modified on 11 September 2014 at 15:03 ••• 1,102 views

Release66:QMMM Transition States

From NWChem

Revision as of 15:03, 11 September 2014 by Silvia (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

QM/MM transition states calculations for qm or qmlink regions can be performed using

task qmmm <qmtheory> saddle

The overall algorithm is very similar to QM/MM optimization calculations, but instead of optimization, transition state search will be performed for qm or qmlink region for specified number of steps ( as defined by maxiter keyword). The remaining classical regions (if any) will be optimized following the standard optimization protocol, which may involve, if specified, ESP charge representation of the QM atoms (a recommended option).


Here is an example of direct transition state calculation on the chloroform and OH- solvated species.

 memory total 2000 Mb

 start clfoh

 permanent_dir ./perm

 scratch_dir   ./data
 charge -1

 basis "ao basis"
  * library "6-31+G*"
 end

 dft
 XC b3lyp
 iterations 5000
 end

 md
  system clfoh_ts
  cutoff 1.5
 end

 qmmm
 bqzone 15.0
 region  qm solvent
 xyz  ts
 maxiter 10 1000
 ncycles 2
 density espfit
 end

 task qmmm dft saddle

The success transition state calculations is strongly dependent on the initial guess. User may consider generation of the latter using QM/MM reaction pathway calculation. Another useful strategy involves precalculation of the Hessian. Following the example presented above one could have precalculated numerical Hessian for the qm region

...
qmmm
region qm
end

freq
animate
end

task qmmm dft freq

and then used this information in the TS calculation

...
driver
 clear
 inhess 2  #read in hessian from perm directory
 moddir 1  #follow 1st mode
end

qmmm
 bqzone 15.0
 region  qm solvent
 xyz  ts
 maxiter 10 1000
 ncycles 2
 density espfit
end

task qmmm dft saddle