Symposium Organizers
Giulia Galli University of California-Davis
Duane Johnson University of Illinois, Urbana-Champaign
Mark Hybertsen Brookhaven National Laboratory
Sadasivan Shankar Intel Corporation
Z1: Interface Issues in Catalysis, Fuel Cells and Hydrogen Storage Materials
Session Chairs
Tuesday PM, April 14, 2009
Room 2000 (Moscone West)
9:30 AM - **Z1.1
Chemistry, Electrochemistry, & Photoelectrochemistry of Si Surfaces.
Nathan Lewis 1
1 Chemistry, California Institute of Technology, Pasadena, California, United States
Show AbstractControl over the energetics and charge-transfer kinetics of Si surfaces is important to enable their use as photoelectrodes for water splitting. Functionalization of semiconductor surfaces is important to suppress recombination and allow for chemical control over the surface chemistry. We will discuss methods for alkylation of Si surfaces that provide chemical and electrical passivation, as well as enable control over the Schottky barrier height of Si/metal contacts of such systems. Control over the pH-dependence of the flat-band potential of such electrodes is critical for obtaining high-photovoltages in H2 production from Si photocathodes, both in crystalline and nanowire-array morphologies.
10:00 AM - **Z1.2
New Materials for Energy Conversion: Fundamental Insights and Improvements from Theoretical Calculations.
Jeffrey Grossman 1
1 Berkeley Nanoscience and Nanoengineering Institute, University of California, Berkeley, Berkeley, California, United States
Show AbstractClassical and quantum mechanical calculations are employed to understand important microscopic mechanisms in photovoltaic and thermoelectric materials and interfaces. Our goal is to predict key properties that govern the conversion efficiency in these materials, including structural and electronic effects, interfacial charge separation, electron and hole traps, excited state phenomena, band level alignment, and novel synthesis approaches. An overview of our work will be presented, with emphasis on how these computational approaches can improve our understanding and lead to more efficient devices.
10:30 AM - Z1.3
DFT Study of the Effect of Alumina Support on Pt Catalytic Activity.
Jennifer Synowczynski 1 , Jan Andzelm 1 , Dionisios Vlachos 2
1 , U.S. Army Research Laboratory, Aberdeen Proving Grounds, Maryland, United States, 2 , University of Delaware, Newark, Delaware, United States
Show AbstractAlumina supported catalytic Pt-nanoclusters have been used to promote a variety of reactions including the steam reforming of methane and Fischer-Tropsch synthesis. Understanding the influence of the Pt/alumina interface is key to facilitating combustion reactions in small scale reactors. Although there are many computational studies which detail the complete reaction mechanism for reactant and product species interacting with the catalytically active metal-cluster, few studies consider the pathways that arise due to presence of the Pt/alumina interface. In this paper, we first study the chemisorption of small Pt clusters on an Al terminated alpha-alumina (0001) surface and identify two unique adsorption structures for Pt trimer clusters and four adsorption structures for atomic Pt. We then investigated the thermochemistry and kinetics for dissociation and surface diffusion processes involving small molecular fragments such as water, hydroxyl, hydrogen and oxygen molecules on both the support and at the support/metal interface. The reaction barriers for dissociation and diffusion processes were calculated using the Density Functional Theory (DFT)- Generalized Gradient Approximation (GGA) and compared to experimental data, when available. We found that the kinetics for oxygen surface diffusion remained relatively unaffected by the Pt/alumina interface for diffusion paths that were approximately 3.7Å from the interface. However, near the interface, we identified several Pt-O-Al and Pt-H-Al complexes that formed, thereby changing the dissociation and diffusion barriers.
10:45 AM - Z1.4
Atomic-Scale Analysis of Equilibrium Surface Segregation in Ternary Compound (III-V and II-VI) Semiconductor Nanostructures
Sumeet Pandey 1 , Tejinder Singh 1 , Dimitrios Maroudas 1
1 Department of Chemical Engineering, University of Massachusetts - Amherst, Amherst, Massachusetts, United States
Show AbstractCrystalline semiconductor nanostructures with dimensions over the range from 2 to 10 nm (quantum dots) exhibit size-dependent luminescence due to quantum confinement of excitons. This leads to unprecedented tunability in band gap, which can be controlled by varying the composition and size of the nanocrystals. In this presentation, we address the problem of equilibrium compositional distribution in crystalline nanoparticles of ternary compound semiconductors. We focus on equilibrium surface segregation in ternary III-V and II-VI semiconductor nanocrystals and discuss its implications on the synthesis of core/shell structures. Our analysis is based on Monte Carlo (MC) and conjugate gradient (CG) methods according to classical force fields for combined compositional and structural relaxation in conjunction with first-principles density functional theory (DFT) calculations. The DFT calculations are carried out within the generalized gradient approximation, using plane-wave basis sets, ultra-soft pseudopotentials, and supercell models, and are used for validation of the classical force fields employed in the analysis.We have conducted MC/CG relaxation simulations for InxGa1-xAs and ZnSe1-xSx slab supercells exposing two free surfaces with either [001], or [111], or [110] crystallographic orientation. For the simulations, we have used properly modified/extended parameterizations of the valence-force-field description of directional bond stretching and bending. The parameterizations were verified by comparisons with DFT calculations of energy differences for various configurations of InxGa1-xAs and ZnSe1-xSx slabs at different values of the compositional parameter, x. The validated classical force fields were then used to analyze surface segregation phenomena and the resulting equilibrium concentration profiles in ternary crystalline nanoparticles with well-defined facets for diameters d > 5 nm. Our relaxation method consists of a multi-stage sequence that includes MC sweeps employing exchanges between In and Ga atoms in InxGa1-xAs structures and between Se and S atoms in ZnSe1-xSx structures, followed by many continuous-space MC sweeps over all atoms for structural relaxation and an MC step for strain/volume relaxation of slabs/particles after each such sweep; in all the stages, trials are accepted or rejected according to the Metropolis criterion. The MC simulation is preceded and followed by energy minimization according to a CG scheme to account for local structural relaxation. The underlying surface segregation phenomena are analyzed in particle-size and composition space and their impact on the experimental synthesis of core/shell nanocrystal structures is discussed.
11:30 AM - **Z1.5
What do we Know about Electrocatalysis in Solid State Fuel Cells?
Sossina Haile 1
1 Materials Science, California Institute of Technolog, Pasadena, California, United States
Show AbstractThe global electrode reactions relevant to fuel cells are well-known. For example in a fuel cell based on a proton conducting electrolyte, the anode and cathode half cell reactions are Anode: H2 → 2H+ + 2e- (1), and Cathode: ½ O2 + 2H+ + 2e- → H2O (2). The analogous reactions in the case of an oxygen ion conducting electrolyte are Anode: H2 + O= → H2O + 2e-(3), and Cathode: ½ O2 + 2e- → O= (4). It is widely held that, unlike the situation with liquid electrolytes, the occurrence of the electrochemical reaction in solid state systems is limited to the triple phase boundary at which the electrode, electrolyte and gas phase are simultaneously in contact with one another. In fact, an increasing number of systems violate this constraint, and indeed the search for high performance electrodes often becomes a search for permeable materials that relax the geometric tyranny of the triple phase boundary. Distinguishing material systems in which such boundaries play a role is essential, but represents only the first step in developing an atomistic level understanding of the electrochemical reaction pathway. We review here recent attempts to uncover mechanistic pathways in fuel electrooxidation and oxygen electroreduction, beginning with the simplest case of hydrogen electrooxidation in both solid oxide and solid acid fuel cells.
12:00 PM - Z1.6
Modeling the Sorption Dynamics of AlH3 using a Reactive Force Field.
Julius Ojwang 1 , Adri van Duin 3 2 , William Goddard 2 , Gert Kramer 1 , Rutger van Santen 1
1 Chemistry, Eindhoven University of Technology, Eindhoven Netherlands, 3 Mechanical and Nuclear Engineering, Penn State University, University Park, Pennsylvania, United States, 2 Materials and Process Simulation Center, California Institute of Technology (Caltech), Pasadena, California, United States
Show AbstractWe have parameterized a reactive force field, ReaxFFAlH3, for aluminum hydride with the objective of describing H2 desorption process in AlH3. We aim to shed more light on the long range transport mechanisms of Al atoms during (de)sorption process and the dynamics governing hydrogen desorption process in AlH3. ReaxFF has already been shown to be able to accurately predict the dynamical and reactive processes in MgH2[1] and NaH[2]. In this presentation the details of the parameterizations of ReaxFFAlH3, the diffusion mechanism of hydrogen atoms and hydrogen molecules in AlH3 and abstraction process of molecular H2 in AlH3 clusters will be discussed. A key feature in ReaxFF is using the bond-order formalism that allows for bond breaking and formation as per Tersoff, Brenner and environment dependent interatomic potential (EDIP) approach. ReaxFF includes polarizable charges which are calculated using electronegativity equalization method (EEM), which provides a geometry dependent charge distribution. ReaxFF calculates non-bonded (van der Waals and Coulomb) interactions between all atoms (including 1-2, 1-3 and 1-4 interactions) making it suitable for systems in which there are both covalent and ionic interactions. It is this last feature, coupled with the ability to continuously create and dissociate bonds, that makes ReaxFF attractive for modeling ReaxFFAlH3 in which there is an interplay of both polar and covalent interactions.Parameterizations of the ReaxFFAlH3 energy expressions was done by fitting to a training set containing the ab initio derived equations of state (EoS) of pure Al and AlH3 condensed phases, reaction energies and bond dissociation profiles on small finite clusters. The parameterized force field, ReaxFFAlH3, is used to study the dynamics governing hydrogen desorption in AlH3. During the abstraction process of surface molecular hydrogen in AlH3 charge transfer is found to be well described by the parameterized force field. A molecular dynamics run is done, which shows that a clear signature of hydrogen desorption is the fall in potential energy surface during heating. Using the force field we have also unambiguously identified a molecular hydrogen trapped in the channels of a cluster of AlH3[3].[1] Sam Cheung, Wei-Qiao Deng, A. C. T. van Duin, and W. Goddard III. J. Phys. Chem. A, 109 ( 2005) 851–859 [2] Ojwang et al., J. Chem. Phys. 128 (2008) 164714 [3] L. Senadheera et al., J. Alloys and Compd. 463 (2008)1
12:15 PM - Z1.7
Proton Transfer in Monoclinic Zirconia and Yttrium-Doped Monoclinic Zirconia: A Theoretical Study.
Yves Mantz 1 , Randall Gemmen 1
1 , DOE National Energy Technology Laboratory, Morgantown, West Virginia, United States
Show AbstractIn solid oxide fuel cells, operated between 773-1273 K, the mechanism of hydrogen oxidation at the anode, typically composed of a nickel/yttria-stabilized zirconia (YSZ) cermet if the electrolyte is YSZ, is complex. Two classes of mechanisms are thought to be operative, spillover and interstitial hydrogen transfer. A slow step of the latter is proton diffusion in the electrolyte, terminated by the formation of water. The YSZ surface region, where proton diffusion is likely, in contact with the anode is a complex structure but may consist of a few layers dominated by a monoclinic zirconia phase 6-nm thick, based on the analysis of experimental data.Accordingly, proton transfer in monoclinic zirconia is studied within the framework of density-functional theory, using the HCTH/120 exchange-correlation functional, as implemented in the CPMD software package. To validate the methodology, the Murnaghan equations of state of the three low pressure zirconia polymorphs are predicted, for the first time using the HCTH/120 functional. The fit parameters are generally consistent with previous theoretical and experimental results obtained. Importantly, monoclinic is predicted to be the most stable low pressure phase of zirconia. From a Born-Oppenheimer molecular dynamics (BOMD) simulation, at an average temperature of 1000 K within the microcanonical (NVE) ensemble, the Zr-O coordination number of monoclinic zirconia is computed to be seven, in agreement with crystallographic data.Subsequently, the preferred binding site of the proton in monoclinic zirconia is identified, oxygen atom, O1, which is favored by 0.4 eV with respect to, O2. During a BOMD simulation at an average temperature of 690 K, the two hydrogen bonds formed between the proton at O1 and its favored oxygen atom neighbors are broken repeatedly, and the proton is forming 1.25 bent hydrogen bonds, on average, with several neighbors. By contrast, during a second BOMD simulation at the higher (average) temperature of 990 K, successive proton transfers are observed between O1 sites. In accordance with the crystal symmetry, a particular O1 site is not favored. However, the O2 sites, which are higher in energy, are not sampled at this temperature.These results are compared to those from two additional BOMD simulations, at 730 and 1030 K, of the proton in monoclinic zirconia with one of the zirconium (+4) atoms substituted by an yttrium (+3) atom. The former simulation is purposely initiated from an equilibrated configuration in which the proton is “freer”, to estimate its vibrational frequency of 3570 cm-1. During the simulation at 1030 K, several proton transfer events are observed. Consistent with the fact that O1 sites closer to the less positively charged dopant are energetically favorable, the proton is spending most of its time near the yttrium atom dopant, suggesting a more complex (microscopic) picture of proton diffusion in the presence of a dopant.
12:30 PM - **Z1.8
Computational Discovery of Novel Hydrogen Storage Materials and Reactions
Christopher Wolverton 1
1 Materials Science and Engineering, Northwestern University, Evanston, Illinois, United States
Show AbstractPractical hydrogen storage for mobile applications requires materials that exhibit high hydrogen densities, low decomposition temperatures, and fast kinetics for absorption and desorption. Unfortunately, no reversible materials are currently known that possess all of these attributes. Here we present an overview of our recent efforts aimed at developing a first-principles computational approach to the discovery of novel hydrogen storage materials. We have developed computational tools which enable accurate prediction of decomposition thermodynamics, crystal structures for unknown hydrides, and thermodynamically preferred decomposition pathways. We present examples that illustrate each of these three capabilities. Specifically, we focus on recent work on crystal structure and dehydriding reactions of borohydride materials, such as Mg(BH4)2, MgB12H12, and mixtures of complex hydrides such as the ternary LiBH4/LiNH2/MgH2 system.References1. V. Ozolins, E. H. Majzoub, and C. Wolverton, "First-Principles Prediction of a Ground State Crystal Structure of Magnesium Borohydride", Phys. Rev. Lett. 100, 135501 (2008).2. C. Wolverton, D. J. Siegel, A. R. Akbarzadeh, and V. Ozolins, “Discovery of Novel Hydrogen Storage Materials: An Atomic Scale Computational Approach”, J. Phys. Condens. Matt. 20, 064228 (2008).3. J. Yang, A. Sudik, D. Halliday, D. J. Siegel, C. Wolverton et al., "A Self-Catalyzing Hydrogen Storage Material" Angew. Chem. Int. Ed., 47, 882 (2008).4. A. R. Akbarzadeh, V. Ozolins, and C. Wolverton, “First-Principles Determination of Multicomponent Hydride Phase Diagrams: Application to the Li-Mg-N-H System”, Advanced Materials 19, 3233 (2007).5. D. J. Siegel, C. Wolverton, and V. Ozolins, "Thermodynamic Guidelines for the Prediction of Hydrogen Storage Reactions and their Application to Destabilized Hydride Mixtures”, Phys. Rev. B 76, 134102 (2007).
Z2: Interface Issues in Catalysis, Fuel Cells and Hydrogen Storage Materials II
Session Chairs
Tuesday PM, April 14, 2009
Room 2000 (Moscone West)
2:30 PM - **Z2.1
Materials-related Aspects of TiO2-based Photocatalysis.
Annabella Selloni 1
1 Chemistry, Princeton University, Princeton, New Jersey, United States
Show AbstractTitanium dioxide (TiO2) is a technologically important material, that is widely applied in photocatalysis and solar energy conversion. Since surfaces have a key role in all these applications, there has been an increasing interest in the fundamental physical and chemical properties of TiO2 surfaces over the last decade. In this talk we will present recent theoretical work on issues relevant to TiO2-based photocatalysis. The connections between our results and the experiment will be highlighted. Topics to be discussed include the relationship between surface structure and reactivity, the role of surface and subsurface defects, and surface sensitization via adsorbed chromophores as used in dye sensitized solar cells.
3:00 PM - Z2.2
Molecular Dynamics Simulations of Water Dynamics on Titanium Oxide.
Paul Kent 1 , Nitin Kumar 2 , Jorge Sofo 2 3
1 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States, 2 Department of Physics, Penn State University, University Park, Pennsylvania, United States, 3 Materials Research Institute, Penn State University, University Park, Pennsylvania, United States
Show AbstractThe surface of crystalline titania (TiO2) is one of the most investigated metal-oxide surfaces[1], but still remains the source of many controversies, particularly those related to its interaction with water. For example, photoinduced superhydrophobicity[2] has been demonstrated, with ramifications for self-cleaning surfaces, but the microscopic origin of this effect is still debated. Many other technologically relevant puzzles remain.The study of water-related phenomena has proven particularly challenging for simulation due to the dynamic nature of the adsorped water layer and the delicate nature of the interactions present near the surface. We have studied the dynamics of water at the (110) surface of titania using a series of large scale quantum molecular dynamics calculations using density functional methods. Using codes optimized for modern supercomputer platforms we have recently been able to perform simulations with sufficient lengthscale (hundreds of atoms), timescale (>10ps) , temporal resolution (0.5fs), and temperature range to make systematic comparison with recent x-ray and neutron scattering data. We find a layered structure at the surface, in agreement with previous simulations, but also find good agreement in the dynamics as observed in quasielastic neutron scattering measurements. The simulations reveal a plethora of information about how the microscopic hydrogen bonds, dissociation, and vibrational frequencies combine to dominate the macroscopic behavior of the samples. A similar methodology can now be applied to more complex geometries and other related materials.This research utilizes the powerful combination of confirmation and validation of experimentally observable features with the ability to deconvolve the results and examine, using simulation data, the underlying atomistic processes. If time allows I will discuss our studies of the structure of oxide nanoparticles, as well as the current challenges in performing these simulations.This research used resources of the National Energy Research Scientific Computing Center, National Center for Computational Sciences, and the Center for Nanophase Materials Sciences, which are sponsored by the respective facilities divisions of the Advanced Scientific Computing Research and Basic Energy Sciences of the U.S. Department of Energy.1. U. Diebold, Surface Science Reports 48, 53 (2003).2. R. Wang et al., Nature 388, 431 (1997).
3:15 PM - **Z2.3
Towards a Predictive First Principles Understanding of Water-sold Interfaces.
Angelos Michaelides 1
1 London Centre for Nanotechnology, University College London, London United Kingdom
Show AbstractSome examples of our recent work in the area of water-solid interfaces will be examined [1-5]. Specifically focusing on the nucleation of ice nanoclusters at metal surfaces and the properties of the liquid water - salt interfaces. In addition a critical assessment of the ability of conventional density functionals to accurately describe such systems will be provided [6].[1] D. Pan et al., Surface energy and surface proton order of ice, Phys. Rev. Lett. 101, 155703 (2008). [2] H. Gawronski et al., Manipulation and control of hydrogen bond dynamics in adsorbed ice nanoclusters, Phys. Rev. Lett. 101, 136102 (2008). [3] L.-M. Liu, M. Krack, and A. Michaelides, DensityOscillations in a Nanoscale Water Film on Salt: Insight from Ab Initio Molecular Dynamics, J. Am. Chem. Soc. 130, 8572 (2008)[4] A. Michaelides and K. Morgenstern, Ice nanoclusters athydrophobic metal surfaces, Nature Mater. 6, 597 (2007)[5] X.L. Hu and A. Michaelides, Ice formation on kaolinite:Lattice match or amphoterism? Surf. Sci. 601, 5378 (2007); ibid, 602, 960 (2008)[6] B. Santra, A. Michaelides, and M. Scheffler, On the accuracy of density-functional theory exchange-correlation functionals for H bonds in small water clusters: Benchmarks approaching the complete basis set limit, J. Chem. Phys. 127, 184104 (2007); B. Santra et al. (in press)
3:45 PM - Z2.4
Water Chemistry and Manipulation on Alkaline Earth Halide Surfaces.
Adam Foster 1 , Tom Trevethan 3 , Sabine Hirth 2 , Michael Reichling 2 , Alexander Shluger 3
1 Laboratory of Physics, Helsinki University of Technology, Helsinki Finland, 3 Department of Physics and Astronomy, University College London, London United Kingdom, 2 Fachbereich Physik, Universität Osnabrück, Osnabrück Germany
Show AbstractWhile scanning tunneling microscopy (STM) offers routine manipulation on conducting samples, only non-contact atomic force microscopy (NC-AFM) can provide atomic resolution and molecular manipulation on insulating surfaces. As yet, mechanical manipulation of molecules with NC-AFM on insulating surfaces remains in its infancy, with very few published studies [1-3]. Our previous studies on CaF2 (111) [2] demonstrated the potential of manipulating defects on the surface, and in recent water dosage experiments, we observed the development of two characteristic defects - one which can be manipulated by the tip and one which cannot. In order to understand this, and interpret the experimental images, we have performed systematic first principles calculations of water chemistry and mobility on the surface using the VASP code. We calculated the binding energies for molecular water and hydroxide groups on the perfect surface, as well as at neutral and charged fluoride vacancies that might act as pinning or dissociation sites for these molecules. Water was found to bind more strongly to vacancies than to the perfect surface, while the binding strength for hydroxides is weaker. Calculations of the reaction and diffusion barriers suggest that molecular water adsorbed at a charged vacancy is accessible to AFM manipulation, while dissociated water at neutral vacancies is more strongly pinned. In order to provide a comprehensive comparison to experiment, we take the lowest energy configurations of water predicted by first principles, and use atomistic methods to simulate NC-AFM imaging and tip-induced manipulation. We also use simulations to suggest explanations for the observed differences in manipulation behaviour seen on BaF2 and SrF2 (111) surfaces. [1] R. Nishi, D. Miyagawa, Y. Seino, I. Yi and S. Morita, Nanotechnology 17 (2006) S142 [2] S. Hirth, F. Ostendorf and M. Reichling, Nanotechnology 17 (2006) S148 [3] S. Fujii and M. Fujihira Nanotechnology 18 (2007) 084011
4:30 PM - **Z2.5
Organic Semiconductors: Computational Chemistry Aided Understanding of Basic Concepts and Design of New Materials.
Jean-Luc Bredas 1
1 Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia, United States
Show AbstractOver the past two decades, the science and engineering of organic semiconducting materials have advanced very rapidly, leading to the demonstration and optimization of a range of organics-based solid-state devices, including organic light-emitting diodes, field-effect transistors, photodiodes, and photovoltaic cells. Particularly attractive for organic semiconductors are flexible plastic substrates that can lead to applications and consumer products with lower cost, highly flexible form factors, and light weight. These attributes, combined with the ability to optimize the physical properties of organic (macro)molecules by fine tuning their chemical structure, constitute the main drivers boosting research and industrial interest in organic photovoltaics. In the operation of all such devices, electron-transport processes, that is, charge carrier mobilities, do play a key role. In this presentation, we will take a molecular/microscopic standpoint and discuss how computational chemistry can help in describing the various parameters that impact charge carrier mobilities in ordered systems: (i) the electronic coupling between adjacent molecules or chains; (ii) the intra- and inter-molecular reorganization energies and resulting electron-phonon couplings; and (iii) the polarization and site energies. In this way, structure/transport properties relationships start emerging.The steps that still need to be taken on the computational side in order to be in a position to predict carrier mobilities from first principles will be described.
5:00 PM - Z2.6
Monte-Carlo Simulations of PdAu Bimetallic Alloy Surface under Realistic CO Coverage.
Bin Shan 1 , Jangsuk Hyun 1 , Neeti Kapur 1 , Sang Yang 1 , Kyeongjae Cho 2
1 Computational Nanoscience, Nanostellar Inc, Redwood City, California, United States, 2 Department of Materials Science and Engineering and Department of Physics, University of Texas at Dallas, Richardson, Texas, United States
Show AbstractAlloying has been one of the strategies to develop non-Pt based CO oxidation catalyst. PdAu bimetallic alloy has recently been shown to yield better reactivity and thermal stability toward CO oxidation for diesel engine applications than pure metal catalysts. One important factor influencing the reactivity of Pd-Au alloy is its surface composition and morphology at the onset of CO oxidation, which is usually CO pre-covered in diesel engine catalysis. We studied surface segregation processes in bimetallic Pd-Au alloy using Monte-Carlo simulations, combining an improved Embedded Atom Method (EAM) potential for the bare alloy energies and an atomistic treatment for adsorbed CO molecules. The EAM potential is parametrized by fitting to DFT data of 66 configurations of PdAu slab with different segregation profiles in addition to bulk alloy properties to improve the segregation energetics. Parameters for the atomistic description of CO are extracted from a set of DFT calculations, which incorporates both first nearest neighbor and second nearest neighbor effects. We have also included a lateral interaction term for CO-CO repulsion to take into account the coverage dependent CO binding energies. The validity of the model has been verified by comparing the energies of a selected set of Pd-Au alloy energies, where the model yields very good agreement with DFT data. In order to explore the surface composition change of PdAu bimetallic alloy as a function of CO coverage, we carried out Monte-Carlo simulations of CO adsorption on a 3:1 ratio PdAu alloy surface. We found PdAu bimetallic alloy exhibits distinctive behaviors as a result of competition between surface energies of respective metals and their interaction strength with CO. Under low CO coverage of less than 0.25ML, Au segregates to the surface due to its lower surface energy; Under moderate CO coverage from 0.25ML to 0.50ML, PdAu keeps its bulk alloy structure and little adsorbate-induced segregation is observed. For high CO coverage of larger than 0.5ML, a phase transition takes place and adsorbate-induced segregation transforms PdAu alloy into an overlayer structure with a pure Pd monolayer on top of a PdAu alloy. However, our simulation indicates the saturation coverage of CO cannot reach one monolayer due to CO-CO repulsion. The maximum CO coverage we can achieve is around 0.75ML with CO forming hexagonal pattern on the surface. The ensemble of CO adsorption configurations give insight into reactive sites for subsequent oxygen dissociation and CO oxidation on PdAu surface, which are of special importance to diesel engine catalysis. The enrichment of Pd on the surface with increasing CO coverage has been confirmed experimentally. To our knowledge, this is the first time that first-principles based EAM model in conjunction with an atomistic detailed description of adsorbate molecules has been used to study the segregation of bimetallic alloys.
5:15 PM - Z2.7
Understanding the Adsorption and Diffusion of Carbon Dioxide in Zeolitic Imidazolate Frameworks: a Computational Study.
Dahuan Liu 1 , Qingyuan Yang 1 , Chongli Zhong 1
1 Dept. of Chem. Eng., Beijing University of Chemical Technology, Beijing China
Show Abstract Zeolitic imidazolate frameworks (ZIFs) are porous crystalline materials with tetrahedral networks that resemble those of zeolites: transition metals (Zn, Co) replace tetrahedrally coordinated atoms (for example, Si), and imidazolate links replace oxygen bridges[1]. They are a subclass of metal-organic frameworks (MOFs) that have shown great opportunity for carbon dioxide capture[2]. This work performed the first computational study to reveal the adsorption sites and diffusion behavior of CO2 in ZIFs that can provide microscopic level information for a better understanding of the characteristics of ZIFs. Similar to our previous works on MOFs[3-5], in the present work ab initio calculations were performed to calculate the atomic partial charges in the frameworks of ZIFs, and various force fields were examined for reproducing the experimental adsorption isotherms of CO2 in ZIFs to find a suitable force field. With the atomic charges calculated and the suitable force field determined, grand canonical Monte Carlo (GCMC) simulations were carried out to study the adsorption isotherms, saturated adsorption values, as well as the adsorption sites of CO2 in ZIFs-68 and 69. Furthermore, molecular dynamics (MD) simulations were performed to investigate the diffusivity of CO2 in these ZIFs and compared with other MOFs, as well as zeolites and carbonaceous materials. The information obtained is very useful for guiding the future design of new ZIFs with improved CO2 storage capacity as well as providing a better understanding of CO2 adsorption mechanism in ZIFs.Key references:1. B. Wang et al., Nature, 2008, 453, 207.2. R. Banerjee et al., Science, 2008, 319, 939.3. Q. Yang and C. Zhong, J. Phys. Chem. B, 2005, 109, 11862.4. Q. Yang and C. Zhong, J. Phys. Chem. B, 2006, 110, 655. 5. Q. Yang, C. Zhong and J. Chen, J. Phys. Chem. C, 2008, 112, 1562.
Symposium Organizers
Giulia Galli University of California-Davis
Duane Johnson University of Illinois, Urbana-Champaign
Mark Hybertsen Brookhaven National Laboratory
Sadasivan Shankar Intel Corporation
Z5: Molecular Electronics
Session Chairs
Thursday AM, April 16, 2009
Room 2000 (Moscone West)
9:15 AM - **Z5.1
Mechanically-Controlled Binary Conductance Switching of a Single-Molecule Junction.
Latha Venkataraman 1
1 Applied Physics, Columbia University, New York, New York, United States
Show AbstractRealization of molecular-scale binary switches is of fundamental importance to nanoscale electronics. While such switching has been reported in atomic quantum point contacts, single-molecule junctions provide the additional flexibility of tuning the on/off conductance states through molecular design. Thus far, switching in single-molecule junctions has been attributed to changes in the conformation or charge state of the molecule. In this talk, we will show our results which demonstrate reversible binary switching in a single-molecule junction by mechanical control of the metal-molecule contact geometry.To fabricate and characterize single molecule junctions, we repeatedly form molecular junctions between the gold tip and a gold substrate using a simplified scanning tunneling microscope in a solution based break-junction technique. This method has proven to be very reliable when using chemical link groups that form a donor-acceptor type bond to gold electrode [1, 2]. Here, we focus on measurements of 4,4'-bipyridine-gold junctions [3]. We measure conductance as a function of relative tip-sample displacement and find that there are two reproducible and distinct conductance states, at ~ 6 × 10-4 G0 and ~1.6 × 10-4 G0. The high conductance state occurs as soon as the gold point-contact is broken, while the low conductance state follows on junction elongation. To experimentally determine the relation between tip-substrate separation and 4,4'-bipyridine junction conductance, we determine the distance required to compress a molecular junction until the conductance reaches that of a gold-gold contact. This measured "push-back" distance shows that for junctions within the low conducting state, the gold-gold separation is around 12-13 Å, consistent with a geometry where the molecule is held vertically between the apex of the tip and substrate. For junctions in the high conducting state the tip-substrate separation ranges from 7-11 Å, indicating that the molecule is probably tilted within the junction. These measurements thus provide direct evidence that the two conductance states occur at different gold-gold separation. We finally show that a single 4,4'-bipyridine junction can be repeatedly switched between its low and high conducting states, by modulating the tip-sample separation, demonstrating a new class of mechanically-activated single-molecule switch with a pyridine-gold molecular link. Our theoretical collaborators have done extensive calculations that demonstrate how the interplay between the changes in the mechanical constraints in the junction and the local pyridine N-Au bonding leads to high and low conducting configurations [3]. [1] L. Venkataraman, et al., Nano Lett. 6, 458 (2006).[2] Y. S. Park, et al., J. Am. Chem. Soc. 129, 15768 (2007).[3] S.Y. Quek, M. Kamenetska, et al,. submitted (2008)
9:45 AM - Z5.2
First principles studies of single-molecule junctions: Conductance and Mechanically-Controlled Switching.
Su Ying Quek 1 , Hyoung Joon Choi 2 , Steven Louie 3 1 , Mark Hybertsen 4 , Latha Venkataraman 5 6 , J. Neaton 1
1 Molecular Foundry, Lawrence Berkeley National Lab, Berkeley, California, United States, 2 Department of Physics and IPAP, Yonsei University, Seoul Korea (the Republic of), 3 Department of Physics, University of California, Berkeley, Berkeley, California, United States, 4 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, New York, United States, 5 Department of Applied Physics and Applied Mathematics, Columbia University, New York, New York, United States, 6 Center for Electron Transport in Nanostructures, Columbia University, New York, New York, United States
Show AbstractA fundamental challenge in nanoscience is to predict and understand the electronic conductance of individual molecules. Recent measurements [1] of the electronic conductance of single-molecule junctions (single molecules bonded to macroscopic gold electrodes by chemical endgroups) provide an opportunity to benchmark standard first-principles methods while quantitatively exploring fundamental mechanisms of molecular-scale charge transport. Here we describe work using a self-energy corrected density-functional theory (DFT)-based scattering state approach [2] to explore the single-molecule conductance of junctions with amine-Au and pyridine-Au links, in the context of recent experiments. Using a physically motivated approximate self-energy correction based on GW calculations of metal-molecule interfaces [3], we explore quantitatively how junction geometry, link chemistry, and length can affect transport properties and lead to novel behavior, such as reversible conductance switching. We find that amine-Au bonding results in a conductance that is insensitive to details of the junction structure, in accordance with single well-defined conductance peaks observed in experiment [4]. The self-energy corrected conductance agrees well with experiment, in contrast to the DFT value which is about 7 times larger [4]. Using first principles calculations on about 60 different bipyridine-Au junction geometries, we demonstrate, together with experiment, that reversible binary conductance switching can result from mechanically-induced changes in the metal-molecule contact geometry. The flexible but stable N-Au bond results in distinct geometries depending on the geometric constraints of the junction separation. At large separations, the molecule prefers a vertical geometry, with the N-Au bond perpendicular to the conducting pi-system. This results in weaker coupling and lower conductance. In contrast, at small separations (when the Au contact is just broken), the molecule cannot be accommodated in a vertical geometry. The N-Au bond orients at an angle to the conducting pi-system, resulting in stronger coupling and higher conductance. [1] L. Venkataraman et al, Nano Lett 6, 458 (2006)[2] H. J. Choi et al, Phys Rev B, 76, 155420 (2007)[3] J. B. Neaton et al, Phys Rev Lett, 97, 216405 (2006)[4] S. Y. Quek et al, Nano Lett. 7, 3477 (2007)
10:00 AM - Z5.3
Simultaneous Force and Conductance Measurements of Single Molecule Junctions
Michael Frei 1 , Maria Kamenetska 1 2 , Max Koentopp 2 , Mark Hybertsen 3 , Latha Venkataraman 1 2
1 Applied Physics and Applied Mathematics, Columbia University, New York, New York, United States, 2 Center for Integrated Science and Engineering, Columbia University, New York, New York, United States, 3 Center for Functional Nanomaterials, Brookhaven National Laboratory, Upton, New York, United States
Show AbstractSimultaneous conductance and force measurements of single molecule junctions are performed by repeatedly forming and breaking junctions between a molecule coated gold substrate and a gold-coated cantilever in a simplified atomic force microscope (AFM), using molecules with Amine, Methyl Sulfide and Dimethyl Phosphine link groups. Previous conductance measurements have shown that alkanes terminated with these links bind selectively to under-coordinated gold atom, providing a reliable and reproducible junction conductance [1-2]. While conductance measurements of single molecular junctions have been investigated extensively, the simultaneous measurement of force provides an additional probe into junction properties as well as its stability.Here, we first present force measurements for gold point contact, carried out in ambient conditions, where we see that sharp drops in conductance, as the contact is thinned out, are accompanied with simultaneous changes in force across the junction. We find here that the force required to break a single gold-gold bond is consistent with previously reported values of 1.5 nN [3]. We will then show results from simultaneous force and conductance measurements of three molecules, 1,4 diaminobutane, 1,4 bis (methyl thiol) butane, and 1,2-bis (dimethyl phosphino) ethane. These molecules are directly evaporated onto the gold substrate, as force measurements cannot be carried out in a solution of the molecules. For the first two molecules, we measure a junction breaking force of about 0.5 nN – 0.7 nN, significantly smaller than that required to break a clean gold-gold bond, indicating that the junction breaks at the Au-N and Au-SMe bond respectively. Furthermore, we find that in these molecules, the final drop in junction conductance occurs simultaneously with the drop in the junction force in a majority of the measured traces. For the third molecule measured (1,2-bis(dimethyl phosphino) ethane), we find that the force required to break the molecular junction is comparable to that required to break a gold-gold bond, consistent with calculations [2]. We find further that for a significant fraction of the traces measured, the molecular junction conductance often drops by an order of magnitude, while the forces do not change significantly. We will discuss the implications of these findings and show how they relate to detailed simulation of the junction elongation process for these links [4]. [1] L. Venkataraman, et al., Nano Lett. 6, 458 (2006).[2] Y. S. Park, et al., J. Am. Chem. Soc. 129, 15768 (2007).[3] G. Rubio, et al., Phys. Rev. Lett. 76, 2302 (1996).[4] M. Kamenetska, M. Koentopp, et al,. in preparation (2008)
10:15 AM - Z5.4
Simulation and Measurement of Phosphine and Amine Linked Single Molecule Junction Evolution Under Stress.
Max Koentopp 1 , Mark Hybertsen 2 , Marie Kamenetska 3 , Adam Whalley 4 , Young Park 4 , Michael Steigerwald 4 , Colin Nuckolls 4 , Latha Venkataraman 3
1 Center for Integrated Science and Engineering, Columbia University, New York, New York, United States, 2 Center for Functional Nanomaterials, Brookhaven National Laboratories, Upton, New York, United States, 3 Department of Applied Physics and Applied Mathematics, Columbia University, New York, New York, United States, 4 Department of Chemistry, Columbia University, New York, New York, United States
Show AbstractReliable measurements of single molecule conductance are performed by repeated breaking of Au point contacts in a solution containing target molecules with link groups that readily form donor-acceptor bonds to specific Au atoms on the electrodes [1-2]. Measured traces present the junction conductance as a continuous function of junction elongation under applied stress. This approach has enabled systematic studies of the dependence of single molecule junction conductance on molecular properties such as length, conjugation and link group [3-4]. We have performed a Density Functional Theory based study of the bonding, structural evolution and conductance of model junctions, comparing the results for butane terminated by two distinct link groups, amine and dimethylphosphine. The electrodes are modeled by pyramidal clusters with 20 Au atoms. We model the junction evolution under stress by displacing one of the Au electrode structures in small steps (0.05 A) and fully relaxing the junction structure at each step, fixing a portion of each Au electrode as a reference. The low bias junction conductance at each step is computed using a Green’s function approach. We find that the junction can be readily formed through a bond between the linking N or P atom and either an Au adatom on the electrode tip face or an Au edge atom where two tip facets meet. We find that, under stress, both link groups can drive displacement of an Au adatom. While the N-Au bond in the amine case can shift from one edge Au atom to another, we find that the P-Au bond in the dimethyl phosphine case remains attached to the same Au atom through the entire elongation process. The P-Au bond is stronger and we observe more extensive plastic deformation of the Au tip structure in some cases. Correspondingly, the maximum force sustained by the N-Au bond is about 0.6-0.8 nN while that sustained by the P-Au approaches 1.5 nN, similar to the measured breaking force for Au point contacts. For all model junctions studied, we see that the calculated zero-bias electron transmission through the junction is only moderately perturbed by changes in the junction structure. From the perspective of experimental traces, this explains why conductance steps can extend over distances of several angstroms, considerably longer than expected for breaking a single bond. We compare our calculated results to an extensive experimental database correlating measured conductance step length to molecule backbone length.[1] L. Venkataraman, et al., Nano Lett. 6, 458 (2006).[2] Y. S. Park, et al., J. Am. Chem. Soc. 129, 15768 (2007).[3] L. Venkataraman, et al., Nature 442, 904 (2006).[4] M. S. Hybertsen, et al., J. Phys.: Condens Matter 20, 374115 (2008).
10:30 AM - Z5.5
First-Principles Studies of Covalently-Bonded Aromatic Molecules on Gold using a GW Approach.
Isaac Tamblyn 1 , Su Ying Quek 2 , Stanimir Bonev 1 , Jeffrey Neaton 2
1 Physics, Dalhousie University, Halifax, Nova Scotia, Canada, 2 Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United States
Show AbstractThere is considerable interest in using organic molecules as components in solar cells and other nanoscale electronic and optoelectronic devices, driving a critical need for the identification, understanding, and design of metal-organic interfaces with optimal electronic structure and charge transport properties. Frontier molecular orbital energies, accessible experimentally via photoemission or transport measurements, dictate the nature of optical absorption, chemical reactivity, and transport at metal-organic interfaces. Metal-molecule contacts are also known to play a critical role in determining the low-bias transport properties of molecular electronic devices. Recent work [1] on the conductance of benzenediamine-Au single-molecule junctions has shown that standard methods based on density functional theory fail to correctly position molecular orbital energies relative to the Au Fermi level, resulting in a pathological overestimate of the conductance for this class of systems. In this work, we use many-electron perturbation theory within the GW approximation to compute quasiparticle energies of aromatic molecules covalently bonded to a gold surface, taking particular care to assess dynamical screening beyond standard plasmon-pole approximations. Our parameter-free approach captures important physics governing electronic level alignment, including charge transfer screening, dipole formation, and surface polarization. We discuss results for benzene on Au(111) bonded via amine (-NH2) and thiol (-SH ) link groups. These data are compared with more approximate model self-energy corrections applied to these systems [1], and also recent experiments.[1] S. Y. Quek et al, Nano Lett. 7, 3477 (2007)
Z7: Frontier of Electronics I
Session Chairs
Thursday PM, April 16, 2009
Room 2000 (Moscone West)
3:00 PM - **Z7.2
Modelling Charge Transport in Molecular Crystals, Polymers and Liquid Crystals.
Alessandro Troisi 1 , David Cheung 1 , David McMahon 1 , Denis Andrienko 2
1 Department of Chemistry, University of Warwick, Coventry United Kingdom, 2 , Max Planck Institute for Polymer Research, Mainz Germany
Show AbstractWe present recent advances in the computations of the absolute mobility of molecular semiconductors and the understanding of the underlying transport mechanism. We have shown in the past that the absolute value of the hole mobility can be computed for crystalline materials such as rubrene using a combination of classical molecular dynamics simulations and quantum chemical methods leading to an excellent agreement with the experiments. The predictive capabilities of this model are further validated by several recent experiments on THz spectroscopy and by new experimental determination of the tail of the density of states in organic crystals. We extended the model to the microscopic description of polymeric materials where the nuclear dynamics requires more sophisticated computations (both classical and quantum chemical). We show how the mobility in P3HT (one of the most common p-type organic semiconductor) is limited by long lived traps for the hole which originate from slightly distorted conformations of the conjugated back bone. We are therefore able to bridge the phenomenological theories of transport in disordered system with a model that describes the nature of the localized states in a detailed fashion. The versatility of the approach is further demonstrated by the study of the charge transport in semiconducting columnar liquid crystals, where with timescales spanning more than 6 orders of magnitudes need to be considered to achieve a satisfactory description of the quantum dynamics. Charge dynamics in the liquid crystalline environment is particularly challenging because it is an intermediate case between a system with only dynamic disorder (the crystal) and a system with mainly static disorder (the polymer). We will show how to develop a model which takes into account both types of disorder. The connection with more conventional theoretical methods and other solid state physics theories will be also is discussed. *This work is supported by EPSRC (UK)1) D. Cheung, A. Troisi, Modelling Charge Transport in Organic Semiconductors: from Quantum Dynamics to Soft Matter, PHYS CHEM CHEM PHYS 10, 5942, 20082) van Laarhoven HA, Flipse CFJ, Koeberg M, Bonn M, Hendry E, Orlandi G, Jurchescu OD, Palstra TTM, Troisi A, On the mechanism of charge transport in pentacene, J CHEM PHYS 129, 044704, 20083) Troisi A, Prediction of the absolute charge mobility of molecular semiconductors: the case of Rubrene, ADVANCED MATERIALS, 19, 2000-2004, 2007 4) Troisi A, Charge dynamics through pi-stacked arrays of conjugated molecules: effect of dynamic disorder in different transport/transfer regimes, MOLECULAR SIMULATION, 32 (9), 707-716, 20065) Troisi A, Orlandi G, Charge transport regime of crystalline organic semiconductors: diffusion limited by thermal off-diagonal electronic disorder, PHYS REV LETT 96 (8), 086601, 2006
3:30 PM - Z7.3
Numerical Formulation of the Effective Medium Approximation: Illustrative Examples and Application to Organic Semiconductors.
Peter Graf 1 , Muhammet Kose 1 , Kwiseon Kim 1
1 , National Renewable Energy Laboratory, Golden, Colorado, United States
Show AbstractThe effective medium approximation (EMA) is an approach to calculating the macroscopic properties (e.g. conductivity, mobility) of disordered materials from an average of known microscopic interactions. Logically, we can divide the approach into two parts. First, an expression based on the microscale physics of constituents embedded in the effective medium is derived. Second, a configurational average of some sort over “configurations” of the system is performed, leading to an equation for the effective parameter, which is then solved. Based on this division, we present a numerical approach to the EMA in which the effective parameter ξe is the solution of an equationg(ξe) = 〈f(σ, ξe)〉 = ∫ f(σ, ξe) ρ(σ) dσ = 0.Here, σ represents possible configurations of the system, i.e. some variable of the physical formulation with respect to which we have knowledge of both i) microscale interactions and ii) the distribution of constituents; f represents our knowledge of the microscale interactions as a function of σ; and ρ represents our knowledge of the distribution of the configurations/constituents with respect to σ. The equation says that the average of the microscale interactions embodied in f, which is a function of the effective parameter ξe, over the possible configurations of the system represented by σ, is zero, and this equation allows us to determine the effective parameter ξe, which, for example, we can compare to experiment, or use in a drift/diffusion type of macroscopic simulation. After presenting the basic formulation and simple illustrative examples, we will apply our EMA formulation to a realistic application of hopping transport in organic photovoltaics simulation. The physics of energy and carrier transport in organic semiconductors are not as well understood as those in their inorganic counterparts. It is a well known fact that the morphology of the organic medium plays an important role in the efficiency of organic solar cells, though there are few theoretical reports in the literature that deal specifically with exciton and/or carrier dynamics in amorphous phases of conjugated materials. The successful simulations of exciton/carrier dynamics are vital to reveal the limiting factors which are responsible for poor photovoltaic activity in most of the current organic photovoltaic devices. Our numerical EMA formulation provides a bridge between detailed quantum mechanical calculations and experimental results.
3:45 PM - **Z7.4
The Physics and Technology of Nanonet Electronics: A Bottom-up Perspective of Carrier Transport in Soft Material.
Muhammad Alam 1
1 , Purdue University, West Lafayette, Indiana, United States
Show AbstractAs the future of Moore’s law of transistor scaling appears uncertain, Electronics is trying to reinvent itself by broadening its focus to other areas including macroelectronics (electronics of large, possibly flexible and transparent displays), bioelectronics (e.g., nanobio sensors for geomomics, proteomics), and energy-harvesting (e.g., solar cells). In this regard, a material based on nanonets of Carbon Nanotubes or Si/ZnO/SiGe Nanowires have attracted considerable attention. The nanonets act as channel materials for thin-film transistors for flexible/transparent electronics, as sensor elements for label-free bio-sensors, and as transparent top electrode for solar cells. The performance of these Nanonet devices have been good (and sometimes impressive) and various laboratories have reported considerable improvements over the years. A lack of predictive transport models, however, has stymied the translation of laboratory experiments to practical, disruptive technology. The classical theory of bulk semiconductors, developed over last 50 years in close collaboration with experimentalists, device physicists, numerical analysts, and computer scientists, do not apply to these new electronic components with spatially inhomogeneous transport properties. In this talk, I will discuss a simple theory of the Nanonets based on 2D percolation, Cantor transform, and fractal dynamics to show how these simple/intuitive bottom-up views is challenging conventional wisdom and allowing optimization of nanonet transistors, biosensors, and solar-cells that would have been impossible even a few years ago..
Z8: Mesoscale Phenomena
Session Chairs
Thursday PM, April 16, 2009
Room 2000 (Moscone West)
4:30 PM - Z8.1
Discrete Dislocation Dynamics Simulation of the Anisotropic Mechanical Properties of Iron and Steel at High Temperature.
Steve Fitzgerald 1 , Sylvie Aubry 2 , Wei Cai 2 , Sergei Dudarev 1
1 Theory and Modelling, EURATOM/UKAEA Fusion Association, Abingdon, Oxfordshire, United Kingdom, 2 Dept. of Mechanical Engineering, Stanford University, Palo Alto, California, United States
Show AbstractSeveral important emerging power generation technologies, in particular nuclear fusion and advanced fission, require the development of high temperature structural materials capable of retaining their properties in an irradiation environment. The most promising candidate materials for these applications are reduced-activation ferritic-martensitic steels, due to their high strength, low activation and favourable swelling and embrittlement properties. However, low-alloy ferritic steels exhibit a catastrophic loss of strength above around 650C, (as was graphically demonstrated by the tragic collapse of the World Trade Center towers in 2001), limiting the achievable efficiency of future power plants. This unusual behaviour can be rationalised by considering the anisotropic elastic properties of iron. Whilst the trigonal shear modulus (C44) falls off linearly with temperature, to reach zero at the melting point, the tetragonal shear modulus (C’) approaches zero at the alpha-gamma phase transition, consistent with the displacive nature of the martensitic transformation. Hence anisotropic elastic effects cannot be neglected; indeed, this directional softening of a crystal could never be captured by any isotropic approximation. To develop a fast simulation method capable of describing the effect of elastic anisotropy on the plastic properties of iron and steels, the parallelized dislocation simulation program ParaDiS has been modified to incorporate anisotropic elastic forces acting on dislocation segments. We present the results of recent simulations, and illustrate the striking differences between the full anisotropic predictions and those of the isotropic approximation. These are compared with recent experimental results on the mechanical properties, together with electron microscope observations of the dislocation microstructures.
4:45 PM - Z8.2
Impact of the Surrounding Network on the Si-O Bond-Breakage Energetics.
Stanislav Tyaginov 1 , Viktor Sverdlov 2 , Wolfgang Goes 1 , Philipp Schwaha 2 , Rene Heinzl 2 , Franz Stimpfl 2 , Tibor Grasser 1
1 Christian Doppler Laboratory for TCAD at the Institute for Microelectronics , TU Wien, Vienna Austria, 2 Institute for Microelectronics , TU Wien, Vienna Austria
Show AbstractMcPherson’s Model (MM) [1] describes the rupture process for the Si-O bond, but considers only single SiO4 tetrahedron; however, it is obvious that consideration of the whole lattice substantially changes the situation. We show that the secondary energetic minimum occurs in a different direction compared to the one predicted within MM. Another, and more serious, conclusion is that the contribution of the whole crystal drastically changes the Si-O bond-breakage energetics.
Similar to the original MM, in our Extended McPherson Model (EMM) we employed the Mie-Grüneisen Potential (MGP). We take into account all possible pair-wise interactions between Si-O, Si-Si and O-O. Since MGP has 3 terms and each ion is characterized by its own “effective charge” one needs 6 independent constants to describe all types of interactions in silica. The convergence of the electrostatic energy leads to the condition of electrical neutrality of a primitive cell and the number of independent constants is reduced to 3. These parameters may be found by calibration to material properties (cohesion energy, elastic constants) and to DFT results. Finally we consider also well-accepted potentials constructed on the basis of DFT results: TTAM and BKS [2,3]. The potential acting from the rest of the crystal on a Si/O ion reveals an energy minimum at its equilibrium position and 3 constants are to be found in the manner to represent these minima (obtained with e.g. TTAM) and the cohesive energy for α-SiO2.
We examined different directions and found that the second saddle point occurs in another direction (compared to MM) for all cases: MGP, TTAM and BKS potentials. This secondary minimum reveals high activation energy ~6 eV for the transition of the Si ion (corresponding to bond-breakage). Such a high barrier reflects the contribution of the surrounding lattice, especially of the neighboring Si atoms. The breakage rate has been calculated for a “virgin” bond and for a bond weakened by Hole Capture (HC). In the last case the rate was found to be close (the difference is within 3-4 orders) to the one calculated within MM without HC.
Such a small discrepancy obtained for principally diverse situations is attributed to the effect of the surrounding lattice on the energetics of Si-O bond-breakage. A small breakage rate (of a “virgin” bond) suggests that the interaction of the electric field with a dipole moment can not provide a considerable breakage probability and such factors as bond distortion and/or energy delivered by particles must be considered as the essential components of bond rupture mechanism. We conclude that bond angle variations, strain, disorder and interactions with mobile hydrogen and hot carriers are inherent contributors to bond-breakage process.
1. J.W. McPherson, JAP, v. 99, No. 083501 (2006); 2. S. Tsuneyuki et al, PRL, v. 61, p. 869 (1988); 3. B.W.H. van Beest et al, PRL, v. 64, p. 1955 (1990).
5:00 PM - Z8.3
Theoretical Study of the Mobility of Perfect Dislocations in Silicon.
Laurent Pizzagalli 1 , Andreas Pedersen 2 , Andri Arnaldsson 2 , Hannes Jonsson 2 , Jean-Luc Demenet 1 , Jacques Rabier 1
1 , PhyMat - CNRS, Futuroscope Chasseneuil France, 2 Faculty of Sciences, University of Iceland, Reykjavik Iceland
Show AbstractThe knowledge of the mobility properties of dislocations is often the key to understand the plastic behavior of a given materials. Traditionally, dislocation mobility is investigated by experiments performed at the macroscale, and continuum theory such as elasticity. However, for 'hard' materials with a large lattice friction, such as covalent materials, the fine structure of the dislocation core and how it is modified during displacement is of prime importance. Unfortunately, even with recent developments in electron microscopy, it is still difficult to determine the core structure from experiments. The investigation of kinks or jogs on dislocations is even more challenging. Atomistic calculations are the appropriate answer for determining dislocation core properties. The information gained can then be used in larger scale approachs such as Kinetic Monte Carlo or Discrete Dislocation Dynamics, for a meaningful comparison with experiments. We have performed first-principles and interatomic potentials calculations of the properties of kinks on perfect dislocations in silicon. These dislocations are observed at low temperature in the brittle regime for many covalent materials, and are expected to play a role in the brittle-ductile transition. Using the Nudged Elastic Band and the Dimer methods, the atomistic mechanisms and associated activation energies for kinks formation and migration have been determined [1]. The role of high stress present in experiments has been considered. A special care has been taken for investigating the effect of an hydrostatic pressure on the stability and mobility of dislocations [2]. In fact, experiments performed at low temperature in the brittle regime are usually made with the help of a large confining pressure, in order to prevent the breaking of the sample. Finally, we have compared the mobility properties of dislocations at high and low temperature in silicon, showing that the effect of stress itself is not sufficient to explain the experimental results [3]. [1] L. Pizzagalli, A. Pedersen, A. Arnaldsson, H. Jonsson, and P. Beauchamp, Phys. Rev. B 77, 064106 (2008).[2] L. Pizzagalli, J.-L. Demenet, and J. Rabier, submitted to Phys. Rev. B (2008)[3] L. Pizzagalli and P. Beauchamp, Phil. Mag. Lett. 88, 421 (2008)
5:15 PM - Z8.4
Close-Range Dislocation Interactions and Plasticity in Small Volumes
Amine Benzerga 1
1 Aerospace, Texas A&M University, College Station, Texas, United States
Show AbstractDislocations are the main carriers of plastic deformation in crystalline materials. Their long-range interactions are accurately accounted for by elasticity. Close-range interactions are directly affected by atomic-level core properties. In this paper, we report on a new dislocation dynamics simulation paradigm applied to investigate the scale-dependence of mechanical properties at micro- and nano-scales in finite volumes. To account for atomistic input close-range interactions are formulated as additional rules parametrized based on fully atomistic simulations. The motion of dislocations, their interaction with obstacles, multiplication andannihilation are the elementary mechanisms that contributeto the phenomenology of plastic flow. In the limit of a large dislocation source density, we show that conventional forest hardening processes are active but the stress-strain response exhibits a clear size-dependence of the strain-hardening rate. The latter is rationalized in terms of evolution of densities of geometric dislocations among the various slip systems. In the other extreme of low dislocation source density, the simulations reveal a new regime of behavior where rare events can have a major impact on the stress-strain response. Various scalings of the flow stress with crystal size emerge in the simulations, which are compared with recent experimental data on nanopillars. Key structural variables and parameters affecting that scaling are identified and discussed.
5:30 PM - Z8.5
Search for New High Performance Piezoelectrics for Thin Film and Nano-Scale Applications.
Panchapakesan Ganesh 1 , Ronald Cohen 1
1 Geophysical Laboratory, Carnegie Institution of Wahsington, Washington, District of Columbia, United States
Show AbstractRecent theoretical simulations using density functional theory (DFT) and novel low temperature high energy x-ray diffraction experiments clearly show the existence of a high pressure morphtotropic phase boundary (MPB) in pure PbTiO3. The experiments show a richer phase diagram than the simulations, with multiple monoclinic phases (Pm and Cm) in the MPB region. Here we examine the MPB region in more detail using high precision DFT calculations within the local density approximation (LDA) and the Wu-Cohen generalized gradient approximation (WC-GGA). We explain why zone-boundary mode is more likely to be stable at higher pressures above ~ 25 GPa and not at moderate pressures of ~10GPa, using the LDA. Our results support the polarization rotation theory and open up fresh possibilities for applying chemical pressure to engineer novel electromechanical materials. In this regard we present results on Pb1/2Sn1/2TiO3, Pb1/2Ge1/2TiO3 and Sn1/2Ge1/2TiO3 that were obtained by substituting the ‘A’ site of ABO3 compounds by atoms with smaller ionic radii. Pb1/2Sn1/2TiO3 has a lattice parameter identical to SrTiO3 allowing one to make thin film devices with novel electromechanical properties.
5:45 PM - Z8.6
Atomistically Informed Continuum Modeling of Beta-solenoid Protein Structures.
Sinan Keten 1 , J. Fernando Rodriguez Alvarado 2 , Markus Buehler 1
1 Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States, 2 Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
Show AbstractBeta-solenoids are a new class of nanotube-like protein structures that are observed in virulence factors, prion proteins and amyloids. We report molecular dynamics (MD) simulation and atomistically informed continuum modeling of the nanomechanics of the beta-helix protein motif, a recently discovered protein building block forming a tube-like structure with a triangular core. We find that the beta- helix structure is extremely extensible and can sustain tensile deformation up to 800 % engineering strain without rupture of the covalently bonded protein backbone. Our atomistic simulation results reveal that the instantaneous tensile strength of the tube is proportional to the rate of H-bond rupture, providing a link between the dynamics of hydrogen bond rupture and the mechanical signature of the protein structure. This finding suggests that concurrent as opposed to sequential breaking of bonds leads to higher mechanical resistance, corroborating earlier results found in studies of beta-sheet protein domains. Inspired by the beta-helical domain of the needle-like cell puncture device of bacteriophage T4, we carry out compressive loading and bending simulations on the full atomistic structure of this domain, and show that this protein motif can withstand extremely large compressive loads, far exceeding the tensile strength of proteins. We systematically characterize the mechanical strength of this protein nanotube using molecular dynamics simulations over a wide range of deformation speeds. We illustrate that the failure strength of the molecule have power law dependence on deformation rate. We observe H-bond rupture initiation as the atomistic mechanism of instability corresponding to the peak load in the force extension curve. We show that the behavior of the protein in small compressive deformation can be approximated by a rate dependent linear elastic modulus, which can be used in context of continuum formulations for mechanical stability. Length-dependent mechanical deformation modes under compressive loading are summarized in a deformation map. Our work provides a link between the structure and functional properties of this beta-topology and illustrates a rigorous framework for bridging the gap between experimental and simulation time-scales for future compression studies on proteins. Our findings illustrate the potential of the beta-helix protein motif as an inspiration for nano-scale materials applications, ranging from stiff nanotubes to self-assembling peptide based fibers.
Z9: Poster Session: Computational Nanoscience
Session Chairs
Davide Donadio
Tianshu Li
Friday AM, April 17, 2009
Salon Level (Marriott)
9:00 PM - Z9.12
Determination of Surface Structure of Cleaved (001) USb2 Single Crystal Surface.
Shao-Ping Chen 1 , Marilyn Hawley 2 , Phil. Van Stockum 3 , Hari Manoharan 3 , Eric Bauer 4
1 Theoretical Division, Los Alamos National Laboratory, Los Alamos, New Mexico, United States, 2 Materials Science and Technology Division, Los Alamos National Laboratory, Los Alamos, New Mexico, United States, 3 Department of Physics and Stanford Institute for Materials and Energy Sciences, Stanford University, stanford, California, United States, 4 Mterials Pysics and Applications, Los Alamos National Laboratory, Los Alamos, New Mexico, United States
Show AbstractWe have achieved what we believe to be the first atomic resolution STM images for a uranium compound taken at room temperature. The a, b, and c lattice parameters in the images confirm that the USb2 crystals cleave on the (001) basal plane as expected. The a and b dimensions were equal, with the atoms arranged in a cubic pattern. Our calculations indicate a symmetric cut between Sb planes to be the most favorable cleavage plane and U atoms to be responsible for most of the DOS measured by STM. Some strange features observed in the STM will be discussed in conjunction with ab initio calculations.
9:00 PM - Z9.13
Phenol Cluster Adsorption onto Single-Walled Carbon Nanotube: a Quantum Mechanical Level Study.
Wenjie Fan 1 , Jun Zeng 2 3 , Ruiqin Zhang 1
1 Centre of Super-Diamond and Advanced Films (COSDAF) & Department of Physics and Materials Science, City University of Hongkong, Hong Kong China, 2 , MedChemSoft Solutions Pty Ltd, Melbourne, Victoria, Australia, 3 College of Chemistry, Sichuan University, Chengdu, Sichuan, China
Show Abstract Non-covalent functionalization is one of the key approaches to enable single-walled carbon nanotubes (SWCNTs) being applied in the areas of biosensors and drug delivery. In this work, we studied the functionalization of SWCNTs with peptides. The binding mechanism and the relative strength of peptides and SWCNTs were revealed at quantum mechanical level by theoretical simulating the adsorption of phenol clusters [(C6H5OH)n, n=1-3] to SWCNTs using a density-functional tight-binding (DFTB) method with an empirical van der Waals force correction. With the inclusion of weak interactions, our calculations showed that the phenol clusters could be spontaneously absorbed to the SWCNTs surface through π…π and H…π stacking at the physisorption distances. After adsorption of peptides on its sidewall, the geometric and electronic structures of SWCNTs are basically undamaged. The diameter dependence of binding energies was investigated, and the calculations demonstrated that, the larger π…π stacking, the larger binding energies will be. The adsorption spectrum and fluorescence spectrum were calculated to provide better understanding of the optical properties and guidance for the experiments. It is shown that the combinations of intra-molecular H-bonding, H…π stacking, and π…π stacking play a key role in binding of phenol to SWCNTs, thus determining and stabilizing the phenol clusters/CNT systems. The findings provide better understanding of the binding features between phenol clusters and SWCNTs and are helpful for designing bio-functionalized SWCNTs biosensors and drug delivery products.
9:00 PM - Z9.14
Onset Of Plasticity In Materials At The Nanometer Scale: The Role Of Surface Defects.
Pierre Hirel 1 , Julien Godet 1 , Sandrine Brochard 1 , Laurent Pizzagalli 1
1 , PhyMat - CNRS, Futuroscope Chasseneuil France
Show AbstractThe study of the response of a solid to a mechanical load is an important domain of materials science. Beyond a certain threshold, a plastic behavior occurs, that allows to relax the applied stresses. In bulk materials, the internal mechanisms associated with the onset of plasticity are well known, such as the multiplication of pre-existing dislocations for instance. However, in systems characterized by one or several dimensions at the nanometer scale (thin layers, nanowires, nanopillars,...), these mechanisms are generally inactive, asking for how and where plasticity occured in these materials. Since one major difference between bulk and nanomaterials is the dominating presence of surfaces/interfaces in the second case, it has been proposed that surfaces and defects could be at the origin of the onset of plasticity in systems with nanoscale dimensions. Unfortunately, due to the very small length and time scales involved, it is almost impossible to observe the onset of plasticity in experiments. Computational science then appears as the only alternative for characterizing and understanding the mechanims at play. We have investigated the onset of plasticity from surfaces in two different model materials, aluminum and silicon, by performing both molecular dynamics and Nudged Elastic Band calculations. In both cases, plasticity occured by formation of a half-loop dislocation from a surface defect. For aluminum, using the data from numerical simulations, we have developed an analytical model and determined the conditions (stress, temperature) initiating plasticity [1]. Silicon is a more complicated system, due to the existence of two bulk plastic regimes as a function of stress and temperature. Here we show that these two regimes can be initiated from surfaces [2], suggesting that both can occur in systems with nanometric dimensions, such as thin films of nanowires. [1] P. Hirel, J. Godet, S. Brochard, L. Pizzagalli and P. Beauchamp, Physical Review B 78, 064109, (2008).[2] J. Godet, P. Hirel, S. Brochard, and L. Pizzagalli, submitted to J. Appl. Phys. (2008).
9:00 PM - Z9.15
Atomic Structure of {001} Hydrogen Induced Platelets in Germanium From First-principles Calculations and High-Resolution Transmission Electron Microscopy.
Marie-Laure David 1 , Laurent Pizzagalli 1 , Frederic Pailloux 1 , Jean-Francois Barbot 1
1 , PhyMat - CNRS, Futuroscope Chasseneuil France
Show AbstractHydrogen is a common impurity in semiconductors such as silicon and germanium. Depending on the concentration, hydrogen can form defect complexes. At high concentrations, obtained by implantation or plasma exposure, extended planar defects lying along the {001} and/or {111} planes are formed: the so-called hydrogen induced platelets. These particular defects are known to be the precursors for layer splitting process used for the production of Silicon-On-Insulator wafers. Although these platelets have been largely investigated both experimentally and theoretically, many questions about their structure remain unanswered. In fact, it is not clear whether vacancies are required for forming platelets, and how hydrogen interact with the native atoms. The typical size of a platelet is few nm, large enough to be a tremendous challenge for first-principles calculations, while remaining in a scale hardly accessible to experiments. Combining calculations and experiments appears as the only relevant way to obtain meaningful results. Using results from first-principles calculations and observations made by High-Resolution Transmission Electron Microscopy (HRTEM), we propose a model for the structure of the {001} platelet in germanium. A large set of possible configurations have been considered in the calculations, including various numbers of vacancies and H2 molecules in the core of the defect. Low energy configurations have then been introduced into a larger model, which has been compared with HRTEM images (obtained with a JEOL 3010 microscope, 300 kV, LaB6, 0.19 nm point resolution). The most favorable configuration involves a dihydride passivation, the presence of vacancies, and a filling of the formed cavity with H2 molecules. The presence of dihydrogen molecules into the platelet leads to a dilation of the surrounding layers, with a calculated maximum amplitude of 0.6 Å. A geometrical phase shift analysis of the TEM contrast leads to a similar dilation value.
9:00 PM - Z9.16
Mixed Ab-initio Quantum Mechanical/Monte Carlo Calculations to Probe Electronic Correlations in Graphite and Graphene Nanostructures.
Simone Taioli 1 , Stefano Simonucci 3 , Lucia Calliari 1 , Massimiliano Filippi 1 , Maria Merlyne De Souza 2 , Maurizio Dapor 1
1 Computational and Theoretical physics, Center for Materials and Microsystems, Trento Italy, 3 Department of Physics, University of Camerino, Camerino Italy, 2 Electrical and Electronic Engineering, The University of Sheffield, Sheffield United Kingdom
Show AbstractA mixed Quantum Mechanical/Monte Carlo method (QMMC) for calculating non-radiative decay spectra in graphite and graphene nanostructures is presented.There are two main steps in this cluster approach. First, accurate full ab-initio calculations are performed to obtain the non-radiative decay rate.Second, a Monte Carlo treatment of the energy loss of the emitted electrons is superimposed to compare the theoretical spectra with `as-acquired' experimental data.To test this approach C K-VV Auger spectra of different carbon structures, such as graphite, nanotubes and graphene, have been calculated as a function of the emission angle and of the different geometric structures.In the Quantum Mechanical calculation we propose a new approach for calculating Auger lineshapes.In this method we select a number of transitions, which account for the quasi-atomic nature of the Auger emission, while long range correlation effects enter naturally in the Hamiltonian as a perturbation.Our method goes beyond the semi-empirical Cini-Sawatzky model, which has been used up to now to quantitatively interpret Auger spectra, by including correlations effects form first principles.In the second step, using the calculated line-shape as electron source, the Monte Carlo method is used to simulate the effect of inelastic losses on the original Auger line-shape.The resulting spectrum can be directly compared to `as-acquired' experimental spectra, thus avoiding background subtraction or deconvolution procedures.The results obtained in the peak energies, relative intensities and energy loss are in good agreement with the experimental data obtained in our group.This approach provides us with some theoretical insights on the curvature effects, on the electronic correlations and on the mechanisms of energy loss in strongly confined 1D systems and suspended graphene sheets.This method can be used as a predictive tool of spectral properties of carbon nanomaterials and to probe the interaction between carbon atomic layers.Preliminary results on the nonradiative recombination of excitons in carbon nanotubes will be presented which reflect the effect arising from the strong spatial confinment and the reduced screening in 1D systems.
9:00 PM - Z9.17
Simulation and Experiments of Fracture in Hybrid Organic-Inorganic Glasses
Mark Oliver 1 , Reinhold Dauskardt 1
1 Materials Science and Engineering, Stanford University, Stanford, California, United States
Show AbstractThin hybrid glass films in which organic and inorganic species are covalently bonded and intermixed at the molecular scale are an increasingly important class of materials for a variety of applications ranging from interlayer dielectrics for microelectronic devices to bonding of high-performance aircraft structures. Currently, a strong theoretical basis for understanding experimentally observed correlations between the composition, processing, and fracture properties of hybrid glasses has not been established, largely due to the complexity of their structures. The advancement of such an understanding requires both experimental and computational tools. Experimental methods exist for characterizing both the local molecular structure (a few nearest neighbors) and the fracture properties of thin hybrid glass films. Using these methods, we have recently gained new insights into the relationship between the molecular structure and fracture properties of hybrid glasses used as high-performance adhesive coupling layers. However, fracture in these materials is highly dependent upon the intermediate range (1-10 nm) structure and connectivity of the glass, which is not easily characterized in detail with experimental methods. Thus, we have developed a modeling approach to generate hybrid organic-inorganic glass networks via molecular dynamics simulations and to model the molecular-scale mechanisms of crack growth within these networks using the mathematics of graph theory. These simulations are not only useful for providing insights into the fundamental mechanisms responsible for experimentally observed behavior but potentially for also predicting how changes in the glass structure will influence fracture properties. We will discuss our efforts to understand the relationship between the molecular- and nano-scale structure of hybrid organic-inorganic glasses and their fracture properties by using a combination of experimental and computational methods.
9:00 PM - Z9.18
DFT-Based FEM Analysis on Extreme Elastic Deformation Behavior of Diamond Crystals.
Hajime Kimizuka 1 , Naohiro Toda 1 2 , Shigenobu Ogata 1
1 Department of Mechanical Science and Bioengineering, Graduate School of Engineering Science, Osaka University, Osaka Japan, 2 Super Hard Materials Development Department, Sumitomo Electric Hardmetal Corp., Hyogo Japan
Show AbstractWe developed a new framework of a finite-element method (FEM) analysis, with a nonlinear constitutive relation based on density functional theory (DFT), as an efficient method to characterize the nonlinear extreme elastic deformation of single-crystal diamond (SCD). Recent technological progress allows manufacturing a pure and perfect SCD for applications used in high-load apparatuses, such as an indenter and an anvil cell. Under high-load conditions, it is expected that a pure and perfect SCD can exhibit extreme elastic deformation, more than 20 % strain before permanent deformation starts, and the deformation behavior is considered to be highly nonlinear and anisotropic. Despite the importance of assessing structural integrity of such devices, there has not yet been a method to evaluate such a deformation behavior in a predictive manner.In the present method, the stress-strain relations are obtained during FEM analysis on the fly based on the DFT total-energy calculations and their numerical database is simultaneously constructed, which enables us to obtain high-precision stress without any empirical parameters even under finite strained conditions. The database significantly improves the total computational efficiency without loss of accuracy. We also carried out parallel-computing method for the database construction process to realize further improvement of computational efficiency. Once the stress-strain database is constructed, total computational cost of DFT-based FEM analysis is reduced by 99 % and above. Using our method, we conducted the shear-deformation tests of SCD under various stress conditions, to examine the external-stress dependence of its deformation characteristics. The present FEM calculations of extreme elastic deformation behavior of SCDs and predictions of their instabilities are validated by comparing with direct DFT calculations. The obtained stress-strain relation under (111)<11-2> shear deformation of an SCD exhibits a highly nonlinear behavior, and can reproduce the maximum shear stress. It is noted that the present results and the direct DFT results agree very well each other. Thus, our method has potential to estimate nonlinear elastic behavior and ideal strength of crystal under various external stress conditions. The effectiveness of the present method is demonstrated through numerical simulations for the uniaxial deformation of a SCD pillar model and also for the load-displacement response during indentation of SCD by various indenters. Finally, their results are compared to experiment and the strength and mechanical behaviors of SCDs under extreme stress conditions are examined.
9:00 PM - Z9.19
An Interactive Software Package for Simulating Nanoscale Patterning on a Solid Substrate.
Michael Wang 1
1 , Albuquerque Academy, Albuquerque, New Mexico, United States
Show AbstractWhen deposited over a solid surface, some chemicals form patterns at the nanoscale. Two major factors cause this pattern formation. The minimums in Gibb’s free energy drive the phase separation of the chemical components. This separation increases the amount surface free energy. To minimize its total energy, the system reacts by reducing the number of phase boundaries. On the other hand, the surface stress produced by concentration variations tends to create finer patterns by increasing the number of phase boundaries. These two opposing factors cause the system to reach equilibrium and form a stable pattern. This pattern formation is described by a set of nonlinear integral-differential diffusion equations that couple the concentrations and the surface stress (Lu, 2006). These equations can be simplified using the Fourier Transformation, which converts the integral-differential equations into a set of ordinary differential equations in Fourier space, which are solved using a semi-implicit method proposed by Chen and Shen (1998). The Fast Fourier Transform is used to link the real and Fourier spaces. A C# program has been developed to simulate this self-assembly process. In this program, the equations are modified to include the effects of temperature on this pattern formation. Model simulations agree qualitatively with the experimental results reported in literature, including the transitions between quantum dots, serpentine stripes, and quantum pits. We have shown that heterogeneous pattern formations can be guided by preexisting patterns. We have also shown that temperature can be used to control the size of patterns. This software can be used to understand and design nanoscale pattern formation on solid surfaces.Chen, L.-Q. and Shen, J., Applications of semi-impicit Fourierspectral method to phase field equations, Comp. Phys. Commun. 108 (1998) 147–158.Lu, W., Theory and Simulation of Nanoscale Self-Assembly on Substrates. J. Comp. Theoretical Nanoscience 3.3 (2006) 342-361.
9:00 PM - Z9.2
Morphological Stability of Epitaxial Core-shell Nanostructures: Nanowires and Nanoparticles.
Moneesh Upmanyu 1 , Hailong Wang 1 , Cristian Ciobanu 1
1 Engineering Division, Materials Science Program, Colorado School of Mines, Golden, Colorado, United States
Show AbstractMorphological instability of misfitting epilayers is a well known phenomenon in thin films. Motivated by the various shell morphologies reported recently in semiconducting core-shell nanowires, we analyze the morphological stability against azimuthal, axial, and general helical perturbations for epitaxial core shell nanowires in the growth regimes limited by either surface diffusion or evaporation-condensation surface kinetics. For both regimes, we find that geometric parameters (i.e., core radius and shell thickness) play a central role in determining whether the nanowire remains cylindrical or its shell breaks up into epitaxial islands, similar to those observed during Stranski-Krastanow growth in thin epilayers. The combination of small cores and rapid growth of the shell emerge as key ingredients for stable shell growth. Our results provide an explanation for the different core-shell morphologies reported in the Si-Ge system experimentally, and also identify a growth-induced intrinsic mechanism for helical nanowire morphologies. In conclusion, we discuss extensions to core-shell nanoparticles, and highlight computational strategies aimed at incorporating additional nanoscale effects, including but not limited to core faceting and surface stresses.
9:00 PM - Z9.20
Polyanion Conduction Mechanism in Solid Scandium Tungstate.
Zhou Yongkai 1 , Stefan Adams 1 , Arkady Neiman 2
1 Mater. Science and Eng. , Nat. University of Singapore, Singapore Singapore, 2 Chemical Department, Ural State University, Ekaterinburg Russian Federation
Show AbstractIn our recent work we could by combined computational, electrochemical and diffraction studies identify WO42- anions (and not as previously reported in literature Sc3+) as the mobile species in the solid electrolyte scandium tungstate, Sc2(WO4)3. Conduction by polyatomic anions is otherwise rare in solid state ionics (except for diatomic OH-). The validity of our forcefield is verified by successful reproduction of the negative thermal expansion over a limited temperature range, as well as of a pressure-dependent orthorhombic to monoclinic structural phase transition Sc2(WO4)3. Though the phase transition is observed in our simulations only when the pressure exceeds 0.8 GPa (compared to the experimental value of 0.25 GPa), the variation of lattice constants and the monoclinic angle α with pressure exhibit similar trends.In our Molecular Dynamics (MD) simulations for initially defect-free Sc2(WO4)3 structure models the diffusion of WO42- groups follows a unique correlated mechanism, which is triggered by a rare high energy step: the generation of a tungstate Frenkel defect. Consequently the simulated activation energy is significantly higher than the experimentally observed value. It may be presumed that the ion transport in scandium tungstate is dependent on the concentration of defects that are produced during the high temperature synthesis of the material. In order to investigate details of the WO42- conduction mechanism, here we report MD simulations of structure models with artificially induced defects, e.g. WO42- vacancy, Frenkel defect and Schottky defect, using the same modified version of the Universal Forcefield. All these simulations produce lower activation energy compared to the initially defect-free model. Simulations of Frenkel defect structures show very low activation energy but the interstitial WO42- initially has a strong preference to return to the vacant WO42- site. Vacancy and Schottky defect models both reproduce the activation energy in experimental conductivity studies more closely. Even though the Schottky defect model includes the creation of Sc3+ vacancies, no Sc3+ diffusion is observed in our simulations. Considering the high formation energy of Sc vacancies, the concentration of Schottky defects will however be very low in Sc2(WO4)3, so that extrinsic WO42- vacancies should be more relevant for the conduction mechanism. Nonstoichiometric samples with varying initial Sc2O3:WO3 ratios (Sc2O3)(WO3)3+x, where -0.1 < x < 0.1, are synthesised and characterised by XRD and impedance measurements. While the samples with the lowest WO3 contents contain a second phase Sc6WO12, the most samples appear to be single phased according to powder XRD. Impedance data suggest that the conductivity of Sc2(WO4)3 increases with defect concentration, both for samples with enhanced concentrations of tungstate vacancies (x<0) and interstitial (x>0). Possibilities to enhance the conductivity by nanostructuring are currently explored.
9:00 PM - Z9.21
Accurate ab initio Calculations of Methane Dimer Interaction Energies and Molecular Dynamics Simulation of Fluid Methane.
Arvin Huang-Te Li 1 , Sheng Der Chao 1
1 Institute of Applied Mechanics, National Taiwan University , Taipei Taiwan
Show AbstractIntermolecular interaction potentials of the methane dimers have been calculated for 12 symmetric conformations using the Hartree-Fock (HF) self-consistent theory, the second-order Møller-Plesset (MP2) perturbation theory, the coupled-cluster with single and double and perturbative triple excitations (CCSD(T)) theory. The HF calculations yield unbound potentials largely due to the exchange-repulsion interaction. In MP2 and CCSD(T) calculations, the basis set effects on the repulsion exponent, the equilibrium bond length, the binding energy, and the asymptotic behavior of the calculated intermolecular potentials have been thoroughly studied. We have employed basis sets from the Slater-type orbitals fitted with Gaussian functions (STO-nG, n=3~6), Pople’s medium size basis sets [up to 6-311++G(3df, 3pd)] to Dunning’s correlation consistent basis sets (cc-pVXZ and aug-cc-pVXZ, X=D, T ). With increasing basis size, the repulsion exponent and the equilibrium bond length converge at the 6-31G** basis set and the 6-311++G(2d, 2p) basis set, respectively, while a large basis set (aug-cc-pVTZ) is required to converge the binding energy at a chemical accuracy (~0.01 kcal/mol). Both the basis set superposition error (BSSE) corrected and uncorrected results are presented to emphasize the importance of including such corrections. Only the BSSE corrected results systematically converge to the destined potential curve with increasing basis size. The binding energy calculated and the equilibrium bond length using the CCSD(T) method are close to the results at the basis set limit. For molecular dynamics simulation, a 4-site potential model with sites located at the hydrogen atoms was used to fit the ab initio potential data. This model stems from a hydrogen-hydrogen repulsion mechanism to explain the stability of the dimer structure. MD simulations using the ab initio PES show good agreement on both the atom-wise radial distribution functions and the self-diffusion coefficients over a wide range of experimental conditions.
9:00 PM - Z9.23
A Density Functional Theory Investigation of Polymorphism in ZnO Thin Films.
Benjamin Morgan 1
1 Chemistry, Trinity College, Dublin, Co. Dublin, Ireland
Show AbstractZnO is an intrinsic n-type transparent conducting oxide1 and there is much interest in the synthesis and growth of thin films of this material. Bulk ZnO adopts the wurtzite (B4) structure, and is thus dipolar. Samples are commonly observed with exposed polar {0001} faces, despite these being inherently unstable. While there has been much debate about the stabilisation mechanism of such polar orientations, there is strong evidence that in macroscopic systems stabilisation is achieved via the presence of non-stoichiometric islands at the surface2,3. For nanoscale thin films alternative stabilisation mechanisms are possible which involve a polymorphic relaxation of the entire system. It has previously been proposed that stoichiometric B4 thin films below a threshold thickness are unstable with respect to a hexagonal-planar layered structure4 (h-MgO); a structural analogue of hexagonal BN; and that such thin films are the precursors to {0001}-terminated wurtzite films. Thicker slabs are predicted to stabilise with the B4 structure via charge transfer between the two polar surfaces making these metallic. Although the h-MgO planar structure is non-polar, the trigonal-planar coordination displayed by the ions is non-optimal. An alternative structure – d-BCT – is non-polar and displays near-tetrahedral coordination. This structure constitutes an extended region with the IDB* grain boundary motif5,6 and has been observed to form spontaneously in previous simulations of nanocrystals7,8,9 (which are B3/B4 in the bulk), as well as having been predicted as a metastable polymorph in bulk LiF10. The first principles calculations presented here indicate that thin films with the d-BCT structure are more stable than h-MgO (which is metastable) at all thicknesses, and display stability relative to B4 over a broader range of thickness. Nudged elastic band calculations have been used to examine the pathways between these polymorphs, and indicate that the h-MgO → d-BCT transition is spontaneous, while d-BCT → B4 interconversion for thicker slabs has a small activation barrier. 1 Janotti, A., Van de Walle, C. G.; Phys. Rev. B 76, 165202 (2007), and references therein.2 Ostendorf, F., Torbrügge, S., Reichling, M.; Phys. Rev. B 77, 041405(R) (2008).3 Kresse, G., Dulub, O., Diebold, U. Phys. Rev. B 68, 245409 (2003).4 Freeman, C. L., Claeyssens, F., Allan, N. L., Harding, J. D.; Phys. Rev. Lett. 96, 066102 (2006).5 Northrup, J. E., Neugebauer, J., Romano, L. T.; Phys. Rev. Lett. 77, 103—106 (1996).6 Liu, Y. Z. et al.; Phil. Mag. Lett. 87, 687—693 (2007).7 Morgan, B. J., Madden, P. A.; Phys. Chem. Chem. Phys. 9, 2355—2361 (2007).8 Hamad, A., Catlow, C. R. A.; J. Crys. Grow. 294, 2—8 (2006).9 Morgan, B. J.; Phys. Rev. B 78, 024110, (2008).10 Doll, K., Schön, J. C., Jansen, M.; Phys. Chem. Chem. Phys. 9, 6128—6133 (2007).
9:00 PM - Z9.24
Resolving Strain Mystery in Silicon Nanocrystals
Dundar Yilmaz 1 , Ceyhun Bulutay 1 , Tahir Cagin 2
1 Physics, Bilkent University, Ankara Turkey, 2 Chemical Engineering, Texas A&M University, College Station, Texas, United States
Show AbstractAtomistic strain in silicon nanocrystals (Si-nc) embedded in an oxide matrix is an active field of research. However, there are many controversial interpretations of experimental data. For instance, there is no consensus even on the type of strain whether it is tensile or compressive. Strain is effective on the optical activity even though direct measurement of it has not been reported yet. On the other hand, with the use of a realistic force field, results of a molecular dynamics simulation may provide more valuable information than the best available TEM image. In the past, there have been theoretical attempts to understand atomistic strain in Si-nc in which the chemistry of the environment was poorly characterized with use of inadequate, simplistic models. Such potentials share the drawback of storing a static bond list instead of simulating chemical reactions. To overcome this disadvantage, we use recently developed reactive force field model by van Duin et al as our simulation environment. In our previous work we reported Si-Si bond length stretching up to 3% just under the surface of nc, while Si-Si bonds at the core region of nc were not stretched. Following this path, to have a deeper understanding, we compute atomistic strain tensor and volumetric strain. Unlikely in our previous findings, atomistic strain calculations indicates the compressive nature of the strain especially at the region just below nc surface. Moreover, our results show that bond length stretching and compressive strain can occur at the same time at the same system. This result seems contradictory at first glance, however our deeper analysis of oxidation effects on bond topology resolves this mystery. To investigate Si-Nc/oxide interface, realistic definition of Si-NC surface is vital. This is achieved by the Delaunay Tessellation technique. With this realistic surface definition, the compressive strain is understood to be a result of distortion of tetrahedrons formed by Si atoms bonded to surface Si atoms in such a way to increase of solid angle subtended by tetrahedral face which is oriented to the surface. Whereas, solid angles subtended by tetrahedral faces which are oriented toward the core region are decreased as a result of oxidation. This orientation dependence of solid angles explains how the oxidation affects the strain distribution. Our detailed study on the atomistic strain resolves the strain dilemma which plagued the coherent interpretation of the experimental data.
9:00 PM - Z9.26
Size Dependent Elastic Properties of Graphene.
Reddy Damodara 1 , Shenoy Vivek 2 , Yong-Wei Zhang 1 3
1 Large-Scale Complex Systems, Institute of High Performance Computing, Singapore Singapore, 2 Dvision of Engineering, Brown University, Rhode Island, New York, United States, 3 Department of Materials Science and Engineering, National University, Singapore Singapore
Show AbstractIn last few years, the graphene has attracted significant attention due to its excellent novel electronic properties and its potential to make nano-electronic devices. Previous study has shown that the size and shape of graphene sheets significantly influence charge transport properties. An important aspect is how to describe its edge stress state and how the edge stress state affects the morphology of graphene sheets. In this work, we used both molecular dynamics method and continuum method to show that the elastic properties of finite graphene are distinctively different from infinite graphene. An energetic model was proposed to include its edge effect, in which edge stress and edge elastic modulus were naturally introduced. Molecular dynamics simulation with the AIREBO potential was used to compute the variation of the potential energy of a graphene sheet at different strain levels and widths. Different types of edge structures, namely, zigzag, armchair, zigzag and armchair edges with hydrogen, and reconstructed zigzag and armchair edges were considered. The edge elastic constants and edge stress were extracted from the energy calculations. It was found that the edge effects are dominant at smaller widths. The edges are in compression when the edge bonds terminated or bonded with hydrogen atoms while the edges are in tension when the edges are reconstructed.
9:00 PM - Z9.27
Effects of Hydrogen Chemisorption on the Structure of Carbon Nanotubes.
Andre Muniz 1 , Tejinder Singh 1 , Michael Behr 2 , Eray Aydil 2 , Dimitrios Maroudas 1
1 Department of Chemical Engineering, University of Massachusetts at Amherst, Amherst, Massachusetts, United States, 2 Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minnesota, United States
Show AbstractThe exposure of carbon nanotubes (CNTs), either single- or multi-walled (SWCNTs or MWCNTs), to hydrogen plasmas has been a subject of great interest with various practical applications ranging from hydrogen storage to synthesis of nanocrystalline carbon. In this presentation, we report results of a computational atomic-scale analysis of the effects of atomic hydrogen chemisorption on the structure of SWCNTs and MWCNTs, using classical molecular-dynamics (MD) simulations and first-principles DFT calculations. The Adaptive Interatomic Reactive Empirical Bond Order (AIREBO) potential is employed in the MD simulations of H-CNT interactions and the resulting structural relaxations. The DFT calculations are performed within the generalized gradient approximation and employ plane-wave basis sets, ultrasoft pseudopotentials, and supercell models.We find that H chemisorption onto SWCNTs affects considerably their structure, leading to deformation and amorphization of the SWCNT graphene wall. These changes cause “swelling” of the nanotube, consistently with experimental observations. The corresponding radial and axial strains depend on the H coverage, temperature, SWCNT diameter, and SWCNT chirality (for small nanotube diameters). Most importantly, we find that there is a critical H coverage (around 20-25%) beyond which, the radial and axial SWCNT strains start increasing linearly with H coverage, and sp3-hybridized carbon atoms prevail; at sub-critical coverages, the strain levels are negligible and sp2-hybridized C atoms dominate. The number of C atoms that form three bonds, but are bonded to sp3-hybridized atoms, increases with the degree of hydrogenation, going through a maximum at intermediate H coverage. These atoms have a dangling bond, which can provide a site for nucleation of nanocrystalline carbon phases (e.g., cubic and hexagonal diamond), which has been observed experimentally by exposure of MWCNTs to hydrogen plasmas.We also investigated the possibility of formation of inter-shell sp3 C–C bonds in MWCNTs induced by atomic hydrogen. These localized inter-shell C–C bonds, formed through reactions of hydrogen atoms with the carbon structure, can act as a seed for nucleation of crystalline phases embedded into the MWCNTs. We report results of a comprehensive protocol of DFT and MD computations, which show that the resulting structures containing these inter-shell C-C bonds are stable and that seeds for the nucleation of different carbon phases (allotropes) can be formed, depending on the alignment between the concentric graphene walls of MWCNTs.
9:00 PM - Z9.28
Ab initio Determination of Intrinsic Diffusion Coefficients in Uranium Dioxide.
Marjorie Bertolus 1 , Boris Dorado 1 , Michel Freyss 1 , Julien Durinck 1 , Serge Maillard 1 , Philippe Garcia 1
1 , CEA, DEN, DEC/SESC, Centre de Cadarache, Saint-Paul-lez-Durance France
Show AbstractAtomic transport in nuclear ceramics is central to unravelling the complexities associated with irradiation driven micro-structural changes in these materials. These changes in turn control many basic material properties such as oxidation, actinide or fission product redistribution, fuel swelling and creep. Numerical modelling, in particular ab initio calculations of the electronic structure or classical molecular dynamics are fast becoming invaluable tools for studying these transport phenomena on an atomic scale. Ab initio methods are now widely used for calculating defect formation energies and are starting to be used for estimating migration energies and identifying migration paths. Very often however, atomistic modelling stumbles upon the difficulty of validating calculation results or hypotheses against reliable experimental data. Generally, it is difficult to experimentally determine reliable basic physical values which can be directly used to assess their theoretical equivalent. Calculated values such as defect formation energies are assessed against experimental data generally inferred from the analysis of transport experiments. An alternative strategy is to develop a theoretical approach based on point defect models which uses both defect formation and migration energies to infer atomic diffusion coefficients. We present here the application of such a strategy to the modelling of intrinsic diffusion coefficients in uranium dioxide using the energies yielded by density functional theory. In particular, we determine oxygen diffusion coefficients as a function of temperature and deviation from stoichiometry and assess calculation results against a set of available experimental data. One of the more novel features included in this methodology is an attempt to estimate the entropic contributions to atomic migration.
9:00 PM - Z9.3
Atomic Ordering in Nano-layered FePt: Multiscale Monte Carlo Simulation.
Rafal Kozubski 1 , Miroslaw Kozlowski 1 , Jan Wrobel 2 , Tomasz Wejrzanowski 2 , Krzysztof Kurzydlowski 2 , Christine Goyhenex 3 , Veronique Pierron-Bohnes 3 , Marcus Rennhofer 4 , Savko Malinov 5
1 Interdisciplinary Centre for Materials Modelling, M.Smoluchowski Institute of Physics, Jagellonian University, Krakow Poland, 2 Interdisciplinary Centre for Materials Modelling, Faculty of Materials Science and Engineering, Warsaw University of Technology, Warsaw Poland, 3 , IPCMS, Strasbourg France, 4 Faculty of Physics, University Vienna, Vienna Austria, 5 School of Mechanical and Aerospace Engineering, Queen's University Belfast, Belfast United Kingdom
Show AbstractCombined nano- and mesoscale simulation of chemical ordering kinetics in nano-layered L10 AB binary intermetallics was performed. In the nano- (atomistic) scale Monte Carlo (MC) technique implemented with vacancy mechanism of atomic migration and diverse models for the system energetics were used. The meso-scale microstructure evolution was, in turn, modelled by means of a Monte Carlo procedure simulating antiphase-domain-boundary motion as controlled by antiphase-boundary energies evaluated within the nano-scale simulations. The study addressed FePt thin layers considered as a material for ultra-high-density magnetic storage media and revealed metastability of the L10 c-variant superstructure with monoatomic planes parallel to the layer surface and off-plane easy magnetization. The layers, originally perfectly ordered in a c-variant of the L10 superstructure, showed homogeneous disordering running in parallel with a spontaneous re-orientation of the monoatomic planes leading to a mosaic microstructure composed of a- and b-L10-variant domains. The domains nucleated heterogeneously on the free surface of the layer and grew discontinuously inwards its volume. Finally, the domains relaxed towards an equilibrium microstructure of the system. Two “atomistic-scale” processes: (i) homogeneous disordering and (ii) nucleation of the a- and b-L10-variant domains showed characteristic time scales. The same was observed for the meso-scale processes: (i) heterogeneous L10-variant domain growth and (ii) domain microstructure relaxation. The above complex structural evolution modelled by means of the multiscale Monte Carlo simulations has recently been observed experimentally in epitaxially deposited thin films of FePt.Work supported by the Polish Ministry of Science and Higher Education (Grant no. COST/202/2006). Financial support granted by the governments of France and Poland within the POLONIUM programme is greatly acknowledged. Two of the authors (R.K. and S.M.) collaborated within the International Fellowship granted to R.K. by Queen’s University, Belfast, UK. Calculations were carried out at PW ICM Warsaw, computational grant G31-5.
9:00 PM - Z9.30
Structural and Energetic Behavior of Pd-Pt Nanoalloy Clusters with 98 Atoms: A Computational Study
Alvaro Posada-Amarillas 1 , Lauro Paz-Borbon 2 , Thomas Mortimer-Jones 2 , Roy Johnston 2 , Giovanni Barcaro 3 , Alessandro Fortunelli 3
1 Dept de Investigación en Física, Universidad de Sonora, Hermosillo, Sonora Mexico, 2 School of Chemistry, University of Birmingham, Birmingham United Kingdom, 3 Istituto per i Processi Chimico-Fisici, Consiglio Nazionale delle Ricerche, Pisa Italy
Show AbstractSeveral computational techniques were implemented to study the structural and energetic behavior of 98-atom bimetallic clusters: A shell optimization routine, a Genetic Algorithm technique, and a Basin-Hopping atom-exchange routine. The interatomic interactions are modeled using the many-body Gupta potential, which is based on the second moment approximation to a Tight-Binding Hamiltonian. For most compositions, the presumed global minima present structures based on defective Marks Decahedra (MD). The most stable structure is the Leary Tetrahedron (LT) for compositions ranging from Pd46Pt52 to Pd63Pt35. According to the excess energy stability criterion, the Pd56Pt42 cluster is the most stable across the entire composition range. The main features of these nanoalloys are shown by the analysis of the binding energy and the corresponding structural details.
9:00 PM - Z9.31
Electronic Structure Characterization of Size-expanded DNA Bases and the Effect of Substituents.
Alvaro Vazquez-Mayagoita 2 , Oscar Huertas 3 , Giorgia Brancolini 5 , Modesto Orozco 4 , Bobby Sumpter 1 , F. Javier Luque 3 , Rosa Di Felice 5 , Miguel Fuentes-Cabrera 1
2 Department of Chemistry, University of Tennessee, Knoxville, Tennessee, United States, 3 Departament de Fisicoquimica, Univesitat de Barcelona, Barcelona Spain, 5 Istituto Nazionale per la Fisica della Materia, University of Modena and Reggi o Emilia, Modena Italy, 4 Institut de Recerca Biomedica, Parc Cientific de Barcelona, Barcelona Spain, 1 Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee, United States
Show AbstractNaptho and benzo-homologated DNA bases have been recently used to build a new type of size-expanded DNAs known as x-, y- and yy-DNA. We have used ab initio techniques to investigate the structure, tautomeric preferences, base-pairing ability, and HOMO-LUMO gap of these bases and the effects of substituents such as Cl, F, OH and Se. For the naptho-bases, the HOMO-LUMO gap is smaller than that of their benzo-counterparts, indicating that size-expansion of DNA bases is an efficient way of reducing this energy difference. Substituents tend to modify the electronic structure and the interactions between base pairs. Our theoretical findings and the available experimental data support the suitability of exploring the limits of size-expanded DNA bases, including chemical modifications, so as to create new DNA structures and nanowires with predetermined properties.Acknowledgment:Work at Oak Ridge National Laboratory (ORNL) was supported by the Center for Nanophase Materials Sciences, sponsored by the Division of Scientific User Facilities, U.S. Department of Energy (USDOE) and used resources of the National Center for Computational Sciences, ORNL, supported by the Office of Science, USDOE.
9:00 PM - Z9.32
Diameter Dependence of Aligned Growth of Carbon Nanotubes on a-Plane Sapphire Substrates
Alexander Badmaev 1 , Koungmin Ryu 1 , Xiaolei Liu 1 , Song Han 1 , Chongwu Zhou 1
1 Electrical Engineering, University of Southern California, LA, California, United States
Show AbstractAligned carbon nanotubes have great potential for advanced nanotube transistors and integrated circuits. In this article, we studied the carbon nanotube alignment mechanism using a chemical vapor deposition growth on a-plane sapphire substrates. We synthesized carbon nanotubes of different diameters by controlling the catalyst size and observed that nanotubes of smaller diameters had a higher degree of alignment. In addition, a surprising observation was that misaligned nanotubes had a preferred orientation. Furthermore, we developed a numerical simulation method to calculate interaction energy between a-plane sapphire surface and carbon nanotubes of different diameters. The calculated results were in good agreement with our experimental observations, which confirmed the observed diameter-dependent alignment and the preferred orientation for misaligned nanotubes.
9:00 PM - Z9.33
Modeling DNA Functionalized Nanoparticles
One-Sun Lee 1 , George Schatz 1
1 Chemistry , Northwestern University, Evanston, Illinois, United States
Show AbstractDNA-functionalized gold nanoparticles (DNA-NPs) have unique optical, thermodynamic and structural properties. A wide range of DNA-functionalized gold nanoparticles have been used in applications to DNA and protein detection. DNA-NPs also can be used as elemental building blocks in materials synthesis. However, the atomic structure of DNA-NPs is not known, and this has proven to be a hindrance in understanding how this system functions. Here, we report molecular dynamics simulations of DNA-NPs at the atomistic level. The model of single DNA-NP is developed, and the interaction between DNA-NPs is scrutinized. The effective radius of DNA-NPs and the salt concentration around the DNA-NP is studied according to the interaction distance. Our model shows a comparable result with the previous experiments and simulations.
9:00 PM - Z9.34
Peak/Plateau Strength in Nanoscale Multilayer Thin Films: Constrained vs Unconstrained Dislocation Nucleation
Qizhen Li 1 , Peter Anderson 2
1 , University of Nevada, Reno, Reno, Nevada, United States, 2 , The Ohio State University, Columbus, Ohio, United States
Show AbstractThe peak/plateau strength of multilayer thin films is analyzed in terms of the stress to bow out a dislocation loop from an interface. Comparison of approximate analytic models to experimental data suggests that the bow out is “constrained” by nearby interfaces, at least for e-NbN/Mo and e-Ni/Cu films. Estimates of the interfacial pinning distance to form the bow out range ~20b for e-NbN/Mo and ~70b for e-Ni/Cu, around peak/plateau strength.
9:00 PM - Z9.35
Experiments and modeling of clay-interlayer-polymer interactions in Polymer Clay Nanocomposites : A multiscale modeling Approach.
Kalpana Katti 1 , Dinesh Katti 1 , Debashis Sikdar 1 , Priyanthi Amarasinghe 1 , Shashindra Pradhan 1
1 , North Dakota State University, Fargo, North Dakota, United States
Show AbstractHere we present a multiscale approach to modeling polymer clay nanocomposites. A combination of steered molecular dynamics and experiments in spectroscopy and nanoindentation provide input to development of 3D finite element models of polymer clay nanocomposites. These models are based on our recently proposed ‘altered phase model’ of nanocomposites. In this model steered molecular dynamics, FTIR photoacoustic spectroscopy and nanoindentation experiments in addition to polymer characterization studies indicate that the polymer phase is significantly altered by the nanoscale dispersion of nanoclays as well as the clay structure is influenced by the polymers. Our multiscale models quantitatively describe the weak non bonded interactions between the various constituents of polymer clay nanocomposites. The new altered phase model, as well as the multiscale modeling techniques described, are applicable to not just other polymer clay nanocomposite systems but also particulate nanocomposites in general. Other applications of clays include barrier behavior that is exploited for geoenvironmental applications where clay layers are used as barriers for environmental toxins in landfills. We will present results from FTIR experiments that indicate that the molecular interactions of montmorillonite (MMT) with organic fluids and water. Molecular dynamics simulations are done with clays at different water content. Interaction energies are calculated for clay layers, sodium ions and intercalated water to evaluate the effect of non-bonded interactions on swelling properties. Further, the rate of intercalation inside clay gallery was studied using a clay model solvated in a box of water. Simulations were run for up to 4 ns to obtain a quantitative measure of water intercalation inside clay galleries with respect to time and also the variation of montmorillonite d-spacing with time. These studies describe synergetic treatment of simulations and experiments to exploit applications and design with nanoclays.
9:00 PM - Z9.37
Parallel Polarizability of Metallic Carbon Nanotubes.
Trinh Vo 1 , Paul von Allmen 1 , Anupama Kaul 1
1 NASA-Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, United States
Show AbstractThe polarizability of carbon nanotubes (CNT) is essential to the concept of recently proposed electronic devices such as memory switches and for the electrophoretic migration used in purification processes. The polarizability of CNT is anisotropic with the component parallel to the tube larger than the perpendicular component. Published electronic structure calculations have established the scaling properties of the parallel and perpendicular polarizability for semiconducting tubes and for perpendicular polarizability for metallic tubes. We will present a simple formula for the parallel polarizability of metallic tubes obtained by fitting electronic structure calculations to an electrostatic model of the CNT. A tight-binding method and first principles density functional calculations are used to obtain the parallel polarizability as a function of CNT length and radius. A boundary element solver gives the polarizability of a hollow metallic cylinder with the thickness of the walls tuned to fit the polarizability obtained from the electronic structure calculations. The electrostatic model is then used to produce a simple formula that gives the polarizability as a function of the CNT radius and length.
9:00 PM - Z9.38
Exploring Basics of Atomic Force Microscopy with the Computer Simulator.
Sergey Belikov 1 , Sergei Magonov 2
1 , MikroMasch USA, Santa Barbara, California, United States, 2 , Agilent Technologies, Chandler, Arizona, United States
Show AbstractQuality of AFM results depends on characteristics of the probe, dynamics of the cantilever, and geometry of the tip and its apex. Essential issues of AFM experiments in different modes (quasistatic contact mode, dynamical amplitude modulation (AM) and frequency modulation modes) can be simulated by equations of motion of the probe and assuming a particular type of the tip-sample force interactions [1]. A computer program was developed to simulate the AFM probe and its motion in AM mode based on Krylov-Bogoljubov-Mitropolsky asymptotic method. The AFM simulator evaluated a crystallographic model of the bc plane on polydiacetylene crystal [2], spherical particles in various nano-scale arrangements (from single lying objects to the closed-packed ordered in a hexagonal lattice), linear structures with a pitch along the main direction and heterogeneous systems composed of the components with different stiffness. A substrate for these structures was made with a large hard sphere placed below the structures so that the top sphere surface was a practically flat surface. The simulated probe was oscillated with free amplitude of 20nm and set-point amplitudes of 20nm, 19nm and 10nm. Therefore, the simulations were performed at different force levels starting with the “zero-force” level. In most experiments the probe with stiffness of 1N/m was applied. The tip is modeled by a sphere. The tip with radius of 150pm represents atomically-sharp probe (the radius of Si atom is 132pm), which is most likely not realistic to make and to use. The tips with radius of 1nm, 5nm and 10nm represent the commercial Si probes. The modeling of AFM with the test structures revealed the changes of the images, which are related to variations of tip size and the tip-sample forces. In the simulated images we analyzed a particular pattern and changes in lateral dimensions and vertical corrugations. This is important for understanding the relationship between the images and true objects and the ability of AFM microscopes, which have S/N limitations, to detect such image variations. The comparison of the simulations with the high-resolution AFM images of similar objects provides a strong confidence in this approach that can be essential for the optimal choice of the probe and imaging routines. Such simulations have unique potential to lead the AFM design and control algorithms, and increase rigorousness and efficiency of educational training both theoretical and practical.REFERENCES 1. Belikov S., Magonov S. “Classification of Dynamic AFM Control Modes Based on Asymptotic Nonlinear Mechanics” ACC 2009, submitted.2. Belikov S., Magonov S. “True Molecular-Scale Imaging in Atomic Force Microscopy: Experiment and Modeling” Jap. Jour. Appl. Phys. 2006, 45, 2158-2165
9:00 PM - Z9.4
Towards Single Molecule Detection on Graphene: What we Can Learn from ab initio Studies.
Thomas Connolly 1 , David Carey 1
1 , University of Surrey, Guildford United Kingdom
Show AbstractThe emergence of stable single layer and few layer graphene sheets has seen a renaissance in the study of two dimensional carbon materials. In recent studies single molecule detection on micron sized flakes of graphene has been reported in terms of changes in the local carrier concentration during adsorption or desorption events. The 2D nature of graphene makes its surface readily amendable to the study of molecular physisorption by ab initio methods. Previously, we have studied the adsorption of molecular hydrogen on graphene [1] using density functional theory in both the local density approximation (LDA) and generalised gradient approximation (GGA). Here we report similar studies this time using CO and NO molecules. The binding energy and optimum binding intermolecular separations have been calculated for a number of possible sites in both the LDA and GGA. In addition we have found a significant difference in binding energy depending on the relative orientation of the O atom in the molecule with binding favoured when the O atom is orientated furthest away from the graphene layer.
[1] Daniel Henwood and J David Carey, Phys. Rev. B 75, 245413 (2007).
9:00 PM - Z9.40
Graphene Ribbon Turned into Double-helix, and Other Spontaneous Warping.
Ksenia Bets 1 , Boris Yakobson 1
1 MEMS, Rice University, Houston, Texas, United States
Show AbstractIn pristine graphene ribbons, the disruption of aromatic bond network results in positive edge energy of 1-2 eV/Å. Furthermore, depopulation of the covalent orbitals tends to increase the bonds and thus to elongate the edge, with the effective force of fe ≈ 1 eV/Å (larger for armchair than for zigzag, according to ab initio calculations). This latent force can have quite striking macroscopic manifestation in case of narrow ribbons, as it favors their spontaneous twist, when the parallel edges form a double helix, resembling the DNA, with a pitch Lt of about 15-20 lattice parameters. Through simulations, we investigate how the 1/Lt ~ dA/dz (torsion) decreases with the width of the ribbon, and observe its bifurcation: the twist of the wider ribbons abruptly vanishes and instead the corrugation localizes near the edges. In this case, as well as for very large dimensions, the length-scale of the sinusoidal “frill” at the edge is fully determined by the intrinsic parameters of graphene, its bending stiffness D = 1.5 eV and the edge force fe, so that Le ~ D/fe. Analysis reveals other warping configurations and suggests their sensitivity to the chemical passivation of the edges, with possible sensor applications.
9:00 PM - Z9.41
First-Principles Calculation of Carrier Scattering, Mobility, and Lifetimes in Semiconductors.
Vincenzo Lordi 1 , Paul Erhart 1 , Daniel Aberg 1 , Kuang-Jen Wu 1
1 , Lawrence Livermore National Lab, Livermore, California, United States
Show AbstractWe have developed first-principles methods based primarily on density functional theory to calculate carrier scattering rates and carrier lifetimes in semiconductors, focusing primarily on the effects of nanoscale defects in the materials. We have recently applied these methods to study the role of defects in limiting carrier transport in aluminum antimonide (AlSb), a promising material for high-resolution room-temperature gamma radiation detection; however, the methods are generally applicable and also have been applied to a broader range of systems including nanostructures. Carrier scattering rates are calculated within perturbation theory using a Fermi’s golden rule formalism. Carrier lifetimes are analyzed within Schottky-Read-Hall theory using recombination rates calculated considering several possible mechanisms: radiative recombination, phonon-assisted recombination, and the Auger process. The recombination rates are calculated within a fully first-principles framework with no empirical parameters. With respect to application to room-temperature radiation detection, AlSb nominally possesses an ideal set of properties, including an indirect band gap of 1.6 eV, a large average atomic mass, and potentially high electron and hole mobilities; however, defects and impurities have largely hindered the material’s performance for this application by limiting carrier mobilities and/or lifetimes and reducing resistivity. Based on the methods above, we present an analysis of the effects of native defects and important impurities on the mobilities and lifetimes of carriers in AlSb, which led to experimental improvement in detector-grade material. We show that, among the major impurities present in the material, oxygen and carbon exhibit very strong carrier scattering and lead to significant mobility degradation. On the other hand, substitutional impurities with increasing atomic mass within a column of the periodic table exhibit monotonically decreasing carrier scattering. Based on these results, a synergistic theoretical and experimental effort enabled the optimization of the impurity content in this material to minimize oxygen content and use tellurium n-type doping to compensate unintentional p-type doping by carbon, resulting in high-mobility, high-resistivity material with improved radiation detection performance. Furthermore, a detailed analysis of carrier lifetimes due to trapping on deep levels from native defects and impurities will be presented and discussed in relation to material performance. Open questions and future directions in the state-of-the-art of accurate prediction of mobilities and lifetimes in semiconductors based on fully first-principles methods will be discussed. This work performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.
9:00 PM - Z9.5
Trends of Nanoclusters' Growth by Physical Vapour Deposition Studied by Atomistic Simulation.
Abuhanif Bhuiyan 1 2 , Steven Dew 1 , Maria Stepanova 2
1 ECE, University of Alberta, Edmonton, Alberta, Canada, 2 , National Institute for Nanotechnology, Edmonton, Alberta, Canada
Show AbstractPhysical vapor deposition (PVD) is one of the most flexible, efficient, and clean techniques to fabricate nanopatterns. In particular, self-assembled arrays of nanocrystals can be synthesized by PVD. Synthesis of arrays of nanocrystals is one of central processes in nanoelectronics since nanocrystalline structures are capable of dramatic improving performance of electronic materials. Like the metals and their oxides have application in efficient field emission sources for flat panel displays, magnetic nanocrystals have importance for ultrahigh density magnetic storage devices. Nanocrystals reduce also charge leakage which enhances performance of flash memory and semiconductor quantum dots have important application in photonics. Efficient fabrication of nanostructures in array configurations can be attained through processes of self-assembly. However size, shape and density of self-assembled nanocrystals are highly sensitive to the process conditions such as duration of deposition, temperature, substrate material, etc. To efficiently synthesize nanocrystal arrays by PVD, the process control factors should be understood in detail. Detailed numerical modeling of self-assembled synthesis of nanocrystals is a subject of constantly growing interest. Popular approaches to simulate nanocrystalline film growth include Molecular Dynamics (MD) simulation, Kinetic Monte Carlo (KMC) simulation, and analytic coarse-grained theories. KMC favorably combines flexibility, predictive power, and numerical efficiency, and as such is very well suited for both basic understanding of nanocrystal growth and applications for in-silico aided design of nanofabrication processes. Thus, KMC simulations of film deposition are an important tool for understanding the mechanisms of this process. In this paper, we report an original KMC modeling that explicitly represents PVD synthesis of self-assembled nanocrystals on substrate surface. We investigate, how varying critical process parameters such as deposition rate, duration, and temperature affect the morphologies of self-assembled metallic islands on substrates, and compare our results with experimentally observed surface morphologies generated by PVD. The symmetry of the growing metal clusters by PVD method depends on the type of lattice structure used in the model. We implemented our model on two different substrates, one mimicking primitive square lattice type for analyzing the results of metal deposition on amorphous substrate, and the second one representing a FCC crystalline structure to simulate metal deposition on crystalline structure. The developed model is capable of representing the PVD deposition process for a given set of conditions, and can be employed to study the impact of these conditions on the morphology of the deposited islands. Our simulations correlate well with experimental results reported in the literature and existing numeric models.
9:00 PM - Z9.6
A Mie-based Simulation Study on Spectral and Angular Radiation Properties of Gold-silica-gold Multilayer Nanoshells.
Ying Hu 1 , Ryan Fleming 1 , Vengadesan Nammalvar 1 , Rebekah Drezek 1 2
1 Bioengineering, Rice University, Houston, Texas, United States, 2 Electrical and Computer Engineering, Rice University, Houston, Texas, United States
Show AbstractWhile gold nanoparticles feature a small size factor and are favored in biomedical imaging studies, their plasmonic resonant wavelength cannot be tuned into the near-infrared (NIR) bioimaging window. Silica-gold core-shell conventional nanoshells (CNS) offer NIR plasmonic resonance at small shell thickness-to-core radius ratios. However, coating on sub-100 nm silica cores with an ultra-thin gold layer is difficult to achieve. This study investigates whether extra tunability can be obtained by introducing an additional gold core in the silica layer. Xia et al. were the first to report the synthesis of ~50 nm multilayer nanoshells (MNS) that may exhibit NIR extinction peaks [Nanotechnology, 17 (2006) 5435–5440]. Chen et al. simulated the ultrasharp resonant peaks of similar MNS with an overall diameter of 10 nm [J. Biomed. Opt., 10(2) 024005]. The goal of this paper is to examine the spectral and angular scattering properties of gold-silica-gold MNS in the size region where successful particle syntheses have been reported and can be achieved based on currently available protocols. A Mie-based computation code was developed to calculate light scattering from concentric spheres. MNS were found to bear an extra degree of tunability from the inner gold core. This optical tunability can be understood as an interaction between the conventional nanoshell bonding mode and the gold core sphere mode. The thickness of the intermediate silica layer determines the degree of interplay between the two modes. An increase in the inner gold core radius on an otherwise fixed geometry decreases the intermediate silica layer thickness and increases the plasmon interaction, thus red shifts the extinction peak. Furthermore, the amount of wavelength shift with respect to the resonant wavelength of an equivalent CNS without the gold core was found independent of the overall particle size and to follow a universal scaling principle. The extinction spectra of MNS were found sensitive to the surrounding medium, with the extinction peaks both red shift and increase in magnitude as the dielectric constant of the medium increases. MNS with a larger gold core, a thinner silica layer, or a thinner outer gold shell were found more absorbing than scattering. Both scattering intensity and angular radiation pattern of MNS differ from CNS due to spectral modulation from the inner core. Angular radiation plots suggest that some MNS may provide more backscattering at wavelengths where CNS predominantly forward scatter.
9:00 PM - Z9.8
Three Dimensional Dynamics Simulator for Sintering and Grain Growth Based on Quantum Chemical Molecular Dynamics.
Ai Suzuki 1 , Katsuyoshi Nakamura 2 , Ryo Sato 2 , Kotaro Okushi 2 , Michihisa Koyama 2 , Hideyuki Tsuboi 2 , Nozomu Hatakeyama 2 , Akira Endou 2 , Hiromitsu Takaba 2 , Carlos. A. Del Carpio 2 , Momoji Kubo 3 , Akira Miyamoto 1 2
1 New Industry Creation Hatchery Center(NICHe), Tohoku University, Sendai Japan, 2 Department of Chemical Engineering & Applied Chemistry, Tohoku University, Sendai Japan, 3 Fracture and Reliability Research Institute, Graduate School of Engineering, Tohoku University, Sendai Japan
Show Abstract1.INTRODUCTION/ Catalyst deactivation for long time, especially, the loss of active site of precious metals along with specific surface area of supports is a continuing concern in the automotive catalysts. In order to evaluate theoretically deterioration speed of catalysts prior to a series of duration tests, three dimensional dynamics simulator for sintering and grain growth of supports was developed. In this review, typical automotive catalysts such as Pt/gamma-Al2O3, Pt/ZrO2 and Pt/CeO2 were targets to apply. 2.METHOD/ We built up sub-μm order meso-scale catalyst models by referring experimental TEM images for fresh conditions of Pt/gamma-Al2O3, Pt/ZrO2 and Pt/CeO2. All particle average size of Pt were 1.0 nm. The three dimensional sintering simulator is kinetic monte carlo scheme base, so Pt and support metal oxide are dealt with particle-base. Therefore, μm order large scale catalyst model and hour order long time span calculations can be realized. Three dimensional dynamics simulator for sintering and grain growth can reflect micro scale binding energies which expressing the strengths of Pt-support or metal-oxygen in support materials as diffusion parameters. Various types of diffusion dominate both on the surface and inside the metal oxides in automotive catalysts. For example, there are surface diffusion of supported Pt on support metal oxides and grain boundary diffusion of support metal oxide itself. These kinds of diffusions drive thermal sintering of Pt and grain growth of support and their characteristics of diffusions could be successfully calculated by our original quantum chemical calculations method. The binding energies of support metal oxides are correlated with grain boundary diffusion energies of supports. Whereas, adsorption energies between Pt and supports are correlated with surface diffusion energies of Pt. 3.RESULT/ We investigated the Pt-oxide-support interaction on optimized geometries of Pt/gamma-Al2O3, Pt/ZrO2 and Pt/CeO2. As a result of macro scale sintering simulations which were carried out at 1073 K for 5 hours, Pt-support interaction was in the order; Pt/CeO2 > Pt/ZrO2 > Pt/gamma-Al2O3 represented as adsorption energies. Activation energies for sintering of Pt that were experimentally suggested were approximately linear relationship with calculated adsorption energy of Pt on supports. This causes the difference of Pt sintering. Pt particles supported on the gamma-Al2O3 severely sintered. Pt particles supported on ZrO2 sintered but not so severely progressed than that on gamma-Al2O3. However, Pt supported on CeO2 does not sinter and remains almost intact as fresh condition. On the other hand, the degree of grain growth of support is in the order; ZrO2 > CeO2 > gamma-Al2O3. As a result, sintered particle size of Pt and the amount of specific surface loss of support metal oxides quantitatively agree well with experimental duration test results.
9:00 PM - Z9.9
Accurate Determination of Donor Levels in Si Nanowires.
Adam Gali 1 , Balint Aradi 2 , Thomas Frauenheim 2 , Riccardo Rurali 3
1 Department of Atomic Physics, Budapest University of Technology and Economics, Budapest Hungary, 2 , Bremen Center for Computational Materials Science, Bremen University, Bremen Germany, 3 , Universitat Autonoma de Barcelona, Barcelona Spain
Show AbstractSilicon nanowire (SiNW) is a very promising material in the area of nanodevices, nanosensors, and most recently, in solar cells. The accurate determination of the dopant levels is critical in the optimization of these devices, and it is of high importance to clear the role of typical effective mass like donors or acceptors in SiNW. However, the experimental identification of these dopant levels in nanowires is not straightforward while the accurate calculation of these ionization energies is challenging. In two recent studies tight binding (TB) and standard density functional theory (DFT) calculations predicted quite different behavior for 4 nm≤d≤20 nm regime: while TB predicted that doping efficiency starts to decrease already at d≥20 nm, DFT calculations predicted that the dopant levels are already very shallow for d≥3 nm wires. The results are not conclusive in this important issue due to the deficiancies of both methods. State-of-the-art time-dependent DFT (TDDFT) or many-body perturbation theory (MBPT) would be the best choice to study the dopant states in SiNW but they are computationally prohibitive even for very small defective wires, like d≈1.0 nm. In our recent study we developed such a methodology that could be directly applied for calculating the electronic band gap of hydrogenated SiNW with high accuracy. We applied this method for typical EMT donors in hydrogenated <110> and <111> silicon nanowires showing that: i) in accordance with one-dimensional hydrogen theory, the donor dopants show typical effective mass states (EMT) even at d≈1.5 nm, and its wave function is higly localized in the quantum confinement regime; the EMT state breaks down only at about d≤1.0 nm ii) we provide the hyperfine constant of phosphorus to identify them in SiNWs and we establish the connection between its hyperfine constant and ionization energy iii) the ionization energy of dopants are deeper than the values from DFT, but shows a decreasing difference with larger wires iv) the doping efficiency for d≥8 nm is comparable with that of bulk silicon.