Ab Initio global optimization of the structures of SinH, n=4-10, using parallel genetic algorithms

OFELIA OÑA; VÍCTOR BAZTERRA; MARÍA C. CAPUTO; MARTA B. FERRARO; JULIO FACELLI

Isla Margarita, Venezuela

Congreso; Congreso internacional de químicos teóricos de expresion altina; 2005

IVIC, Venezuela.

The study of the structure and physical properties of atomic and molecular clusters is an extremely active area of research due to their importance, both in fundamental science and in applied technology. Existing experimental methods for structural determination seldom can obtain the structure of atomic clusters directly. Therefore the calculation, using theoretical structures and comparison with experimental values of their physical and optical properties is the most common way to obtain structural information of atomic clusters. During the last two decades a number of researchers have characterized amorphous hydrogenated silicon (α-Si:H) by different techniques and determined the physical and chemical properties of amorphous silicon films [1, 2].  The role of defects in the properties of amorphous and microcrystalline silicon (α-Si and μc-Si, respectively) is very important because these materials are employed in solar cells and optoelectronic devices [3].  As the hydrogen diffuses and modifies the structure, the energy spectrum and the properties of the devices change. To understand these phenomena it is important to understand how hydrogen affects the local electronic structure and geometry in these systems. Since systems like amorphous and microcrystalline silicon are very difficult to handle computationally, understanding about the effects due to hydrogen addition may be achieved by calculations on small hydrogenated silicon clusters. These calculations critically depend on the available structures of such clusters.  Prasad et al. [4], Yang et al. [5] and references therein constitute a very complete bibliography record of the state of the art in the determination of the structures of hydrogenated silicon clusters. Due to computational limitations, the authors either limited themselves to calculations using local minimizations when using Density Functional Theory (DFT,) methods for the energy calculations or used approximated methods to calculate the cluster energies when using global search techniques. Unfortunately, this has been the state of the art in the determination of atomic cluster structures [6], computational limitations have forced researchers to use either local optimisations or approximate energy calculations.  Our recent work in Sin clusters [7] shows the need for global optimisations, while the work on SinCu clusters [8] clearly demonstrates the drawbacks of using approximate energy calculations when the energy hyper surface provided by the approximate method does not closely match the one predicted from first principles. We have concluded that it is of importance to develop the necessary computational capabilities to make possible to perform global searches of atomic cluster structures using ab initio methods. In this work we report the results of ab initio global optimisation of the structures of  SinH, n=4-10, atomic clusters using parallel genetic algorithms. Driving the global search with the MGAC parallel implementation of the genetic algorithms and using the DFT (Density Functional Theory) as implemented in the CPMD (Carr Parinello Molecular Dynamics) code to calculate atomic cluster energies and perform the local optimisation of their structures, we have been able to demonstrate that it is possible to perform global optimisations of the structure of atomic clusters using ab initio methods.  The results show that this approach is able to find many structures that were not previously reported in the literature. Moreover in most cases the new structures have considerable lower energies than those previously known. The results clearly demonstrate that this calculation are now possible and in spite of their larger computational demands provide more reliable results.   [1]          H. Fritzsche, Annual Review of Materials Research 31, 47 (2001). [2]          R. O. Jones, B. W. Clare, and P. J. Jennings, Phys. Rev. B 64, 125203 (2001). [3]          A. Hiraki, T. Imura, M. Iwami, et al., Solar  Energy Materials 2, 125 (1979). [4]          D. Balamurugan and R. Prasad, Bull. Mater. Sci. 26, 123 (2003); R. Prasad, Bull. Mater. Sci. 26, 117 (2003); D. Balamurugan and R. Prasad, Phys Rev. B 64, 205406 (2001). [5]          J. C. Yang, X. Bai, C. P. Li, et al., J. Phys. Chem. A 109, 5717 (2005). [6]          J. Zhao and R.-H. Xie, J. Comp. and Theor. Nanoscience 1, 117 (2004). [7]          V. E. Bazterra, O. Oña, M. C. Caputo, et al., Phys. Rev. A 69, 053202 (2004). [8]          O. Oña, V. E. Bazterra, M. C. Caputo, et al., J. Mol. Struct. (THEOCEHM) 681, 149 (2004).