TY - GEN
T1 - Multiscale molecular dynamics simulations of nanostructured materials
AU - Tsuruta, Kenji
AU - Uchida, Atsushi
AU - Totsuji, Chieko
AU - Totsuji, Hiroo
PY - 2007/1/1
Y1 - 2007/1/1
N2 - We present some attempts to simulate nanoscale phenomena, which involve different length-scales and time-scales, using multiscale molecular-dynamics approaches. To simulate realistically an impurity-segregated nanostructure, we have developed the hybrid quantum/classical approach. The method can describe seamlessly both dynamical changes of local chemical bonding and nanoscale atomic relaxations. We apply the method to hydrogen diffusion in Si grain boundary. We find that the hydrogen is strongly trapped in (001)Σ5 twist boundary below 1000K, whereas it starts diffusing along the grain boundary above 1000K. For long-time processes in nanostructure formation, we apply the stochastic-difference-equation method to accelerate the simulations for microstructure evolution. The method bridges the states separated by high-energy barriers in a configuration space by optimizing an action, defined as an error accumulation along a reaction pathway. As an example, a SDE simulation is performed for Cu thin-film formation via nanocluster deposition. We show that the method can be applied effectively to search for the long-time process which involves a rare event due to a large potential barrier between two atomic configurations.
AB - We present some attempts to simulate nanoscale phenomena, which involve different length-scales and time-scales, using multiscale molecular-dynamics approaches. To simulate realistically an impurity-segregated nanostructure, we have developed the hybrid quantum/classical approach. The method can describe seamlessly both dynamical changes of local chemical bonding and nanoscale atomic relaxations. We apply the method to hydrogen diffusion in Si grain boundary. We find that the hydrogen is strongly trapped in (001)Σ5 twist boundary below 1000K, whereas it starts diffusing along the grain boundary above 1000K. For long-time processes in nanostructure formation, we apply the stochastic-difference-equation method to accelerate the simulations for microstructure evolution. The method bridges the states separated by high-energy barriers in a configuration space by optimizing an action, defined as an error accumulation along a reaction pathway. As an example, a SDE simulation is performed for Cu thin-film formation via nanocluster deposition. We show that the method can be applied effectively to search for the long-time process which involves a rare event due to a large potential barrier between two atomic configurations.
KW - Hybrid molecular dynamics
KW - Hydrogen diffusion
KW - Multiscale modeling
KW - Silicon grain boundary
KW - Stochastic-difference- equation method
KW - Thin-film formation
UR - http://www.scopus.com/inward/record.url?scp=38349040021&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=38349040021&partnerID=8YFLogxK
U2 - 10.4028/0-87849-428-6.2804
DO - 10.4028/0-87849-428-6.2804
M3 - Conference contribution
AN - SCOPUS:38349040021
SN - 0878494286
SN - 9780878494286
T3 - Materials Science Forum
SP - 2804
EP - 2809
BT - Supplement to THERMEC 2006, 5th International Conference on PROCESSING and MANUFACTURING OF ADVANCED MATERIALS, THERMEC 2006
PB - Trans Tech Publications Ltd
T2 - 5th International Conference on Processing and Manufacturing of Advanced Materials - THERMEC'2006
Y2 - 4 July 2006 through 8 July 2006
ER -