1992
DOI: 10.1002/qua.560440505
|View full text |Cite
|
Sign up to set email alerts
|

Algorithmic tools in the study of semiempirical potential surfaces

Abstract: The most efficient optimization methods implemented in the semiempirical package AMPAC are presented. They concern the minimization of the energy or of the gradient norm by either pseudo-Newton or quadratic procedures, the search for transition states, and the intrinsic reaction coordinate in conjunction with variational transition-state theories. Nonlocal methods such as simulated annealing are also introduced. 0 1992 John Wiley & Sons, Inc.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
51
0

Year Published

1994
1994
2017
2017

Publication Types

Select...
5
4

Relationship

0
9

Authors

Journals

citations
Cited by 83 publications
(51 citation statements)
references
References 92 publications
0
51
0
Order By: Relevance
“…The most marked difference lies in the orientation of the chain at C(2). According to these authors, the most stable structure is that in which H (20) forms a H-bond with O(1). The AM1 energy difference between this structure and the most stable one is 0.3 kcal/mol; such a small difference suggests that both conformations are equally probable in the gas phase.…”
Section: Dmentioning
confidence: 98%
See 1 more Smart Citation
“…The most marked difference lies in the orientation of the chain at C(2). According to these authors, the most stable structure is that in which H (20) forms a H-bond with O(1). The AM1 energy difference between this structure and the most stable one is 0.3 kcal/mol; such a small difference suggests that both conformations are equally probable in the gas phase.…”
Section: Dmentioning
confidence: 98%
“…The geometries for stationary points were identified by minimization of the energy with respect to all geometric parameters using the Broyden-FletcherGoldfarb-Shanno algorithm [ 14-1 81. The approximate geometries for the transition states (TS) were refined by minimizing the gradient norm of energy using the Powell algorithm [19], followed by the full Newton algorithm (LTRD) [20]. At a later stage, stationary points were characterized by FORCE or LTRD frequency analysis, a zero index for a critical point (where 'index' refers to the number of strictly negative eigenvalues of the Hessian) being assigned to reactants, products, and intermediates, and a unity value to each TS.…”
mentioning
confidence: 99%
“…[18] Geometries of stationary points were determined by minimization of energy with respect to all geometric parameters. Transition states were located by the chain method [19] and characterized by one, and only one, negative eigenvalue of the Hessian matrix. Finally, the transitions states found were shown to belong to the studied reaction by intrinsic reaction coordinate (IRC).…”
Section: Theoretical Studiesmentioning
confidence: 99%
“…The procedure we use is an extension of the simulated annealing method to potential energy surfaces which present several significant minima. It is similar to the procedure developed by D. Liotard [17] based on a random exploration of the potential energy surface by using the metropolis algorithm [18]. The obtained configurations are sorted out and the selected configurations are used as starting points for a local minimization method (quasi-Newton method, BFGS [19]).…”
mentioning
confidence: 99%