Transition state search with DL-FIND
There are two algorithms available in DL-FIND to search for transition states: partitioned rational function optimisation (P-RFO) and the dimer method.
The P-RFO method
P-RFO is a conventional transition state optimisation algorithm which involves a full Hessian calculation to find the transition state mode to follow. It is therefore only suitable for the optimisation of small molecules.
The example optimisation (prfo.chm) follows a similar pattern to the minimisation example, except it searches for a maximum in one dimension (i.e. a transition state).
First, we create an initial geometry as usual. The input geometry to DL-FIND must be in Cartesians, so if you start with a z-matrix as in the example then you need to convert it to Cartesians using z_to_c.
z_create zmatrix=z {
zmatrix angstrom
n
x n nx
h1 n nh x xnh
h2 n nh x xnh h1 120.
h3 n nh x xnh h1 240.
constants
nx 1.0
variables
nh 1.0
xnh 85.0
end
}
z_to_c zmatrix=z coords=c
Alternatively you can of course create an initial Cartesian geometry using c_create or read it in from an XYZ file as in the previous DL-FIND examples.
Once the starting geometry is set up, DL-FIND is called with the option optimiser=prfo to specify a P-RFO transition state optimisation.
dl-find coords=c result=opt_prfo.c \
coordinates=dlc optimiser=prfo \
theory=mndo : accuracy=high
The optimised result in opt_prfo.c is then written out to XYZ format.
The dimer method
The dimer method defines two connected geometry images which are rotated to determine the direction to the transition state. This avoids the need to calculate a Hessian and is therefore recommended for optimising systems with many degrees of freedom.
The dimer method can be chosen by specifying dimer=true as in the example dimer.chm. The recommended optimiser for use with the dimer method is lbfgs:
dl-find coords=c result=opt_dimer.c \
coordinates=dlc dimer=true optimiser=lbfgs \
theory=mndo : accuracy=high
You should find that the dimer method converges to the same final energy and structure (opt_dimer.c) as P-RFO. Due to the approximations made by the dimer method you may find it takes a few more cycles to converge than P-RFO, but this is usually far outweighed by the cost of calculating the Hessian in P-RFO.
For the full list of options that can be selected when using the transition state search methods, please consult the DL-FIND manual pages.
Both the P-RFO and dimer methods require a reasonable initial guess at the transition state structure in order to converge efficiently. In the next tutorial we will look at a method which only requires knowledge of the reactant and product states.