IT-1 Magnetic properties of FeNi Alloys

Mesbahuddin Ahmed
Department of Physics, University of Dhaka, Dhaka – 1000, Bangladesh
email : netproj@bdcom.com

The magnetic properties of pure Fe and its alloys with Ni are calculated using Tight-Binding Linear Muffin-Tin (TB-LMTO) method in a local density approximation (LSDA). First we generate the density of states of pure Fe and pure Ni and then we find it with different combination of both the transition elements. The TB-LMTO equations are solved using the augmented space recursion technique. We also evaluate the total energy and the Fermi energy.
 
 

IT-2 Third-Generation Muffin-Tin Orbitals

O.K. Andersen, T. Saha-Dasgupta, S. Ezhov, D. Savrasov, L. Tsetseris,
O.Jepsen, R.W. Tank, C. Arcangeli, and G. Krier.
Max Planck Institute for Solid State Research, D-70506 Stuttgart, Germany
email : oka@fkf.mpg.de

Muffin-tin orbitals (MTOs) have been used for a long time in ab initio calculations of the electronic structure of condensed matter. Over the years, several MTO-based methods have been developed. The ultimate aim is to find a generally applicable electronic-structure method which is intelligible, fast, and accurate. Our recent progress in that direction will be reported. In order to be intelligible, an electronic-structure method should employ a minimal and flexible basis of short-ranged orbitals. The method should be able to describe the valence electrons in sp-bonded materials using merely four short-ranged s- and p-orbitals per atom and, for insulating phases, using merely occupied orbitals such as bond orbitals. Another example is materials with strong electronic correlations. For such materials, one must first construct a realistic Hamiltonian, and this requires an accurate single-particle basis which can be partitioned into correlated and non-correlated orbitals, without introducing too many of the former. A flexible basis of short-ranged orbitals is thus asked for. Now, a small basis of short-ranged orbitals is a prerequisite for a method to be fast (e.g. of order N), but may be a hindrance for the accuracy, because the orbitals of a smaller basis are in general more complicated than those of a larger basis. Most other ab initio methods, such as plane-wave pseudopotential, LAPW, PAW, and LCAO methods aim at simulation, and are therefore primarily accurate and robust. But they are neither fast nor intelligible in the above-mentioned sense, because they employ basis sets with about hundred functions per atom. With those methods, understanding is therefore attempted after the calculation, by means of projections onto e.g. charge densities, electron-localization functions, partial waves, Wannier functions in case of insulators, a.s.o.. With MTOs, this is not necessary. We shall explain what 3rd-generation MTOs are and, by the examples of diamond-structured Si, the CuBr - Ge series, and the cuprate high-temperature superconductors, demonstrate that they are intelligible and accurate.

Andersen O.K., T. Saha-Dasgupta, S. Ezhov, L. Tsetseris, O. Jepsen, R.W.Tank, C. Arcangeli, G. Krier: 
"Third-generation MTOs" Psi-k Newsletter #45 (June 2001), 86-119.
Pavarini E., I. Dasgupta, T. Saha-Dasgupta, O. Jepsen, O.K. Andersen: "
Band-Structure Trend in Hole-Doped Cuprates and Correlation with Tcmax" 
Phys. Rev. Lett. 87 (2001), 047003, 1-4
Andersen O.K., T. Saha-Dasgupta: "Muffin-Tin Orbitals of Arbitrary Order,"
Physical Review B 62 R16219-R16222, 2000.
Andersen O. K., Saha-Dasgupta T., Tank R. W., Arcangeli C., Jepsen O.,
Krier G.: "Developing the MTO formalism," in: Electronic Structure and
Physical Properties of Solids. The Uses of the LMTO Method, ed. H. Dreyssé.
Berlin/Heidelberg: Springer (2000). Lect. Notes Phys. 535 3-84, 2000.
 
 

IT-3 The projector augmented wave method:  ab-initio molecular dynamics with full wave functions

Peter E. Bloechl,
Institute for Theoretical Physics, 
Clausthal University of Technology, Germany
email : Peter.Bloechl@tu-clausthal.de

The ab-initio moeclular dynamics method of Car and Parrinello has reshaped the way we do electronic structure calculations today. Initially it has been limited to the pseudopoten- tial approach, which, despite a number of advantages, lacks the rigor of all-electron methods. The PAW method is an extension of and a link between all-electron and pseudopotential approaches, combining the strengths of both worlds. It allows efficient ab-initio simulations for all atoms in the periodic table with the full wave functions. I will introduce the basic concepts of the PAW method and describe some representative applications.
 
 

IT-4 Classical wave band gap material from coated spheres

C.T. Chan
Physics Department
Hong Kong University of Science and Technology
Kowloon, Hong Kong
email : phchan@ust.hk

We use theory and computation as a tool for designing man-made photonic and sonic band gap materials using coated spheres. Photonic and sonic gap materials are artificial composites designed to have unusual dispersions. The photonic band gap material can control, confine and guide light the same way as electronic semiconductor can control and regulate the flow of electrons. Sonic band gap material can be used as a nearly perfect sound shield. These functional materials do not exist in nature. They are first conceived through physical intuition. The ensuing physical properties are then calculated and predicted by state-of-the-art computational techniques. Theory then gives a prescription and design of the material. Computation can predict the properties of these advanced composites and can provide a blueprint for the material, which is then fabricated and characterized accordingly. The measured properties generally agree well with those predicted by computation.
 
 

IT-5 Total Energy Calculation of Perovskite BaTiO₃ by Self-Consistent Tight Binding Method

B. T. Cong ^a, P. N. A. Huy ^a, P. K. Schelling ^b , J. W. Halley ^c
(a)- Faculty of Physics, Hanoi University of Science,
334 Nguyen Trai Street , Hanoi, Vietnam.
email : cong@cms.edu.vn
(b) Argonne National Laboratory, USA
(c) School for Physics and Astronomy, University of Minnesota, USA

We present result of numerical computation on some characteristics of BaTiO₃ such as total energy, lattice constant, density of states, band structure... using self-consistent tight binding method. Besides the strongly Ti-O bond between 3d on titanium and 2p orbital on oxygen states, we also include the little hybridization between the Ba 6s and O 2p states. The results are compared with the one of other more sophisticated methods.
 
 

IT-6 Electronic structure of magnetic superconductor RuSr₂GdCu₂O₈

Indra Dasgupta*
Department of Physics, Indian Institute of Technology
Powai, Mumbai 400 076, India
email : dasgupta@phy.iitb.ernet.in

We have investigated the novel magnetic and electronic structure of the magnetic superconductor RuSr₂GdCu₂O₈, using first principles tight-binding linearized muffin tin orbital (TB-LMTO) method. We discuss the possibility of mixed valency of Ru in this compound and its role on the observed magnetic properties. In addition, we report the results of the massive downfolding procedure which allows an ab-initio way to construct low energy, few-band, tight-binding(TB) model Hamiltonians describing the hybridization of the ruthenate and cuprate bands crossing the the Fermi level. On the basis of our model Hamiltonian we argue, the probable physical scenario that can make the co-existence of ferromagnetism and superconductivity possible in this compound.

* Work done in collaboration with T. Saha-Dasgupta, S.N. Bose Centre, Calcutta;
O.K. Andersen and O. Jepsen, Max Planck Institut, Stuttgart; V.I. Anisimov, Institute of
Metal Physics, Ekaterinburg; G. Baskaran, Institute of Mathematical Sciences, Madras.
 
 

IT-7 Transport properties in magnetically doped nanotubes

Keivan Esfarjani
Department of Physics, Sharif Institute of Technology, Tehran, Iran
email : k1@sharif.edu

In order to investigate novel device properties in nanometer scale, we have considered a carbon nanotube which has strong mechanical properties and a very high conductivity, and studied the effect of magnetic dopants placed inside the tube. First, the effect of a single dopant has been considered within the mean-field approximation and using the tight-binding theory. The ferromagnetic dopant will polarize the tube assuming that there is considerable overlap between their wavefunctions. This is very similar to the Anderson Model except that our model is more realistic and includes all five 3d orbitals of the transition metal element. Next, we consider the effect of a chain of dopants inside the tube. Transport properties of the tube will be affected by doping because up spin electrons traveling on the tube will be scattered differently with the ferromagnetic chain than down spin electrons. This will lead to a spin walve effect which will be stronger for semiconducting tubes, if the tube is set to be used as a nanotransistor. The results and the physics of this effect will be discussed in the talk.
 
 

IT-8 New Theoretical Approaches to Biomolecular Functions and Laser-induced Material Dynamics and Chemistry

Thomas Frauenheim,
University of Paderborn, Theoretical Physics, Germany
email : frauenheim@phys.uni-paderborn.de

An efficient density-functional-theory (DFT)-based approach to predictive simulations of nanoscale materials and properties is described. Structure formation under technological relevant conditions is controlled by Molecular-Dynamics simulations based on quantum-mechanically calculated interatomic forces. Successful applications to inorganic structures already cover a broad spectrum of problems, ranging from cluster physics and chemistry through the design of novel Carbon materials to semiconductor technology (surface-growth, interface formation, defect and device engineering).

Approaching new frontiers in materials science the Quantum-method is generalized and improved to cover also weak interactions (hydrogen-bonds and dispersion forces) becoming crucial for modelling of soft materials. By coupling the Quantum- to Molecular Mechanics force-field simulations (QM/MM-interface) and applying new so-called Order-N-techniques for solving the Quantum-problem we provide the theoretical basis for addressing large biomolecular systems in natural environments and studying their functions. Recently, these methods have been used in large-scale simulations on long-time dynamics of proteins and DNA-dodecamers. Projects on studying the biological activity of the Retinal Schiff Base and the related conformational response of their receptive protein environment are outlined.

New developments in time-dependent density-functional-theory (TD DFT) further allow to couple the electron and ion dynamics via a generalized quantum-classical Lagrangian and to study the interaction of ultra-short intense laser pulses with nanoscale materials. As the result we can follow the non-adiabatic structural dynamics of the systems in electronically excited states. Potential applications of the method range from ultra-short time laser spectroscopy to studies of photoisomerization and photo induced chemical processes.

We also describe development of simulation tools to study the current transport in molecular structures. In our approach the system is described via DFTB while boundary conditions to account for current carrying states are accounted via Green function techniques. Thermal average of transport properties are obtained by performing Molecular Dynamics simulations. The developed tool has been applied to Nanotube-Devices, DNA and organic/inorganic heterojunctions.
 
 

IT-9 Density functional theory of materials modelling at different length scales

Swapan K. Ghosh
Theoretical Chemistry Section, Chemistry Group, RC&CD Division,
Bhabha Atomic Research Centre, Mumbai 400 085, India
e-mail : skghosh@magnum.barc.ernet.in

Materials modelling has been one of the important areas of research in various disciplines of science and engineering. Depending on the interest and application one is concerned with, an appropriate length scale is chosen for a theoretical description of materials. One of the concepts that have been found to serve as a vital ingredient in the theoretical tools useful in the study of materials at all the length scales of interest is the concept of density.

In the microscopic length scale, it is the electron density that has played a major role in providing a deeper understanding of chemical binding in atoms, molecules and solids. In the intermediate mesoscopic length scale, an appropriate picture of the equilibrium and dynamical processes has been obtained through the single particle number density of the constituent atoms or molecules. A wide class of problems involving nanomaterials, interfacial science and soft condensed matter has been addressed using the density based theoretical formalism as well as atomistic simulation in this regime. In the macroscopic length scale, however, one usually considers matter as a continuous medium and a description using local mass density, energy density and other related density functions has been found to be quite appropriate.

In spite of the differences in the nature of the density variables used in all these descriptions, the corresponding theoretical frameworks have been found to possess an underlying unified structure. Besides attempting to project the many-particle picture to a single particle one, this density functional theory based description provides a unified theoretical framework for quantum as well as classical systems encompassing the diverse length scales involved in materials modelling, the basic features and some recent developments of which form the subject matter of this talk.
 
 

IT-10 A New Constant Pressure Molecular-Dynamics Method and Its Application To a Single Nanotube

X. G. Gong
Department of Physics, Fudan University, 200433-Shanghai, China
Institute of Solid State Physics, Chinese Academy of Science, 230031-Hefei, China
e-mail : gong@theory.issp.ac.cn

By writing the volume as a function of coordinates of atoms, we propose a new constant-pressure molecular dynamics method with parameters free. This method is specially appropriate for the finite system, in which the periodic boundary condition does not exist and the conventional constant pressure molecular dynamics method can not be directly used. As the applications of this new method, we have studied the behavior of a single carbon tube under the pressure and find a pressure induced hard-soft transition of carbon tube. Some results on other finite system, such as melting of nano particles under pressure, will be also discussed.
 
 

IT-11 The ABINIT software project

Xavier GONZE
Unite PCPM, U. Catholique de Louvain, B-1348 Louvain-la-Neuve, Belgium
Douglas C. ALLAN
Corning Inc., Fundamental Res., SP FR 5, Corning, NY 14831 and the ABINIT group
email : gonze@pcpm.ucl.ac.be

The computation of electronic structure, total energy, forces and many related properties of condensed matter, thanks to density-functional theory (DFT), is a field in constant progress. A DFT software project that wants to stay at the frontier of knowledge cannot be the work of a single individual, neither of a small group. Also, up-to-date software engineering concepts can considerably ease the harmonious development of such software.

The ABINIT project relies upon these ideas : reliability, portability, readability and freedom of sources are emphasized, in the course of developing a sophisticated plane-wave pseudopotential code. More than 200 automated tests secure existing capabilities (concept of self-testing), despite heavy development efforts and the associated bug generation; thanks to MAKE and PERL scripts, and CPP directives, the unique set of Fortran90 source files (about 100000 lines) can generate sequential (or parallel) object code for many platforms, under Unix/Linux, DOS/Windows and MacOS; strict coding rules have been followed to make the source readable; the documentation is extensive, including online help files, tutorials, HTML-formatted sources. Moreover, the whole package is distributed under the GNU General Public License, often nicknamed 'copyleft'. For more information, see <http://www.pcpm.ucl.ac.be/ABINIT>.
 
 

IT-12 The Role of Clusters in the Design of Nano-Scale Systems

P. Jena
Virginia Commonwealth University
Richmond, VA 23284-2000, USA
email : pjena@vcu.edu

Atomic clusters consisting of a few to a few thousand atoms constitute a new phase of matter intermediate between atoms and solids. Unlike conventional nanostructured materials, the size and composition of these clusters can be controlled one atom at a time. The properties of such clusters brought about by their large surface-to-volume ratio, unique geometry, low dimensionality and reduced coordination, exhibit novel behavior quite unlike that in the bulk. For example, metallic elements can be made to form ionic bonds while nonmagnetic and anti-ferromagnetic materials can become ferromagnetic or ferrimagnetic. This talk will introduce the principles for designing these clusters and discuss a concept where clusters can be viewed as super-atoms – adding a third dimension to the periodic table. Recent experimental evidence to support this idea will be presented. Examples of cluster assembled materials will include high-energetic materials involving Al(MnO₄)_3, alkali metal clusters isolated in zeolites, transition metal clusters supported on organic and metallic substrates, and manganese-oxide clusters passivated by acetate ligands. Ultimately the properties of crystals composed of clusters as the building blocks will be discussed. It is hoped that the synergy between theory and experiment will lead to the synthesis of cluster assembled materials with unique and tailored properties, thus creating new opportunities in materials science at the dawn of the new millinnium.
 
 

IT-13 Melting in finite size systems

Dilip G. Kanhere
Department of Physics, University of Pune, Pune 411 007, India
e-mail : kanhere@unipune.ernet.in

Recent advances in ab initio treatment of finite size systems have opened up many exciting possibilities of realistic simulations of clusters to study a variety of phenomenon. Melting of atomic clusters is one such phenomenon, whose understanding at microscopic level is currently actively pursued using experimental and simulation techniques. The present talk focuses on issues related to ab initio methods such as simulation time scales, the choice of potentials, the characterization parameters etc. We advocate multiple histogram technique for calculating specific heat which is directly comparable to experiments.

We have employed the density functional molecular dynamics to investigate the phenomenon of melting of small sodium clusters. We will present and interpret specific heat of these clusters, calculated via 40 ps simulation runs per temperature. We have also investigated the effect of impurity (Sn) on small Lithium clusters. We find rather substantial increase ( by a factor of two) in the melting temperature. These results will be presented along with the methodology of multiple histogram technique.
 
 

IT-14 How to Realize Prediction of Materials Properties by Computer Simulation

Yoshiyuki Kawazoe
Institute for Materials Research, Tohoku University,
Sendai 980-8577, Japan
email : kawazoe@imr.edu

Ab initio methods widely used in solid state physics and recently extended to materials research to analyze physicochemical properties of materials are shifting to a new generation after three decades of the local density approximation (LDA) and 15 years of Car-Parrinello method. Several progresses have already been achieved in the electron gas theory such as GW approximation, which now is gradually possible based on the present day state of art of supercomputer power to be applied to industrial materials to estimate absolute band gap value to predict color of photoemission, ionization potential, and electron affinity. So called ab initio molecular dynamics based on the Car-Parrinello formalism only seeks for the most stable structure of the system restricted on the Born-Oppenheimer surface, and therefore the time-dependent Schrödinger equation should be solved numerically to trace chemical reactions to estimate reaction paths and branching ratios.

In the long history of materials research, useful materials properties have been tried to be achieved mainly experimentally, because they are too complex in nature and simple model calculations by quantum mechanics have never been able to predict them. Theoretical treatments based on the band structure calculations with LDA have been applied mainly to solve these problems, where the electronic structures are calculated assuming the crystal structures determined traditionally by X-ray experiments. It has been too time consuming to determine the crystal structures by ab initio method without experimental information. This situation, however, has been dramatically changed recently, because the processing speed and the memory size of the new generation supercomputers (1TFLOPS and 1TB) have reached the range to apply ab initio method to realistic materials and also because of new inventions of theoretical progresses to reduce the necessary computer resources dramatically.

In this talk, recent results based on the mixed-basis all-electron formulation developed originally by my research group are introduced focused on two interesting examples; (1) GW approximation applied to microclusters to estimate absolute HOMO (Highest Occupied Molecular Orbital) - LUMO (Lowest Unoccupied Molecular Orbital) gap to predict the expected wave length of the light emitted from cluster assembled new materials, and (2) molecular electronics, which is expected to solve the difficulty of present silicon technology by overcoming the problem of submicron limit and realize higher density and higher speed of processors and memories.
 
 

IT-15 Ab Initio Calculations and Design of Ceramic Interfaces

Masanori Kohyama,
Interface Science Research Group,
Special Division of Green Life Technology,
National Institute of Advanced Industrial Science and Technology,
1-8-31, Midorigaoka, Ikeda, Osaka, 563-8577, Japan
e-mail : m-kohyama@aist.go.jp

It is of great importance to investigate ceramic interfaces such as grain boundaries and ceramic/metal interfaces. Grain boundaries dominate various properties of ceramics, and are associated with various functions of ceramics. It is crucial to fabricate ceramic/metal interfaces with desirable properties for the structural and electronic applications of ceramics. It is desirable to understand such ceramic interfaces at the atomic and electronic scales in order to develop the methodology to design such interfaces. Currently, it is possible to perform first-principles calculations of ceramic interfaces by virtue of the development of the density-functional theory and recent novel computational schemes such as the first-principles molecular-dynamics method. In this paper, we present our recent first-principles studies of SiC grain boundaries [1] and SiC/metal interfaces [2]. For SiC grain boundaries, first, we have clarified stable atomic configurations of coincidence tilt boundaries, which are in good agreement with recent high-resolution electron microscopy observations [3]. The effects of polar and non-polar interfaces and interfacial C-C and Si-Si wrong bonds have been analyzed. Second, the mechanical properties of such grain boundaries have been examined by using 'ab initio tensile tests', where the tensile strength and fracture are clarified through the behaviour of electrons and ions. We have observed the critical bond stretching for the Si-C bond breaking and the effects of atomic-scale inhomogeneity associated with the wrong bonds on the tensile strength and fracture. This kind of ab initio tensile or shear tests are promising tools to investigate the basic mechanical properties of ceramic interfaces. For SiC/metal interfaces, we have dealt with Si-terminated and C-terminated interfaces of SiC(001)/Al, SiC(001)/Ti and SiC(111)/Ti interfaces in order to clarify the effects of surface species, metal species and interface planes. Each interface has quite different features such as atomic configurations, bonding nature, adhesive energies and Schottky-barrier heights (SBH). For the C-terminated interfaces, there exist strong covalent and ionic interactions, which induce large adhesive energies. This large reactivity of the C-terminated interface is in good agreement with the experiments. We have found that the SBH is greatly affected by the different interface dipole associated with the different interface species. This result provides valuable insights into the true mechanism of the SBH.

[1] M. Kohyama, Mater. Chem. Phys. 50, 159 (1997); M. Kohyama et al., MRS Symp. Proc. 491, 287 (1998); M. Kohyama, Phil. Mag. Lett. 79, 659 (1999); submitted to Phys. Rev. B (2001).
[2] J. Hoekstra and M. Kohyama, Phys. Rev. B 57, 2334 (1998); M. Kohyama and J. Hoekstra, Phys. Rev. B 61, 2672 (2000); S. Tanaka and M. Kohyama, to appear in Phys. Rev. B (2001).
[3] K. Tanaka and M. Kohyama, "Electron Microscopy 1998", Vol.II, edited by H.A. Calderon Benavides et al. (IOP, 1998), p.581; "Grain Boundary Engineering in Ceramics", edited by T. Sakuma et al. (American Ceramic Society, 2000), p.231; to appear in Phil. Mag. A (2001).
 
 

IT-16 Surface diffusion, growth, and the formation of semiconductor nanostructures

Peter Kratzer
Fritz-Haber-Institut der Max-Planck-Gesellschaft
Faradayweg 4-6, D-14195 Berlin, Germany
e-mail : kratzer@fhi-berlin.mpg.de

In the design of new materials, modeling of the properties of materials and of the processes used to fabricate them has gained considerable importance. In contrast to conventional modeling techniques, that typically rely on empirical parameters and thus are limited in their applicability to a specific material and a narrow range of conditions, the ab initio approach to modeling has the potential to overcome these limitations and will eventually enable us to perform simulations which have predictive character. However, in many processes, e.g. in epitaxial growth of thin films, the properties of interest develop over time scales of the order of seconds and involve several thousands of atoms, while the ruling microscopic processes operate in the atomic length and time domains of 0.1 - 1 nm, and femto- to pico-seconds. Hence, modeling is challenged by the need to bridge this gap in the length and time scales by many orders of magnitude.

I shall report about recent progress towards this goal due to the use of multi-scale modeling techniques, by combining the information about atomic-scale processes obtained from density functional theory (DFT) with techniques suitable to treat the longer length scales or time scales of the problem at hand. The examples will be taken from the epitaxy of group III-V-semiconductor materials:
- calculations of the structural properties of semiconductor nanostructures (InAs islands on GaAs) that result from an energetic balance between elastic strain relief and the cost of forming additional crystal facets. Here, the elastic energy gain can be obtained from continuum elasticity theory, while the surface energies of crystal facets come from DFT slab calculations. 
- DFT studies of surface diffusion (In atoms on GaAs) in the presence of long-range strain fields due to heteroepitaxially grown nanostructures. 
- simulations of molecular beam epitaxy of GaAs ranging from the atomic scale up to the scales of micrometers and seconds, using the kinetic Monte Carlo technique with microscopic rate constants derived from DFT calculations.
 
 

IT-17 Novel caged clusters of silicon: Fullerenes, Frank-Kasper polyhedra and cubic

Vijay Kumar
Institute for Materials Research, Tohoku University, 2-1-1 Katahira, Sendai, Japan
and Dr. Vijay Kumar Foundation, Chennai 600 078, India
email : kumar@imr.edu

Current interest in miniaturization of devices has led to intense research on the development of novel nano-forms of semiconductor materials. Silicon clusters have been studied in great detail [1] in the small size range of a few tens of atoms. These results have shown that clusters with 15-20 atoms have prolate structures. Fullerenelike structures with a few silicon atoms inside have also been proposed [2] for Si₄₅. Recently it has been found [3] that one metal atom can affect the structure and properties of silicon clusters drastically. These results obtained from computer experiments based on ab initio ultrasoft pseudopotential method, have led to the findings of novel forms of metal encapsulated silicon clusters with fullerene, Frank-Kasper polyhedral and cubic shapes. The size and shape of these clusters depend upon the size of the metal atom. It is found that 14, 15, and 16 atom silicon clusters are very stable. The embedding energy of metal atom is very large (about 12 eV) that leads to the stability of these clusters. The HOMO-LUMO gap of these clusters can be varied in the range of 1 – 2.35 eV. The latter lies in the visible range. Also interaction between clusters is rather weak. These properties make these species attractive for the development of self-assembled materials. Our calculations showed magic behavior for M@Si15 and M@Si16 (M = Cr, Mo, and W) in agreement with experimental findings. Further results [4] on doped clusters of germanium and tin will be discussed in the light of these findings.

[1] V. Kumar, K. Esfarjani, and Y. kawazoe, in Clusters and Nanomaterials, Editors Y. Kawazoe, T. Kondow, and K. Ohno, Springer Series in Cluster Physics (Springer-Verlag, Heidelberg, 2001), p. 9.
[2] U. Rothlisberger, W. Andreoni, and M. Parrinello, Phys. Rev. Lett. 72, 665 (1994).
[3.] V. Kumar and Y. Kawazoe, Phys. Rev. Lett. 87, 045503 (2001); V. Kumar and Y. Kawazoe, Jap. J. Appl. Phys. (in press); V. Kumar and Y. Kawazoe in Reviews in Modern Quantum Chemistry – A celebration of the Contributions of R.G. Parr, Rd. K.F. Sen, World Scientific (in press); V. Kumar and Y. Kawazoe, submitted for publication.
[4.] V. Kumar and Y. Kawazoe, submitted for publication
 
 

IT-18 Numerical study of a Zn impurity in high T_c superconductors

Ting-Kuo Lee
Institute of Physics, Academia Sinica, Taipei 11529, Taiwan
and
National Center for Theoretical Sciences, Hsinchu, Taiwan
email : tklee@phys.sinica.edu.tw

The magnetic polarization induced by nonmagnetic impurities such as Zn in high T_c cuprate compounds is studied by the variational Monte Carlo simulation. The variational wave function is based upon the results of Bogoliubov de Gennes mean field approach for the two-dimensional t-J model. A Jastrow factor is introduced to account for the induced magnetic moment and its correlation with the spins and holes surrounding the impurity. A substantial energy gain is obtained by forming an antiferromagnetic polarization covering 4 or 5 lattices sites around the impurity. We also found the doping dependence for the induced magnetic moment to be consistent with experiments.
 
 

IT-19 Is carbon nanotube a good candidate for ultimate energy storage?

Young Hee Lee
Department of Physics, Sungkyunkwan University, Suwon, 440-746, Korea
email : leeyoung@yurim.skku.ac.kr

Carbon nanotubes have been considered for energy storage of battery, supercapacitor, and hydrogen storage. Yet, the principles and practical achievements for energy storage are on strong debate. In this report, I will give a brief summary of what has been done so far and what are the problems from both theoretical and experimental point of views. We will describe the maximum storage capacity of hydrogen, storage mechanism, and the effect of alkali metal doping from density functional theory. Some experimental results will be explained in terms of theoretical point of view.
 
 

IT-20 Linear scaling (Order-N) methods in electronic structure

Richard M. Martin
Department of Physics, University of Illinois at Urbana-Champaign
Urbana, IL 61801, USA
email : rmartin@uiuc.edu]

Traditional methods for calculation of electronic structure of materials scale non-linearly with the size of the system, i.e., the number of atoms N in the molecule or in a cell with periodic boundary conditions for extended matter. Nonlinearity arises because electrons are delocalized and each eigenstate overlaps all the other eigenstates. Of order N³ operations are required to find all the eigenstates, and even if clever algorithms are used the scaling is still at least N². On the other hand, a basic tenet of our understanding of matter is that many properties are extensive, i.e., they are properties of some local region and are independent of distant regions. Classical models have this behavior: for large systems all energies and forces can be found by calculations that scale linearly with N since each atom interacts with only a finite number of neighbors if there are only short range forces; and long range Coulomb forces can be handled in special ways.

Under what circumstances can one formulate Order-N quantum mechanical methods that scale linearly with N? The basic idea is that such algorithms can take advantage of the fact that the density matrix decays as a function of distance. All the methods start from this point, which was perhaps best described by Kohn as "nearsightedness"[1]. In an insulator the decay is exponential making such algorithms robust and feasible. In a metal the day is algebraic at zero temperature and exponential at finite T; even though the range is still very long Order-N methods can be useful in the cases where they are needed - complex disordered metals. Several different methods have been proposed based upon the "divide and conquer" idea[2]; the density matrix[3]; Wannier function type methods[4,5]; and finite T power series operator expansions[6]. A recent review describes the methods[7]. When are Order-N quantum mechanical methods useful and needed? There are two basic areas for use. One is large complicated systems like huge molecules. Examples of large calculations so far are full molecular dynamics simulations of a turn of DNA using the SIESTA program[8] and calculations of giant fullerenes[9]. The other is complex condensed matter systems where is more advantageous to make algorithms that calculate the properties of a region only in terms of a neighborhood of that region instead of relying upon cells with artificial boundary conditions and brute force computer power to make the cells large enough that the boundary conditions are not important. Some typical applications of Order-N methods will be briefly described.

[1] Kohn, W., Chem. Phys. Lett. 208, 167 (1993).
[2] Yang, W., Phys. Rev. Lett. 66, 1438 (1991).
[3] Li, X.-P., W. Nunes, and D. Vanderbilt, Phys. Rev. B 47, 10891 (1993).
[4] Ordejon, P., D. Drabold, M. Grumbach, and R. Martin, Phys. Rev. B 48, 14646 (1993); Ordejon, P., D. Drabold, R.M. Martin, and M.P. Grumbach, Phys. Rev. B 51, 1456 (1995)
[5] Mauri, F., G. Galli, and R. Car, Phys. Rev. B 47, 9973 (1993); Kim, J., F. Mauri, and G. Galli, Phys. Rev. B 52, 1640 (1995).
[6] Goedecker, S., and L. Colombo, Phys. Rev. Lett. 73, 122 (1994).
[7] Goedecker, S. Rev. Mod. Phys. 71, 1085 (1999).
[8] Artacho E, Sanchez-Portal D, Ordejon P, et al., Phys. Stat. Sol. B 215, 809 (1999).
[9] Xu, C., and G. Scuseria, Chem. Phys. Lett. 262, 219 (1996).
 
 

IT-21 First principles phase transformations in alloys

Abhijit Mookerjee
S.N. Bose National Centre for Basic Sciences
JD Block, Sector-3, Salt Lake City, Kolkata 700 093, India
email : abhijit@boson.bose.res.in

We shall report here a first-principles, density functional based approaches to the study of phase transformations in alloys. We shall, in particular, base our energy estimates on the tight-binding linearized muffin-tin orbitals technique of Andersen and coworkers, coupled with the recursion method of Haydock and shall describe disorder effects through the augmented space formalism introduced by me. The entropy estimates will be through the cluster variation method. Phase stabilities will be studied using various approaches : the embedded cluster methods, the concentration wave method of Khatchaturyan. We shall illustrate the above with various magnetic and non-magnetic alloy systems.
 
 

IT-22 Environment-dependent bond-order potentials: new developments and applications

D. Nguyen-Manh (1), D.G. Pettifor (1), D.J.H. Cockayne(1),
M. Mrovec(2), S. Znam (2), and V. Vitek(2) 
(1) Department of Materials, Oxford University, Park Road, Oxford OX1 3PH, UK
(2) Department of Materials Science and Engineering, University of Pennsylvania,
Philadelphia, Pennsylvania 19104-6272, USA
Email: manh.nguyen@materials.oxford.ac.uk

The Bond-Order Potentials (BOPs) idea formulated by Pettifor [1] employs the orthogonal two-center tight-binding (TB) representation for the bond energy and the Harris-Foulkes approximation for the repulsive pairwise contribution. In the last ten years, although many efforts have been focused on theoretical calculations of the bond-order expression, the BOPs still suffers from the uncertainty of how best to choose the semi-empirical TB parameters that enter the scheme. In this talk, we review recent developments [2] to obtain the reliable and transferable BOPs which help to extend the accuracy and applicability to technologically important multi-components systems. Firstly, we have found that a simple pair potential is unsuitable for describing the environmental screening effects due to s and p orbital overlap repulsion in transition metal alloys and therefore the inability to reproduce the negative Cauchy pressures exhibiting in strong covalent systems. By adding an environmental repulsive term, the Cauchy pressure problem has been removed and we now able to get the BOPs for studying extended defects, dislocations and mechanical properties of high-temperature intermetallic Ti-Al alloys. In particular, new results on core structures and possible dissociation of different type of dislocations will be discussed. Secondly, we present the first derivation of explicit analytic expressions for environmental dependence of s , p and d ? bond integrals by inverting the non-orthogonal matrix. We illustrate the power of this new formalism by showing that it is not only captures the transferability of bond integrals between Mo, Si and MoSi₂ but also predicts the large discontinuities between first and second nearest neighbors for pps , ppp and ddp even though absence of any discontinuity for the dds ?bond integral. A new environment-dependent BOPs has been developed for bcc-Mo indicating that the core structure of ½<111> screw dislocations is narrower than structures found in previous studies in agreement with recent ab-initio calculations. Finally, the new formalism will allow us to study the problem of medium range order found recently in amorphous materials with covalent bonding at large and realistic nanoscale. For the case of a-C where the issue of sp2 versus sp3 is very crucial for modeling amorphous structure we found that the s ?and p ?bond integrals are not only transferable between graphite and diamond structures but they also strongly anisotropic due to inter-plan bonding between graphite sheets.

[1] D.G. Pettifor, Phys. Rev. Lett., 63, 2480, (1989); D.G. Pettifor and I. Oleinik, Phys. Rev. Lett., 84, 4124, (2000).
[2] D. Nguyen-Manh, D.G. Pettifor, S. Znam and V. Vitek. in Tight-Binding Approach to Computational Materials Science, Eds. by P.E.A. Turchi et.al., MRS. Sym. Proc., 491, 353, (1998); D. Nguyen-Manh, D.G. Pettifor and V. Vitek, Phys. Rev. Lett., 85, 4136, (2000).
 
 

IT-23 Theoretical Study on Surface Dynamics

Takahisa Ohno
National Institute for Materials Science
1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan
email : ohno.takahisa@nims.go.jp

This talk will focus on the application of theoretical methods to adsorption dynamics on semiconductor and metal surfaces. The adsorption dynamics of Ge atoms on the hydrogen-terminated Si(001) surface will be discussed in terms of first-principles total energy calculations and kinetic Monte Carlo simulations. Work on the adsorption dynamics of F atoms on the Si(001) surface by using a hybrid method between a quantum mechanics and a molecular mechanics will demonstrate the effect of the dissipation of the F adsorption energy to the Si substrate. The novel adsorption spectra of NO molecules on the Pt(111) will be explained by calculations of the dynamic dipole moment.
 
 

IT-24 Microscopic properties of the defects and hydrogen impurity in ferroelectric perovskites and their effect on the spontaneous polarization

Chul Hong Park
Research Center for Dielectric and Advanced Matter Physics
Pusan National University, Gumjung, Pusan, 609-735, Korea
email : cpark@pusan.ac.kr

Recently perovskite ferroelectric compounds are intensively investigated, since the perovskite ferroelectric compounds is employed in developing non-volatile ultra high-density semiconductor memories for its spontaneous polarizations and extremely high dielectric constants. The perovskite oxide is also of great interest as a nonlinear optical material. Usually the thin-film of perovskites oxide has high concentration of defects and the defects lead to some important reliability problems of the ferroelectric device such as fatigue and imprint.

In the talk, we discuss the microscopic properties of defects such as O-vacancy and hydrogen impurity in ferroelectric perovskite through first-principles pseudopotential total-energy calculations. An important consideration is the interaction between defects and host polarization. Many defect-related phonomena such as aging, fatigue, imprint and hydrogen-induced degradation of ferroelectrics will be discussed based on the calculational results. Especially the role of hydrogen will be discussed in details and we will give an explanation to the industrial problem that hydrogen contamination during passivation severely damages the ferroelectric devices with the loss of ferroelectricity.
 
 

IT-25 Theory, modelling and simulation of materials

Michael R. Philpott
Department of Materials Science, National University of Singapore
email: chmmp@nus.edu.sg

The properties and dynamics of several different material systems important in science and technology have been studied using molecular dynamics (MD):

Immiscible liquid-liquid interfaces

Diamond, graphite and amorphous carbon films

Metal clusters in aqueous electrolyte solution

Each is examined with a different variant of MD. In all three the motions of the nuclei are assumed to follow classical mechanics. In the metal cluster the electron motion is treated quantum mechanically using density functional theory and combined with the assumed classical motion of the nuclei using the method of Car and Parinello [1]. For immiscible liquid systems we have studied interfaces containing ions between water and benzene derivatives. In the bond order simulation of amorphous carbons the Brenner potential was used. It is shown how the nature of the carbon deposited depends critically on the energy of deposition [2]. Empirical potentials are avoided in the Car-Parinello molecular dynamics (CPMD) method by suitably combining density functional theory with MD. Our strategy to extend the time scale of CPMD places constraints on the nuclear and electronic motion in order to reduce the computational time. We call this modification chemistry in a box MD. It is useful when all the electrons involved in chemistry are localized and when the molecules in the surroundings not involved in chemistry can be treated using empirical potentials. We illustrate the method with applications to metal particles in aqueous electrolyte and the association of LiBr in water [3].

Acknowledgement : Over the years it has been my good fortune to work with outstanding computational scientists. In particular I want to acknowledge J.N. Glosli, J. Belak and S. Izvekov.

References
R. Car and M. Parrinello, "Unified approach for molecular dynamics and density functional theory", Phys. Rev. Let. 55, 2471 (1985).
J.N. Glosli, J. Belak, and M.R. Philpott , "Molecular Dynamics Modeling of Ultra Thin Amorphous Carbon Films", in "4th International Symposium on Diamond Materials." Electrochemical Society Symposium Proceedings 95-4, 25-37 (1995). (Electrochem. Soc. Inc. Pennington, NJ, 1995).
S.V. Izvekov, M.R. Philpott and R.I. Eglitis_, "Ab initio simulation of metal cluster surrounded by Electrolyte." J. Electrochemical Soc. 147, 2273 – 2278 (2000).
 
 

IT-26 Ground state structures and properties of small hydrogenated silicon clusters

Rajendra Prasad
Department of Physics
Indian Institute of Technology, Kanpur 208 016, INDIA
email : rprasad@iitk.ac.in

We present results for ground state structures and properties of small hydrogenated silicon clusters using the Car-Parrinello molecular dynamics with simulated annealing. We discuss the nature of bonding of hydrogen in these clusters. We find that hydrogen can form a bridge like Si-H-Si bond connecting two silicon atoms. We find that in the case of compact and closed silicon cluster hydrogen bonds to the silicon cluster from outside. To understand the structural evolutions and properties of silicon clusters due to hydrogenation, we have studied the cohesive energy and first excited electronic level gap of clusters as a function of hydrogenation. We find that first excited electronic level gap of Si_n and Si_nH_m fluctuates as function of size and this may provide a first principles basis for the short-range potential fluctuations in hydrogenated amorphous silicon. The stability of hydrogenated silicon clusters is also discussed.
 
 

IT-27 Boron Induced Modifications of Silicon Surfaces and Nanotubes

Marian W. Radny
School of Mathematical and Physical Sciences, The University of Newcastle
Callaghan, NSW, 2308 Australia
email : phmwr@alinga.newcastle.edu.eu

The recent discovery that MgB₂ and metallic Boron are high temperature superconductors raises the interesting question of whether elemental boron will manifest some unexpected behaviour in other situations. In this talk we will discuss the unexpected, experimentally verified, atomic and electronic structure of the Boron modified Si(111) surface and Boron doped single walled, open ended, finite, carbon nanotubes. The discussed systems were calculated within the ab initio methods (Gaussian 98, VASP) for atomic clusters and periodic slabs.
 
 

IT-28 On the electronic structure and equation of state in high pressure studies of solilds

R.S. Rao and B.K. Godwal
High Pressure Physics Division, Bhabha Atomic Research Centre, Mumbai-400085, India
email : rsrao@apsara.barc.ernet.in

One of the main goals of the theoretical studies in condensed matter physics has been the prediction of phase transitions. The ab-initio theoretical calculations have been contributing prominently in this endeavour, especially in the phenomena occurring at high pressures. Advances in the experimental high pressure techniques, enabling the acquisition of high resolution data at tuned pressures have contributed significantly in verifying the theoretical predictions. The efforts made to resolve the controversies between the theoretical predictions and the experimental data have provided rich dividends in some of the theoretical developments, especially in improving the exchange-correlation terms in the calculations based on the density functional theory.

Many of the high pressure studies concentrate on the structural transitions, especially at very high pressures. However, considerable attention has recently been given to isostructural electronic topological transitions (ETT) at moderate pressures in elemental solids and compounds, leading to interesting interplay between theoretical and experimental efforts to resolve the related controversies. We have carried out various high pressure studies on elemental solids and intermetallics, like high resolution X-ray diffraction data using diamond anvil cell, and synchrotron source along with high-resolution imaging plate detectors; resistance and thermoelectric power measurements, and corroborated them with ab-initio electronic structure calculations. Our detailed high pressure studies have shown that manifestations of ETTs can be seen in the changes in the structural parameters (without structural transition), like anomaly in the variation of axial ratio in the hexagonal phase; and anomalies in the variation of transport properties (electrical resistivity and thermoelectric power); and most significantly, the change in the slope of universal equation of state. Our work also shows the importance of temperature effects in the electronic structure calculations, as one deals with the proximity of the electron energy band extremum to Fermi level (on the order of mRy) in the case of ETT, in contrast to many other physical properties which may be related to conduction bandwidth (on the order of Ry) etc.
 
 

IT-29 Microscopic study of Electronic behavior of double perovskite Sr₂FeMoO₆ and low-dimensional vanadate LiV₂O₅ using N-MTO method

T. Saha-Dasgupta
S.N. Bose National Centre for Basic Sciences,
JD Block, Sector-3, Bidhan Nagar, Kolkata 700 098
e-mail : tanusri@bose.res.in

In recent years a generalized form of the linear-muffin-tin orbital (L-MTO) based electronic structure method, namely the N-MTO method, has been proposed and implemented [1], where the flexibility of the basis set can be manipulated to arrive at a minimal-orbital Wannier-like description of the electronic states starting from otherwise complicated LDA band-structures. The method has been proved to be very successful in providing understanding of complex crystalline materials at the level of first-principles derived model Hamiltonians, filtering out the information far away from the energy region of interest.

In the present talk we will discuss the application of this method to two systems, the double perovskite compound Sr₂FeMoO₆ [2] which is recently shown to exhibit room-temperature colossal magneto-resistive effect and the low-dimensional vanadium based transition metal compound LiV₂O₅ [3] as compared to structurally similar NaV₂O₅ exhibiting the influence of the charge-ordering and the corresponding crystallographic distortions on magnetic interactions in these low-dimensional compounds.

[1] O.K. Andersen and T. Saha-Dasgupta, Phys. Rev. B 62, R16219 (2000)
[2] D.D. Sarma, P. Mahadevan, T. Saha-Dasgupta, S. Ray and A. Kumar, Phys. Rev. Lett, 85, 2549 (2000)
[3] R. Valenti, T. Saha-Dasgupta, J. V. Alvarez, K. Pozgajcic and C. Gros, Phys. Rev. Lett, 86, 5381 (2001)
 
 

IT-30 Multiscale Modeling Strategies in Material Science

Vijay B. Shenoy
Department of Mechanical Engineering
Indian Institute of Technology, Kanpur 208 016, INDIA
email : vbshenoy@iitk.ac.in

The problem of prediction of finite temperature properties of materials poses great computational challenges. The computational treatment of the multitude of length and time scales involved in determining macroscopic properties has been attempted by several workers with varying degrees of success. This paper will review several multi-scale modeling strategies including atomistic and continuum methodologies. Special attention will be given to the recently developed quasicontinuum method which is an attempt to bridge the length scales in a single seamless model with the aid of the finite element method. Attempts to generalize this method to finite temperatures will be outlined.
 
 

IT-31 Superconductivity in MgB₂ and Other Diborides: Role of p-Electrons

Prabhakar P. Singh
Department of Physics, Indian Institute of Technology
Powai, Mumbai 400 076, India
email : ppsingh@phy.iitb.ernet.in

Using density-functional-based methods we have studied the fully-relaxed, full-potential electronic structure of the new superconductor MgB₂ [1] and various other diborides [2], including TaB₂ [3]. Our results, to be described in terms of (i) density of states (DOS), (ii) band-structure, and (iii) the DOS and the charge density around the Fermi energy E_F, clearly show the importance of B p-band for superconductivity. In particular, we find that around E_F, the charge density in MgB₂ is planar and is associated with the B plane. For BeB₂ and NaB₂, our results indicate qualitative similarities with MgB₂ but significant quantitative differences in their electronic structure due to differences in the number of valence electrons and the lattice constants a and c. Our results for the charge density around E_F in TaB₂ show a striking similarity to MgB₂ as far as the B plane is concerned. A comparison of the band-structures of TaB₂ and VB₂, coupled with l-character analysis, indicates that TaB₂ has substantially more p –character than VB₂ along A-L and H-A directions near E_F.

[1] J. Nagamatsu et. al., Nature (London), 410, 63 (2001).
[2] Prabhakar P. Singh, Phys. Rev. Lett. 87, 87004 (2001).
[3] Prabhakar P. Singh, cond-mat/0104580.
 
 

IT-32 Magnetism of ultra-thin ferromagnetic film and nano magnets with antiferromagnetic substrate

Ruibao Tao
Department of Physics, Fudan University, Shanghai, China
email: rbtao@fudan.edu

The ferromagnetic thin film with antiferromagnetic substrate and nano magnets has very high potential in the applications. For example, the exchange bias (EB) in the ferromagnetic thin film with an antiferromagntic substrate has been applied in the GMR magnetic read heard. Due to this key issue, the study on antiferromagnetic domain (AFMD) is rapidly growing in current research[1]. Meanwhile, the molecule magnetism with nano size, such as Fe8, Mn12 and V15 et al., are also attractive area to the people due to its possible application in future as a quantum multi-bites recording media [2]. In this presentation, we studied on the mechanism of the AFMD and explored an intrinsic mechanism by means of the simulation of spin dynamics. We first present the picture of AFMD induced by an intrinsic mechanism from the competition between dipole-dipole interaction and anisotropy. A phase diagram of AFMD from homogenous phase to Bloch type AFMD and Neel type one is presented. The detail domain structure of AFMD is received. In our calculation, following model Hamiltonian is suggested where the lattices of the film are located at (x, y) plane,

.

Where (n, i) denotes the lattice at i-th site of the n-th layer, h the applied magnetic field. By means of this model Hamiltonian, we also studied the effect of the order direction of a compensated antiferromanetic substrate on the magnetic resistance (MR) of ferromagnetic film that can be extended to increase the sensitivity of normal MR read head. Meanwhile, we have studied the effect of antiferromagnetic capping on the hysteresis loop of magnet with nano size. It has a single domain. The detail magnetic structure of single domain and its evolution from the homogenous domain to the vortex states and the transition among the vortex states are presented. With a special arranged AFM substrate, a quantum model with two states we have achieved, and it can be used as a multi-bites quantum recording media. The similar behavior [2] in the molecule magnets is compared. Such designed nano ferromagnets with antiferromatic substrate can be called as an artificial super-molecule magnet which might have the potential in the future application of quantum computing as a multi-bites recording media.

The work is supported by National Natural Science Foundation of China and Shanghai Center of Applied Physics.

References
[1] A. Scholl, J. Stohr, J. Luning et al., Science 287, 1014 (2000)
[2] Michael N. Leuenberger and Daniel Loss, Nature 410, 789 (2001)
 
 

IT-33 Variational Quantum Monte Carlo Approach to High Tc Superconductivity*

Nandini Trivedi
Tata Institute of Fundamental Research, Homi Bhabha Road, Mumbai 400 005, India
 

The high T_c cuprates are widely believed to be described by a single band Hubbard model. However the lack of detailed quantitative results for this model has made it difficult to assess the extent to which this simple description suffices. With this motivation we study the Hubbard model with parameters relevant to the cuprates, using variational Monte Carlo with projected d-wave states. We show that our variational wavefunction exhibits three phases: an RVB insulator at half-filling, a d-wave superconductor (SC) for hole doping 0 < x < 0.35 and a Fermi liquid for larger x . We show that strong correlations naturally give rise to two energy scales: The SC order parameter tracks the observed nonmonotonic T_c(x), while the variational parameter D_var(x) scales with the (p ,0) ``hump'' and T* seen in photoemission. The Fermi surface is a large hole barrel essentially unaffected by correlations. However, strong correlations lead to incoherence in the spectral function and from the singular behavior of its moments we obtain quantitative estimates of the nodal quasiparticle weight Z and Fermi velocity v_F. Remarkably, Z~ x, though the Fermi velocity remains finite as x® 0, which has important implications for the self energy. The Drude weight D_low and superfluid density have magnitudes and doping dependences that are consistent with experiments and we predict that D_low~ Z, the nodal quasiparticle weight.

* In collaboration with Arun Paramekanti and Mohit Randeria.
[1] A. Paramekanti, M. Randeria, and N. Trivedi, cond-mat/0101121 (Phys. Rev. Lett. to appear).
 
 

IT-34 Towards development of efficient computational approaches for time dependent quantum electronic processes

M. Tsukada and N. Watnabe
Department of Physics, Graduate School of Science, University of Tokyo
Hongo 7-3-1, Bunkyo-ku, Tokyo 1130033, Japan
e-mail : tsukada@phys.s.u-tokyo.ac.jp

In spite of remarkable progress of the density functional theory for the description of solid and molecular systems, practical calculations for larger systems required for the nano-structure materials have not been straight forward. This is because the conventional methods using either plane wave basis or the atomic orbital basis suffer from their own disadvantages for the efficiencies in calculations of large scale systems. We developed a direct real space finite element method, which is very efficient for the larger systems both for the static and dynamic problems.

In this presentation, after introducing the advantages of the real space finite element method for the density functional calculations of large atomistic systems, I discuss the recently developed finite element approach for the solution of the time-dependent Schroedinger or Kohn-Sham equation. This method is implemented by several innovative techniques of computational physics such as Suzuki’s exponential product, Cayley’s form, the finite element method and the so called adhesive operator. This method conserves the norm of wave functions, and can utilize adaptive mesh refinement, and is very suitable for the parallel type computers.

We demonstrate the advantages of this method for calculating electron dynamics in the nano-scale structures with various case studies on simple examples.
 
 

IT-35 Computational Approach to the Understanding of the Linear and Nonlinear Properties of Practical Nonlinear Optical Crystals

Ding-sheng Wang
Institute of Physics, Chinese Academy of Sciences, Beijing, 100080 China
email: dswang@aphy.iphy.ac.cn

Four most important application related criteria of nonlinear optical crystals, i.e., the harmonic generation efficiency, power damage threshold, acceptance angle and transparency window, are known to be determined by (1) the nonlinear susceptibility; (2) the multiphoton absorption rate; (3) the birefringe; and (4) the energy gap, of the crystals.

With the development of the state-of-the-art band calculation scheme, and massively parallel processing in the high performance computing, we are now able to calculate all these physical properties or the complex practical nonlinear optical crystals, which contains usually 30-100 atoms per unit cell from the first principles with an accuracy acceptable for materials development/design.

With the important series of borates crystals (BBO and CBO developed in China, and CBO by a Chinese-Japanese collaboration) as examples, we will show how the computation approach answers the questions which were long in demand by, and proved a challenge to, the experimentalists.

[1] Electronic structure and optical properties of LiB₃O₅, CsB₃O₅, and BaB₂O₄ crystals calculated from first principles, Jun Li, Chun-gang Duan, Zong-quan Gu, and Ding-sheng Wang, Phys. Rev. B57, 6925 (1998)
[2] First principles calculation of the second harmonic generation coefficients of borate crystals, Chun-gang Duan, Jun Li, Zong-quan Gu, and Ding-sheng Wang, Phys. Rev. B60, 9435 (1999)
[3] Mechanism for linear and nonlinear optical effects in b -BaB₂O₄, J. Lin, M. H. Lee, Z. P. Liu, and C. T. Chen, Phys. Rev. B60, 13380 (1999)
[4] Mechanism for linear and nonlinear optical effects in LiB₃O₅, CsB₃O₅, Z.S.Lin, J.Lin, Z.Z.Wang, et al., Phys. Rev. B62, 1757-1764 (2000)
 
 

IT-36 Modeling Electronic Processes in Organic Light Emitting Diodes

S. Ramasesha
Solid State and Structural Chemistry Unit
Indian Institute of Science,
Bangalore 560 012, India
email: ramasesh@sscu.iisc.ernet.in

It has been demonstrated in the last decade that some conjugated organic polymers which show strong fluorescence can be used as active materials in flexible light emitting diodes (1, 2). In order to improve efficiency of the organic light emitting diodes (OLEDs), it is important understand the electronic structure and the electron-hole recombination processes in these systems. Statistically, it appears that the triplet to singlet exciton formation ratio is 3:1. However, experiments suggest the ratio to be larger in favour of the singlets (3). This and other experimental evidences suggest that pi-electrons in conjugated systems are strongly correlated, and need to be modeled using models which take into account explicit electron-electron interactions (4, 5). The minimal model for these semiconducting systems is the Pariser-Parr-Pople (PPP) model which includes long-range electron correlations. In this talk, we will discuss our studies on the low-lying electronic excitations in conjugated polymers using various many-body techniques which throw light on the fluoresence properties of these polymers (6, 7). We also discuss the many-body quantum dynamics of electron-hole recombination and triplet-triplet scattering (8). We will present some results of a classical Monte Carlo simulation of the efficiency the OLEDs in multilayer devices.

(1) J.H. Burroughes, D. D. C. Bradley, A.R. Brown, R.N. Marks. K. Mackay, R.H. Friend, P.L. Burn and A.B. Holmes, Nature, 347, 539 (1990),
(2) D. Braun and A.J. Heeger, Appl. Phys. Lett., 58, 1982 (1991).
(3)Y. Cao, I.D. Parker, G. Yu, C. Zhang and A.J. Heeger, Nature, 397, 414 (1999)
(4)S. Ramasesha in Chemistry of Advanced Materials} ed. C.N.R. Rao, Blackwell, London (1993)
(5) S. Ramasesha, S.K. Pati, Z. Shuai and J.L. Br'edas, Adv. Quantum Chem. 38, 121 (2000)
(6) Z.G. Soos, S. Ramasesha and D.S. Galvao, Phys. Rev. Letts., 71, 1609 (1993).
(7) S. Ramasesha and Kunj Tandon, in Density Matrix eds. I. Peschel et al, Springer (1998), Dresden
(8) M. Wohlgenannt, Kunj Tandon, S. Mazumdar, S. Ramasesha and Z.V. Vardeny, Nature, 409, 494 (2001) 
 
 

IT-37 Semiflexible Equilibrium Polymers: Lattice and Continuum Models 

Apratim Chatterji and Rahul Pandit
Department of Physics, 
Indian Institute of Science, Bangalore 560 012.
Email: rahul@physics.iisc.ernet.in

There is a growing class of polymeric systems in which the lengths of the polymers are not fixed but fluctuate, via scission and fusion, till an equilibrium length distribution obtains. If such polymers are also semiflexible, then ordered phases can form as the density of polymers increases or the temperature decreases. We give an overview of the statistical mechanics of such phase transitions, including transitions from isotropic to polymer-nematic phases in the context of lattice and continuum models that we have developed. We also examine nonequilibrium phenomena such as shear alignment of such polymers. 
 
 

IT-38 Simulation of catalytic materials in media: gas, liquid and solid

Rajappan Vetrivel
GE India Technology Centre, Whitefield Road, Bangalore – 560 066.
E-mail: rajappan.vetrivel@geind.ge.com

Catalysis is an interesting but difficult phenomenon to simulate due to molecular interactions happening in gas, liquid and solid phases. Simulating the catalytic behavior of materials involves calculating an infinite number of ‘ill-defined’ diatomic interactions. Reproducible methods have been developed to simulate the influence of solid systems. The first and the classical one is to use a large unit cell and apply periodic boundary conditions [Born & von Karman, 1912]. The other relatively modern approach is to model the solids as separate quantum and electrostatic regions [Mott and Littleton, 1938] as well as find a suitable interface. Reliable methods have been developed to simulate the influence of liquid systems. The methods vary from simulating simple fixed-volume ‘box-like’ system containing the required number of solvent molecules to reproduce liquid phase [Rzepa et al, 1991] to more complicated Conductor-like Screening Model (COSMO) [Klamt and Schuumann, 1993]. Whereas the simulation techniques to study the influence of the surrounding gas systems is in its initial phase. A ‘cluster approach’ to represent the exposed surface and ‘quantum dynamics approach’ are being explored. The opportunities and challenges in these approaches will be the focus of the seminar.

Born, M. and von Karman, T., 1912, Z. Physik, 13, 297.
Klamt, A. and Schuumann, G., 1993, J. Chem. Soc. Perkin Trans. 2, 799.
Mott, N.F. and Littleton, M.J., 1938, Trans. Faraday Soc., 34, 485.
Rzepa, H.S., Yi, M.Y., Karelson, M.M. and Zerner, M.C., 1991, J. Chem. Soc.. Perkin Trans. 2, 635.
 

IT-39 Design extraordinary photonic materials through computer simulations

Lei Zhou
Phys. Dept., The HK Univ. Sci. Tech., Clear Water Bay, Kowloon, HK
E-mail: phzhou@ust.hk

We use theory and computation as a tool to design man-made materials with unusual photonic properies (band gaps, negative refraction index, etc.). A planar band gap material is designed based on a specific class of fractal patterns, which can totally reflect an electromagnetic wave with wavelength much larger than its own size. Moreover, the transmission properties of such a material can be activelly modulated. A material with both negative permittivity and permeability is shown to have an effective negative refraction index (often called left-handed material), and has been demonstrated to have many potential applications. We have designed a material which not only has a negative effective refraction index but also has an effective impedance matching the vaccum, that is, a transparent left-handed material. A new mechanism for creating photonic band gaps is also discovered, which is uniquely related to such left-handed materials. All these designed materials have been successully fabricated in laboratory, according to the accurate compuations based on finite difference time domain method. The measured properties of such materials generally agree very well with those theoretically predicted.