Self-consistent Charge Density-functional Based Tight-binding Research Articles

Elastic properties of single amylose chains in water: a quantum mechanical and AFM study.

Recent single-molecule atomic force microscopy (AFM) experiments have revealed that some polysaccharides display large deviations from force-extension relationships of other polymers which typically behave as simple entropic springs. However, the mechanism of these deviations has not been fully elucidated. Here we report the use of novel quantum mechanical methodologies, the divide-and-conquer linear scaling approach and the self-consistent charge density functional-based tight binding (SCC-DFTB) method, to unravel the mechanism of the extensibility of the polysaccharide amylose, which in water displays particularly large deviations from the simple entropic elasticity. We studied the deformations of maltose, a building block of amylose, both in a vacuum and in solution. To simulate the deformations in solution, the TIP3P molecular mechanical model is used to model the solvent water, and the SCC-DFTB method is used to model the solute. The interactions between the solute and water are treated by the combined quantum mechanical and molecular mechanical approach. We find that water significantly affects the mechanical properties of maltose. Furthermore, we performed two nanosecond-scale steered molecular dynamics simulations for single amylose chains composed of 10 glucopyranose rings in solution. Our SCC-DFTB/MM simulations reproduce the experimentally measured force-extension curve, and we find that the force-induced chair-to-boat transitions of glucopyranose rings are responsible for the characteristic plateau in the force-extension curve of amylose. In addition, we performed single-molecule AFM measurements on carboxymethyl amylose, and we found that, in contrast to the results of an earlier work by others, these side groups do not significantly affect amylose elasticity. By combining our experimental and modeling results, we conclude that the nonentropic elastic behavior of amylose is governed by the mechanics of pyranose rings themselves and their force-induced conformational transitions.

Theoretical study of vacancy diffusion and vacancy-assisted clustering of antisites in SiC

Using the self-consistent-charge density-functional-based tight-binding (SCC-DFTB) method, we have investigated the migration of vacancies at high temperatures, taking into account the entropy contribution to the Gibbs free energy. We have found that the energy barrier for sublattice migration of ${V}_{\mathrm{Si}}$ is lower than that for ${V}_{\mathrm{C}},$ so that ${V}_{\mathrm{Si}}$ should anneal out at lower temperatures, which agrees with the experimental observations for as-grown SiC. In irradiated material, vacancies and interstitials are expected to appear in high concentrations. In the early stages of annealing, their recombination can also result in antisites. We show that at annealing temperatures at which vacancies are mobile, their motion can mediate the migration of antisites. The vacancy-assisted diffusion of carbon antisites is much faster than that of silicon antisites and may lead to the formation of carbon antisite clusters. We propose a mechanism for the formation of antisite clusters during the migration of silicon vacancies. The role of these complexes in the formation of very stable defects, observed in irradiated SiC, is discussed, additionally.

Interstitial-based vacancy annealing in 4H–SiC

The migration of carbon interstitials through the 4H–SiC lattice and their recombination with vacancies has been investigated theoretically within the self-consistent charge density functional based tight-binding (SCC-DFTB) method. For vacancy–interstitial pairs created by irradiation, the capture radius of silicon and carbon vacancies has been examined, showing that interstitial migration through the otherwise perfect lattice starts getting important for distances larger than four nearest-neighbor atomic distances.

Structures, Energetics and Electronic Properties of Complex III—V Semiconductor Systems

A parallel implementation of the selfconsistent-charge density-functional based tight-binding (SCC-DFTB) method is used to examine large scale structures in III—V semiconductors. We firstly describe the parallel implementation of the method and its efficiency. We then turn to applications of the parallel code to complex GaAs systems. The geometries and energetics of different models for the √19 × √19 reconstruction at the (1-1-1-) surface are investigated. A structure containing hexagonal rings of As at the surface consistent with STM experiments is found to be stable under Ga-rich growth conditions. We then examine voids in the bulk material which are mainly caused by the movement of dislocations. Void clusters of 12 missing atoms are found to be energetically favorable. This is in very good agreement with recent positron annihilation measurements. Additionally, we investigate the diffusion of C in p-type material and suggest a diffusion path with an activation energy of less than 1 eV which is consistent with experimental studies. Finally, focusing on GaN we provide atomistic insight into line defects in wurtzite GaN threading along the growing c-axis. We highlight the stability and electronic properties of screw and edge dislocations, discuss reasons for the formation of nanopipes and relate the yellow luminescence observed in highly defected materials to deep acceptors, VGa and VGa–(ON)n, trapped at threading edge dislocations.

A parallel code for a self-consistent charge density functional based tight binding method: Total energy calculations for extended systems

We present a parallel version of a selfconsistent-charge density-functional based tight-binding (SCC-DFTB) method for total energy calculations and geometry optimizations of clusters and periodic structures. On single processor machines the SCC-DFTB method has been successfully applied to systems up to several hundred atoms with an accuracy comparable to sophisticated selfconsistent field density-functional theory (SCF-DFT) methods. The new parallel code allows to treat systems which are larger by an order of magnitude in reasonable time. The freely available ScaLAPACK and PBLAS libraries are used for linear algebra operations. We tested the scaling of our program for a realistic system (III–V semiconductor surface) with different sizes and give a short outlook on current applications.