systems of this size remain challenging for most current computational approaches and require algorithms specifically designed for highly parallel computational platforms. We examine the structural properties of the interface with respect to the thickness of the metal overlayer, e.g., the corrugation of the profile of the Pb overlayer. The combined influence of the Si substrate and quantum confinement results in a rich profile for a transition between a thin overlayer (less than a few monolayers), where the corrugation is strong, and the bulk region (more than a half-dozen layers), where the overlaid Pb film is atomically flat. This work proves the feasibility of handling systems with such a level of complexity.

Interaction between K adatoms on graphene is investigated by first-principles calculations based on density function theory and analytical analyses based on the $\mathbf{k}\ifmmode\cdot\else\textperiodcentered\fi{}\mathbf{p}$ perturbation theory. The calculation shows that there is a strong repulsion between K adatoms. The main origin of this strong repulsion is not from the dipole-dipole interaction as suggested for K adatoms on graphite surface, but comes from the screened Coulomb interaction. Potassium adatom on graphene donates its $s$ electron and becomes ${\mathrm{K}}^{+}$. The positively charged K adatom induces electron density oscillation on graphene which is responsible for the screened Coulomb repulsion between the K adatoms.

We observe that Cu clusters grow on surface terraces of graphite as a result of physical vapor deposition in ultrahigh vacuum. We show that the observation is incompatible with a variety of models incorporating homogeneous nucleation and calculations of atomic-scale energetics. An alternative explanation, ion-mediated heterogeneous nucleation, is proposed and validated, both with theory and experiment. This serves as a case study in identifying when and whether the simple, common observation of metal clusters on carbon-rich surfaces can be interpreted in terms of homogeneous nucleation. We describe a general approach for making system-specific and laboratory-specific predictions.

Adsorption of Fe atom and ${\mathrm{Fe}}_{n}$ ($n=2\ensuremath{-}7$) clusters on graphene was studied by first-principles calculation and chemical bonding analysis. Various adsorption configurations, i.e., Fe clusters placed at different lateral positions and with different orientations with respect to the graphene lattice, have been optimized to locate the ground-state geometry of the Fe clusters on graphene. The calculation shows that Fe on graphene prefers a three-dimensional growth mode with a critical size of two atoms. The calculation also shows that ${\mathrm{Fe}}_{n}$ clusters on graphene exhibit ferromagnetic order and large magnetic moment. However, due to charge redistribution caused by the interaction with graphene, some ${\mathrm{Fe}}_{n}$ clusters on graphene have smaller magnetic moments compared to the corresponding free-standing clusters. Adsorption of ${\mathrm{Fe}}_{n}$ clusters also induces magnetic moments in graphene, and the induced magnetic moment on each carbon atom in graphene is correlated with the distortion of the graphene lattice.

We generalized the commonly used Gutzwiller approximation for calculating the electronic structure and total energy of strongly correlated electron systems. In our method, the evaluation of one-body and two-body density matrix elements of the Hamiltonian is simplified using a renormalization approximation to achieve better scaling of the computational effort as a function of system size. To achieve a clear presentation of the concept and methodology, we describe the detailed formalism for a finite hydrogen system with minimal basis set. We applied the correlation matrix renormalization approximation approach to a H${}_{2}$ dimer and H${}_{8}$ cubic fragment with minimal basis sets, as well as a H${}_{2}$ molecule with a large basis set. The results compare favorably with sophisticated quantum chemical calculations. We believe our approach can serve as an alternative way to build up the exchange-correlation energy functional for an improved density functional theory description of systems with strong electron correlations.

When the thickness of ultrathin metal films approaches the nanometer scale comparable to the coherence length of the electrons, significant effects on the structure stability and the electronic properties of the metal films emerge due to electron confinement and quantization of the allowed electronic states in the direction perpendicular to the film. Using first-principles calculations, we showed that such quantum size effects can induce oscillatory electrostatic potential and thus alternating electric field on the surface of the wedge-shaped Pb(111) films. The alternating electric field has significant influence on surface reactivity, leading to selective even- or odd-layer adsorption preference depending on the charge state of the adatoms, consistent with the odd-layer preference of higher Mg coverage on wedge-shaped Pb(111) films, as observed in experiment.

We have performed extensive first-principles calculations to study the binding and diffusion of a Ge ad-dimer on the Si(001) surface. The diffusion of a Ge ad-dimer along the lowest-energy-barrier pathway is found to be piecewise but with strong correlation, and its energy barrier is 0.77 eV. Such a correlated piecewise diffusion pathway is also found to be favored by a Si ad-dimer on Si(001), with an energy barrier of 1.02 eV in excellent agreement with experimental measurements. (c) 2000 The American Physical Society.

Car-Parrinello molecular dynamics simulated annealings were carried out for clusters ${\mathrm{Si}}_{n},$ ${\mathrm{Ge}}_{n},$ and ${\mathrm{Sn}}_{n}$ $(nl~13).$ We investigate the temperature regions in which these clusters transform from a ``liquidlike'' phase to a ``solidlike'' phase, and then from the ``solidlike'' phase to the ground-state structures. Additional simulated annealing was also performed for the cluster ${\mathrm{C}}_{13}$ which is selected as a prototype of small carbon clusters. In addition to the discovery of structures for Sn and Ge clusters, our simulation results also provide insights into the dynamics of cluster formation.

A transferable orthogonal tight-binding model is developed for molybdenum, with special emphasis on applications in molecular-dynamics studies. The elements of the Hamiltonian matrix and the repulsive potential are allowed to depend on the environment in order to account for the effects of the neglected three-center matrix elements, for the neglected nonorthogonality effects and for the variation of the finite set of basis orbitals in different configurations as well. To check the accuracy of the model, the structural energy differences, the elastic constants, the phonon spectrum along high-symmetry lines in the Brillouin zone, the formation and migration energy of a vacancy, the formation energy of an octahedral interstitial atom, surface energies, and relaxations, as well as reconstructions of a (100) surface, are calculated and compared with ab initio data and experimental results.

We present a tight-binding model which goes beyond the traditional two-center approximation and allows the hopping parameters and the repulsive energy to be dependent on the binding environment. Using carbon as an example, we show that the approach improves remarkably the transferability of the tight-binding model. The properties of the higher-coordinated metallic structures are well described by the model in addition to those of the lower-coordinated covalent structures. \textcopyright{} 1996 The American Physical Society.

We performed an experiment to study the thermal stability of metal on graphene. We show that Fe, Gd, Dy, and Eu on graphene exhibit island growth morphology. The 3D islands of Fe, Gd, and Dy show a small decrease in island density and an increase in the height/width aspect ratio upon thermal annealing. By contrast, coarsening of Eu islands to a close flat film is observed after annealing to 365 K. By investigating the fundamental interactions (i.e., adsorption energies and diffusion barriers) between the metal adatoms and graphene using ab initio calculations, we predict that most of the 3$d$ and group 10 transition metals, noble metals, as well as rare earth metals on graphene should exhibit a 3D growth mode as observed in experiment. Most of the metal nanostructures on graphene should also be stable against aggregation. The 3D morphology and strong thermal stability of Fe, Co, Ni, Pt, and Gd islands on graphene can serve as good candidates for surface-supported catalysis applications.

Directed assembly of an array of Ru nanoclusters (NCs) is achieved by deposition of Ru at around room temperature on a single layer of graphene supported on Ru(0001). In this system, directed assembly is guided by the periodic moire structure of the buckled graphene sheet. Behavior is analyzed utilizing both scanning tunneling microscopy and atomistic lattice-gas modeling together with kinetic Monte Carlo simulation. We elucidate the kinetics of NC nucleation and growth, specifically assessing the coverage dependence of the NC density and height distribution. In addition, we provide a detailed characterization of the development of short-range spatial order within the NC array, identifying a tendency for row formation.

Amorphous carbon (a-C) structures over a wide range of densities are generated using tight-binding molecular-dynamics simulations. The a-C networks obtained by quenching low-density liquids consist of mostly threefold coordinated atoms. These a-C structures correspond to those obtained experimentally by evaporation of sputtering of graphite. However, the a-C networks generated by quenching high-density liquids are found to be dominated by tetrahedral bonding sites, which resemble the diamondlike a-C films produced by the mass-selected ion beam deposition technique. Our study shows that the shape and position of the first peak of the radial distribution function in a-C are very sensitive to the relative concentration of sp, ${\mathit{sp}}^{2}$, and ${\mathit{sp}}^{3}$ bondings. The peak position shifts towards larger distance as the percentage of the fourfold sites increases, which gives a good indication of the relative population of threefold to fourfold sites in the system.

The nucleation and growth of Fe on graphene is highly unusual. A constantly increasing island density indicates the presence of strong, predominantly repulsive, adatom interactions. We study these interactions by first-principles calculations to identify their origin. We find that the interactions consist of a short-range attraction and longer-range repulsion. We show that electric dipole-dipole interaction can contribute only part of the repulsive force (\ensuremath{\sim}30$%$) between Fe adatoms, and the repulsion due to the elastic interaction is small. We suggest that the dominant contribution to the repulsive energy would originate from the indirect (or RKKY) interactions mediated through the delocalized electrons on graphene.

Using ab initio molecular dynamics simulations, we find direct evidence that icosahedral clusters have a strong tendency to aggregate in Zr${}_{1\ensuremath{-}x}$Cu${}_{x}$ liquids. Remarkably, the dynamic arrest effect is much more significant in aggregated icosahedral clusters than in nonaggregated ones by a factor of 2 or more. Simulations using a lattice model demonstrate that the dynamic arrest effect resulting from the aggregation of icosahedral clusters may be a fundamental characteristic of the glass transition. Our studies provide an atomistic structural mechanism for dynamic arrest and glass formation.

Adsorption of the alkali-, group-III, and 3$d$-transition-metal adatoms (Na, K, Al, In, V, Fe, Co, and Ni) on graphene was studied systematically by first-principles calculations. The bonding character and electron transfer between the metal adatoms and graphene were analyzed using the recently developed quasi-atomic minimal basis set orbitals (QUAMBOs) approach. The calculations showed that the interaction between alkali-metal adatoms and graphene is ionic and has minimal effects on the lattice and electronic states of the graphene layer, in agreement with previous calculations. For group-III metal adatom adsorptions, mixed covalent and ionic bonding is demonstrated. In comparison, 3$d$-transition-metal adsorption on graphene exhibits strong covalent bonding with graphene. The majority of the contributions to the covalent bonds are from strong hybridization between the ${d}_{{x}^{2}\ensuremath{-}{y}^{2}}$ and ${d}_{\mathit{yz}}$ orbitals of the 3$d$-transition-metal adatoms and ${p}_{z}$ orbitals of the carbon atoms. The strong covalent bonds cause large in-plane lattice distortions in the graphene layer. Charge redistributions upon adsorptions also induce significant electric dipole moments and affect the magnetic moments.

US Department of Energy, Basic Energy Sciences, Division of Materials Science and Engineering [DE-AC02-07CH11358]; NSF [EAR-0757903, EAR-0810272, EAR-1047629, ATM-0426757]

It is well known that conversion of $\text{Si}(111)\text{\ensuremath{-}}(7\ifmmode\times\else\texttimes\fi{}7)$ into the $(\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3})R30\ifmmode^\circ\else\textdegree\fi{}$ phase of adsorbed Ag requires a change in the Si density, and causes formation of islands and holes at the surface. By mass balance, the ratio of areas of islands and holes $({R}_{\text{IH}})$ should be approximately 1. However, we find that the ratio is significantly higher, depending on preparation conditions. A possible explanation would be that there are different types of $(\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3})R30\ifmmode^\circ\else\textdegree\fi{}$ structures. However, neither scanning tunneling microscopy nor density-functional theory (implemented as a genetic algorithm search) supports this explanation. We propose that the edges of the islands contain excess Ag which becomes available to expand the holes, when the island perimeter decreases. Under certain conditions, excess Ag is also made available by dissolution of small islands that are Ag rich.

An atomistic cluster alignment method is developed to identify and characterize the local atomic structural order in liquids and glasses. With the ``order mining'' idea for structurally disordered systems, the method can detect the presence of any type of local order in the system and can quantify the structural similarity between a given set of templates and the aligned clusters in a systematic and unbiased manner. Moreover, population analysis can also be carried out for various types of clusters in the system. The advantages of the method in comparison with other previously developed analysis methods are illustrated by performing the structural analysis for four prototype systems (i.e., pure Al, pure Zr, ${\text{Zr}}_{35}{\text{Cu}}_{65}$, and ${\text{Zr}}_{36}{\text{Ni}}_{64}$). The results show that the cluster alignment method can identify various types of short-range orders (SROs) in these systems correctly while some of these SROs are difficult to capture by most of the currently available analysis methods (e.g., Voronoi tessellation method). Such a full three-dimensional atomistic analysis method is generic and can be applied to describe the magnitude and nature of noncrystalline ordering in many disordered systems.

We have calculated the T{sub 0} curves for several Al-rare-earth binary alloys to assess the importance of the transport-based resistance to crystallization in the overall glass formation process and the general effectiveness of thermodynamic prediction of glass-forming ability. Our results show that the experimentally observed glass-forming compositions for Al-(Ce, Gd, Ho, Nd, Y, Dy) alloys strongly correlate with the composition range bounded by the T{sub 0} curves associated with the relevant crystalline phases. This indicates that sluggish material transport, together with the tendency for clustering and other types of ordering at medium-range scale, is a key factor governing glass formation in these systems.

We have carried out molecular-dynamics simulation of liquid Si using the tight-binding Hamiltonian proposed by Goodwin, Skinner and Pettifor. The simulation results on the structural and dynamical properties (radial and angular distribution functions, the electronic density-of-states, and the diffusion constant) are presented and compared with results from classical-potential simulation and ab initio molecular dynamics (Car-Parrinello method).

We developed a global structure optimization method, genetic algorithm, for a fast and efficient prediction of grain-boundary structures. Using this method we predicted the most stable structures and a number of low-energy metastable structures for Si[001] symmetric tilted grain boundaries with various tilted angles. We show that most of the grain-boundary structures can be described by the structural unit model with the units being the dislocation cores and perfect-crystal fragments. The energies of the grain-boundary structures obtained from the genetic algorithm optimization are evaluated by tight-binding calculations using the environment-dependent Si tight-binding potential developed previously and found to be in very good agreement with the first-principles calculation results.

We consider the effect of thermal phonon displacements on the coherent transport in carbon nanotubes. The atomic displacements are generated using tight-binding molecular dynamics simulations, and the conductances are computed using a nonequilibrium Green's function technique. Atomic displacements due to lattice vibrations lead to different levels of conductance reduction and fluctuation on the massive and massless bands of a metallic nanotube. Different conduction regimes are studied by examining the resistance on different length scales. The temperature-induced displacements have a significant impact on the ballistic or diffusive transport in carbon nanotube.

Scanning tunneling microscopy experiments reveal that small Pb islands with unstable heights, e.g., four layers, on a Si(111) surface decay during coarsening, whereas large islands do not decay but grow to a stable height. This bifurcation in evolution is analyzed by incorporating quantum size effects into theoretical models for island growth dynamics with appropriate geometries. The effective energy barrier for Pb atoms to reach the top of four-layer islands is estimated at about 0.26 eV.

The temperature-dependent thermal conductivity ~(T) of amorphous silicon has been calculated from equilibrium molecular-dynamics simulations using the time correlations of the heat flux operator in which anharmonicity is explicitly incorporated. The Stillinger-Weber two- and three-body Si potential and the Wooten-Weaire-Winer a-Si model were utilized. The calculations correctly predict an increasing thermal conductivity at low temperatures (below 400 K). The ~(T), for T) 400 K, is affected by the thermally generated coordination-defect states. Comparisons to both experiment and previous calculations will be described. I. INTRODUCTION In spite of extensive studies, a number of outstanding problems do remain in understanding the lowtemperature properties of amorphous materials. Among the intriguing features have been the linear specific heat of glasses, which is generally believed to be due to the presence of localized two-level states. The nature of the vibrational modes in glasses and the excess vibrational density of states at low frequencies is also an interesting aspect. In this paper we study the thermal conductivity of amorphous materials, using amorphous silicon as a prototype. The temperature-dependent behavior of the thermal conductivity is amorphous materials has the following three characteristic regimes. (1) At low temperatures (T( 1 K), tc( T) is proportional to T', which was explained by Anderson, Halperin, and Varma' as being due to the scattering from localized two-level states; (2) at intermediate temperature (1 ( T ( 30 K), a plateau was seen, which has attracted many theoretical investigations; (3) at high temperature (T ~ 30 K), where tc(T) increases smoothly to a limiting value, in contrast to the crystalline insulators where it decreases with 1/T. Birch and Clark and Kittel gave qualitative explanations of regime (3) using the kinetic formula tc=Cvl/3 that is applicable only in the Boltzmann regime where one can assign velocities U to propagating modes. Allen and Feldman (AF) have examined the validity of the kinetic formula in their recent calculation of the thermal conductivity in amorphous silicon. Cahill and Pohl argued, based on Einstein model, that regime (3) can be explained by locally uncorrelated harmonic oscillators with relaxation times of the order of the period of vibration. However, this model is only valid in the limit of highly disordered crystals. In this study we will study regime (3) with a different approach from previous works. There exist many theoretical models for the structure of amorphous Si. A convenient theoretical model is the four-coordinated Wooten-Winer-Weaire (WWW) a-Si model. Biswas et al. obtained a vibrational density of states for the WWW model that agreed well with experiment. They have also obtained amorphous silicon configurations with molecular-dynamics (MD) simulations by cooling the molten silicon configuration. It is also known that Stillinger-Weber (SW) potential is valid for a wide range of properties of a-Si. Since the WWW model has only four-coordinated silicon atoms, this will be a more convenient starting point for the thermal-conductivity calculations than molecular dynamics models of a-Si (Ref. 8) that have coordination defects. In addition, the use of the WWW model together with the SW Si potential allows a direct comparison with the recent calculation of Allen and Feldman. Allen and Feldman have used the novel procedure of using the Kubo formula for calculating thermal conductivity in a-Si. Matrix elements of the heat-current operator were calculated between harmonic vibrational states. Although the zero-temperature WWW structure was used, the temperature entered through the quantum occupation of the vibrational states. AF obtained a tc(T) that increases in temperature up to 300 K and then saturates at a value close to experiment. Low-temperature values (T (300 K) for tc(T) were significantly smaller than experiment. We present in this paper an alternate approach that explicitly incorporates anharmonicities and accounts for temperature-dependent structural changes. Theoretical techniques developed by AF have served as a useful guide in the present work.