Fragment recognition in molecular dynamics
We investigate the properties of three methods of fragment recognition in microscopic simulations of molecular dynamics. They are (a) the early cluster recognition algorithm (ECRA) which looks for the most bound partitions in phase space, (b) the minimum spanning tree in two particle energy space (M...
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | JOUR |
| Acceso en línea: | http://hdl.handle.net/20.500.12110/paper_05562813_v56_n2_p995_Strachan |
| Aporte de: |
| Sumario: | We investigate the properties of three methods of fragment recognition in microscopic simulations of molecular dynamics. They are (a) the early cluster recognition algorithm (ECRA) which looks for the most bound partitions in phase space, (b) the minimum spanning tree in two particle energy space (MSTE) which looks for those simply connected partitions in which each particle is bound to, at least, one other member of the cluster to which it belongs, and (c) the standard minimum spanning tree (MST) in configuration space. It is found that, if the objective of a given calculation is to study the time evolution of properties related to the fragment distribution, the MST should be discarded, MSTE results will be valid for not too short times, and ECRA results will give the most complete description of such properties. © 1997 The American Physical Society. |
|---|