Double-Ended Searches

OPTIM.2.3 introduced two keywords to provide double-ended pathway searches. The first is NEB, an implementation of the revised nudged elastic band approach.[24,25] The second is CONNECT, which can also find pathways connecting two minima using a different strategy. In both cases the coordinates of the starting minimum are specified in the odata file, and the coordinates of the final minimum are specified in the file finish. CONNECT builds up a complete min-sad-min-$\cdots$-min sequence by systematically searching for transition states and calculating pathways. The PATH keyword is not needed, but can be used to change the number of points files saved in the path.$n$.xyz files, which are saved for each transition state, $n$. The energy profiles are saved in files EofS.$n$.

If CONNECT needs to connect permutational isomers of the same minimum in the course of a run then the initial guess for the transition state geometry is generated using the NEB routine with just two images, to avoid cold fusion problems.

CONNECT generally needs to be augmented by other keywords to specify how the transition state geometries should be guessed (NEB, NEWNEB or FIXD), how the transition state searches should be performed (SEARCH 2 or BFGSTS, with or without NOHESS, NOIT etc.), and how that pathways should be calculated (SEARCH 0, 6 or 7, BFGSMIN, RKMIN or BSMIN). Most of these combinations should work together. A summary file will be produced if DUMPPATH is specified. An xyz file for the overall path will be printed to path.xyz and the energy as a function of path length is printed to EofS. The corresponding xyz and energy files for the individual steps in the path are numbered path.$n$.xyz and EofS.$n$ for transition state $n$.

When NEB is used in a CONNECT run its function is merely to produce a starting guess for the transition state geometry, from which a transition state search of some kind is then initiated. It seems that the higher energy parts of a path converge faster than the lower regions in NEB calculations, and so it is possible to generate a transition state guess using only a few images and sloppy convergence in the NEB part of the calculation.

In OPTIM3 the NEB and CONNECT keywords were augmented by new algorithms, which are selected using NEWNEB and NEWCONNECT. NEWNEB can work on its own and with both CONNECT and NEWCONNECT. However, NEWCONNECT cannot be used with the old NEB. The new algorithms are based on the doubly-nudged elastic band approach and a more sophisticated connection algorithm.[14]

The latest double-ended search methods to be implemented correspond to the GROWSTRING and EVOLVESTRING keywords, which provide implementations of the growing string and evolving string methods.[15,16]