Session W27: Focus Session: Computational Nanoscience IX - Nanowire, Rods & SAMs

Conductance of telescoped double wall nanotubes calculated with ADF program package

In double wall nanotubes (DWNTs), the interlayer current is negligible compared to the intra-layer current. When the inner tube is partially extracted (telescoped) from the outer tube, however, the total current must flow between the layers so that the interlayer interaction drastically influences the conductance. Here the interlayer bonds can be considered as weak covalent bonds rather than van der Waals bonds since they are anisotropic and their number per atom is limited. In this presentation, the transfer integrals between the layers are calculated by ADF program package and their effects on the conductance in the telescoped DWNTs are investigated. They are compared with our previous results.   

Origin of Giant Piezoresistance in Pristine $<$111$>$-Si nanowires

It was found recently that silicon nanowires possess an unusually large piezoresistive coefficient, 350 times higher compared with Si bulk. Using first principles density functional calculations, we demonstrated that this stems from the strain-induced change in band ordering of surface states. The pristine $<$111$>$-Si nanowire is metallic under ambient condition but the mobility of the carrier is extremely small due to the strong localization. The compression shrinks the surface shell and hence shifts the itinerant state across the Femi level, which consequently leads a surge in conductance. The effective masses of those two bands differ by a factor of 100, a number that can roughly account the experimental data. Since the key bands for transport are surface states, the surface modification plays a vital role on the piezoresistance effects, as observed experimentally.   

The structure and stability of thin H-passivated $<$112$>$ silicon nanowires

Recent experiments on the synthesis on monocrystalline nanowires reveal that their axis can only have a limited number of crystalline orientations. Among these orientations, $<$112$>$ is the highest Miller-index wire axis and generates a rectangular cross-sectional shape. Using a combination between genetic algorithm search and density functional theory calculations, we determine the precise shape of the wire cross-section that corresponds to the lowest formation energy per silicon atom. We analyze the deviations of the cross-sectional shape from the Wulff shape, and show how the shape of the nanowires evolves as a function of cross-sectional area and the chemical potential of hydrogen.   

Small World Carbon Nanomaterials: Density Functional Theory Simulations

The possible existence of small, pure carbon molecules based on physical small-world networks is addressed using density functional theory calculations. A ring of atoms with one or more small-world connections between pairs of non-nearest-neighbor sites was chosen for the network topology. The small-world connections are made with and without additional carbon atoms placed along the link. The energy per atom of these small-world carbon systems is compared with benchmark carbon clusters such as the C$_{20}$ ring, bowl, and cage isomers, the C$_{60}$ Buckyball, monocyclic pure carbon rings ranging from C$_{4}$ to C$_{60}$, bare linear carbon chains ranging from C$_{2}$ to C$_{48}$, fullerenes ranging from C$_{20}$ to C$_{60}$, and various all-carbon graphitic fragments. The energy per atom results for these materials provides an indication that some of these pure-carbon small-world nanomaterials are reasonable for real world synthesis.   

Electronic properties of 1D LaB$_{6}$ rods

Metal hexa-borides have varieties of interesting properties and were utilized frequently in technological applications: e.g. LaB$_{6}$ is known to have extremely low work function, thus is used as one of the most popular electron emitter. Our project is initiated by the experimental findings that LaB$_{6}$ nano-rods generated stronger electric current than in the bulk case. Thus we focus on the band structure calculations of quasi-1D nano-rods with various widths and breadths for the purpose of studying the relationship between work function and rod shapes. Our samples consist of up to ten unit cells, i.e. n\textbf{\textit{a}} X m\textbf{\textit{b}} (\textbf{\textit{a}} and \textbf{\textit{b}} are lattice vectors and n X m $\le $ 10). To accomplish our calculations, we applied GGA density functional theory with all electron and relativistic effect included.   

Measuring Order and the Debye-Waller Factor for Porous Arrays

We derive methods that explain how to quantify the amount of order in ``ordered'' and ``highly ordered'' porous arrays. Ordered arrays from bee honeycomb and several from the general field of nanoscience are compared. Accurate measures of the order in porous arrays are made using the discrete radial distribution function (RDF) and the Debye-Waller Factor (DWF) from 2-D discrete Fourier transforms calculated from the real-space data using MATLAB routines. Nanoporous anodized aluminum oxide, hexagonal arrays from functional materials, hexagonal arrays from nanosphere lithography, and arrays from block copolymer lithography (all taken from the literature) are compared to two-dimensional model systems. The DWF is normalized to the first harmonic and depends on N, the number of peaks in the fit for these finite arrays. We optimize N to the classical model for the DWF as a fit to reciprocal space \textbf{\textit{K}}$^{2}$.   

Molecular Simulations of Liquid/Vapor Phase Equilibria for Single Component and Binary Mixtures of Nanoparticles

Self-assembly of nano and colloidal particles into ordered structures is an important technological challenge for the design of future materials and devices. One promising self-assembly technique is the evaporation of nanoparticles suspended in droplets. However, it is difficult to experimentally observe the self-assembly process in the evaporating droplet. Computer simulation provides an avenue with which to address and directly observe the self-assembly of model nano and colloidal particles within the droplet provided an efficient model can be developed. Here we present the liquid-vapor phase envelopes for model particles as a function of particle size. We compare the liquid/vapor phase envelopes and the computational efficiency for several different models including composite particles comprised of Lennard-Jones (LJ) atoms and particles interacting via integrated LJ potentials. Results for binary mixtures of nanoparticles in a solvent of LJ atoms will also be presented. These studies provide a framework for the size range of particles that can be addressed by each model. Sandia is a multiprogram laboratory operated by Sandia Corp., a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under Contract DE-AC04-94AL85000.   

Structural and Electronic Properties of Aromatic Isocyanide Self-Assembled Monolayers on Au(111) Surface

The search for molecular assemblies with interesting transport properties for molecular electronic devices is an active field of research. Isocyanide self-assembled monolayers (SAMs) have received some attention lately, as they may provide a better $\pi$-network for electron transport than other molecular SAMs such as benzenethiols. We have studied the structural and electronic properties of the interface between a gold surface and an aromatic isocyanide SAM, using density-functional theory in the GGA-PBE approximation. Our calculations predict a herringbone arrangement at high coverage, instead of the conventional structure with $(\sqrt{3}\times\sqrt{3})R30^{\circ}$ periodicity. The most favorable geometry is however found at low coverage, where the interaction between molecules is negligible and the barriers between differently tilted geometries are small compared to room temperature. These results explain the disordered patterns recently observed in room temperature STM measurements and point at possible difficulties in using isocyanide SAMs for molecular devices. Our calculations also give insight into the alignment of the molecular energy levels with respect to the Fermi energy of the metal substrate, and the charge redistribution at the interface, which provide essential guide for understanding and predicting transport properties of these SAMs, in case ordering can be achieved.   

Computer Simulation of Ligated Nanoparticle Assembly from Solution

Nanoparticles are becoming increasingly important for the design of novel materials in a wide range of new applications. Ligation of these particles by chemical species provides a means to stabilize them into useful assemblies. It is essential to have a clear physical picture of the way these particles interact. To this end, we are investigating systems of metal nanoparticles ligated with alkyl chains. Our approach is three-fold. First, we are simulating (Monte Carlo) systems of ligated nanoparticles, including all chain/particle interactions in order to develop a model potential. Second, we are simulating (molecular dynamics) systems of these particles interacting via this model potential, varying the alkyl chain length, solvent, core material, and particle volume fraction. Finally, for comparison we are simulating these systems using theoretically derived potentials found in the literature. Initial results indicate a range of morphologies, from fractal aggregates to crystallites, depending on the temperature and potentials involved. Our goal is to provide a guide to researchers in choosing materials and assembly conditions that will lead to desired assembly properties.   

The electronic and structural properties of the self-assembled monolayer of Au-benzene-1,4-dithiol-Au molecules

The electronic and structural properties of the self-assembled monolayer of Au-benzene-1,4-dithiol-Au molecules are calculated by first-principles calculation methods. The Au-S bond length obtained is 2.20{\AA}, which is about 6.8{\%} smaller than the sum of their covalent radii of 2.36 {\AA}. The Au-S-C bond angle obtained is 98.9$^{0}$, which is within the range of known bond angles of S, e.g. $\angle $FSF=98.2$^{0}$ for$^{ }$SF$_{2}$, $\angle $ClSCl =103$^{0 }$for SCl$_{2}$ and $\angle $CSH =96.4$^{0 }$for CH$_{3}$CH$_{5}$-SH. The Au 5d band is dominantly located at -2.3 eV below the Fermi level, E$_{F}$, with a sharp peak in the partial density of states (PDOS). The PDOS's also show that the highest-occupied-molecular-orbital band contains S 3p, C 2p and Au 5d hybridized states, while the lowest-unoccupied-molecular-orbital band contains S 3p, C 2p, Au 6s and Au 5d hybridized states. The dominant Au 6s states are located at about -1.5eV and 1.3eV relative to E$_{F}$. The present result shows that Au 5d states, which are usually ignored in previous theoretical studies, play an important role in the S-Au bonding and contribute significantly to the transport property of the molecule.   

Auger electron-hole scattering leads to efficient $P\rightarrow S$ electronic relaxation in self-assembled (In,Ga)As/GaAs quantum dots

We have applied our pseudopotential approach to predict Auger-type relaxation mechanisms in million-atom quantum dots. The electronic structure of (In,Ga)As/GaAs self-assembled quantum dots shows an excited P state about $30$-$50\;{\rm meV}$ above the lowest excited S state. Measured P-to-S relaxation times for electron-hole exciton range from $2$-$10\;{\rm ps}$. Because the P-to-S energy spacing is comparable to the energy of an optical phonon, it has been argued that polaron relaxation is reponsible for the fast observed relaxation. Here, we show that in the presence of a hole, Auger electron-hole scattering---decay of the electron from P to S accompanied by an energy conserving hole excitation---leads to a fast, {\rm ps}-scale decay without invoking polaron relaxation. To this end, we calculate the P-to-S decay lifetime $\tau(P\rightarrow S)$ of electrons in lens-shaped (In,Ga)As/GaAs dots due to Auger electron-hole scattering. We find that this Auger-type relaxation mechanism leads to $\tau(P\rightarrow S)\sim 1$-$7\;{\rm ps}$ for dots of different size, in agreement with available data.