hydrogen_bond.py: Binding Energy of a Hydrogen-Bonded Complex

The script hydrogen_bond.py can be used to calculate the total binding energy of a hydrogen-bonded complex of two molecules. See Binding Energies of Hydrogen-Bonded Complexes for details on the computational protocol. You can use this script from Maestro, as described in Binding Energies of Hydrogen-Bonded Complexes, or from the command line. The command syntax is as follows:

jaguar run hydrogen_bond.py [options] filenames

where filenames are one or more Jaguar input files or Maestro structure files (only one file type, not a mixture of both). For a description of the options, see hydrogen_bond.py or run the command with the -h option.

For each complex, a job directory called name_hydrogen_bond is created in the current working directory, where name is the stem of the structure file name. The job files for that complex are written to this directory. A file called name_hydrogen_bond.out is written to the current working directory when the job finishes. The content of this file should be self-explanatory:

--------------------------------------------------------------------
Hydrogen bond energy calculation for h2o-h2co.in
Energy units are kcal/mol
 
BSSE-corrected cc-pVTZ(-f) binding energy:  -3.98
  (cc-pVTZ(-f) BSSE correction:   1.08)
BSSE-corrected cc-pVQZ(-g) binding energy:  -4.37
  (cc-pVQZ(-g) BSSE correction:   0.41)
 
              Extrapolated binding energy:  -4.77
--------------------------------------------------------------------

The binding energy calculated with the two basis sets is printed, along with the corresponding BSSE corrections. As the size of the basis set increases, the size of the BSSE is expected to decrease, and the binding energy is expected to increase in magnitude. The binding energy extrapolated to the basis set limit is printed last. For especially weakly-bound systems, it occasionally happens that the energy obtained with the cc-pVTZ(-f) basis set is slightly lower than that obtained with the larger cc-pVQZ(-g) basis set1. In this situation the usual two-point basis set extrapolation formula cannot be used, so the energy is instead corrected using a simple linear correction of the LMP2/cc-pVQZ(-g) energies to the CCSD(T) reference energies. The output file clearly indicates when this has happened. The binding energy will still be quite accurate to within 0.5 kcal/mol when compared to the CCSD(T) energy.