Alexander von Humboldt Foundation Conference
«Technologies of the 21st century: biological, physical, informational and socialaspects»
Saint-Petersburg, Russia, September 27-29, 2005
Atomic-scale Monte Carlo simulation: a useful tool of modern nanotechnology
A.A.Schmidt
St-PetersburgState Polytechnical University, Russia
Nowadays semiconductor industry continues its exponential growth, which was predicted by Gordon Moore back in mid 60s [1]. He predicted that the maximum number of transistors integrated in one circuit would double every other year, thus exponentially increasing the computational power of computers. Such integration growth also led to the decrease of the transistor size from several microns to a hundred of nanometers and even less [2]. At the same time methods of the nano-size fabrication found further application in novel technology branches such as micro- and nanoelectromechanical systems construction, molecular biology and so forth. Wide usage of “man-made” materials [3],molecules [4] and atom-by-atom assemblyare the distinguishing features of modern nanotechnology. It is obvious that such a fine technology demands manipulation and “tuning” of a vast amount of the processingparameters, which makes implementation of solely experimental methods for the determination of the optimal production conditions inefficient. An alternative way is to use computer simulation that is less expensive and time-consuming. At present four major methods are used for the nanotechnology simulation: ab initio density functional theory, molecular dynamics, statistical methods based on Monte Carlo algorithms and continuum mechanics [5]. We will focus on the Monte Carlowhich is named by John von Neumann [6] because it is based on the random numbers sampling and reminds of the city renownedfor its casinos.Monte Carlo has a great advantage, comparative to other methods, because it is capable to provide atomic-scale precision data on the nano-objects evolution at appropriately long time-scale that is in a range of real process time. In the present work Monte Carlomethod was used to study the growth of silicon carbide on the silicon substrate in vacuum [7]. AlthoughSiC on Si hetero-system allows many applications for high-frequency and high-power electronics production, its utilization in semiconductor technology is hampered, because of the certain complexityof the physical processes that take place during the growth process and so make the production of the device with desired properties virtually impossible. Computer simulations allowed us to get an insight to the growth process and to investigate the spontaneous self-ordering of SiC clusters on the surface features (as steps of defects) [8].
[1]
[2]
[3] Z.I. Alferov, Rev. Mod. Phys. 73 (2001) 767.
[4] S. Iijima,Nature 354 (1991) 56.
[5] D.D. Vvedensky, J. Phys.: Condens. Matter 16 (2004) R1537.
[6] D.D. McCracken,Sci. Am.,192 (1955) 90.
[7] A.A. Schmidt, K.L. Safonov, Yu.V. Trushin, V. Cimalla, O. Ambacher, and J. Pezoldt., phys. stat. sol. a, 201, (2004) 333.
[8] V. Cimalla, A.A. Schmidt, Ch. Foerster, K. Zekentes, O. Ambacher and J. Pezoldt, Superlattices and Microstructures, 36, (2004) 345.
A.A.Schmidt
St-PetersburgState Polytechnical University, Russia
Phone: +7 812 5565140. E-mail: