Temperature-accelerated dynamics

Keywords infrequent events transition-state theory accelerated dynamics hyperdynamics parallel-replica dynamics temperature-accelerated dynamics molecular dynamics bond-boost hyperdynamics parallel-accelerated dynamics Cu(100)... 

Sorensen, M.R., Voter, A.F. Temperature-accelerated dynamics for simulation of infrequent events. J. Chem. Phys. 2000, 112, 9599-606. 

Shim, Y., Amar, J.G., Uberuaga, B.P., Voter, A.F. Reaching extended length scales and time scales in atomistic simulations via spatially parallel temperature-accelerated dynamics. Phys. Rev. B 2007, 76, 205439-1-11. 

The search for local minima in the neighborhood of a given local minimum is usually performed by the excitation of the system from this state followed by the relaxation of the system. If the relaxation of the excited system results in a state different from the initial state (and explored earlier), then a new local minimum is found, otherwise the evolution of the excited system is continued. The ways of moving out of the initial state can be different in temperature accelerated dynamics (TAD) by Sorensen and Voter [78], MD is used at high temperatures in the activation-relaxation technique (ART) by Mousseau and Barkema [79] and the local activated Monte Carlo method (LAMC) [80], the system evolves along the direction opposite to the direction of the force in the long-scale kinetic Monte Carlo... 

The basic idea of the algorithm to be described here (i.e. temperature accelerated dynamics ) is that by increasing the temperature, the overall rate at which activated processes take place is increased. However, altering the temperature also has the effect of changing the relative rates of these microscopic processes and can even turn on processes that would for all practical purposes not even have taken place at the lower temperature of interest. As a result, it is necessary to correct for the temperature induced bias that is present at the higher temperature. Hence, some of the high-temperature transitions are filtered out so as to restore the hierarchy of low-temperature transitions. 

Sorensen M. R. and Voter A. F., Temperature-Accelerated Dynamics for Simulation of Infrequent Events, J. Chem. Phys. 112, 9599 (2000). 

A fundamental requirement on all of the computational studies on metal surface dynamics is fhe need fo perform simulafions with realistic potentials and in a feasible amounf of fime. To this end, the temperature-accelerated dynamics method [14,74,75] has arisen as a possible approach for reaching the latter limit. With the exception of quanfum simulations, most classical simulations are based on semiempirical potentials derived either from the embedded atom method or effective medium theory [76-78]. However a recent potential energy surface for hydrogen on Cu(l 10) based on density functional theory calculations produced qualitatively different results from those of the embedded atom method including predictions of differenf preferred binding sites [79]. 

Sorensen MR, Voter AF (2000) Temperature-accelerated dynamics for simulation of infiequent events. J ChemPhys 112 9599-9606... 

There exist several other variants of Voter s hyperdynamics, for example, temperature-accelerated dynamics ... 

Prominent among the methods for exploring the atomic scale dynamics of a system, including relaxation and rare events, are temperature-accelerated dynamics (TAD) [39], hyperdynamics [40] and parallel replica [41], all developed by Voter and coworkers. These techniques build on statistical mechanics principles for infrequent event systems, and as such do not make any prior assumptions regarding the atomistic mechanisms. They are designed to simply allow the system to evolve more quickly from state to state than they would in normal MD, provided that the barriers are relatively high compared to kT. 

For systems in which the dynamics consist of long stays in the basins around potential energy minima interrupted by swift hops between them passing through saddle points, the assumptions of TST are often obeyed and the long time dynamics can be studied with the accelerated MD methods [20] discussed in Sect. 4. hi these approaches, which include parallel replica dynamics, hyperdynamics, and temperature-accelerated dynamics, the rate of esct ie from the minima is artificially enhanced in one way or another. The natural dynamics on long time scales is then reconstructed from such boosted simulations and often dramatic speed-ups can be achieved. 

---