Jmol for NWChem
NWChem 4.6 is supported. Older versions may work too, but have not been tested. Since the different modules in NWChem can give quite different formatted output and has all kinds of print options for each, it is not guaranteed that any type of combination of task(s) will be properly parsed. Most testing has been done with SCF, MP2 and DFT optimizations and frequency calculations.
The IO reads the all structures, sequentially, and some additional information.
Since this can result in a large amount of structures, they are collected into groups of atom sets, most easily selected using the AtomSetChooser.
Every task is collected in its own branch.
Note that if before the end of a task no geometry output is encountered, no branch will be created for it. This occurs in cases where one first does something like:
task dft optimize; task tddft energy;
In case of a geometry optimization, all structures associated with the optimization trajectory are read.
Since NWChem allows for suppression of output, the number of branches generated by interpreting the geometry optimization output can differ. Maximally an Input, Geometry, and Gradients branch may be generated. The structures in that have gradients also have the vector information set to the gradient itself (this is also flagged in the properties of the atom set).
Additional information interpreted per step:
- Step number
- Symmetry group name
When the output of a frequency task is encountered, a new 'Frequencies' branch containing all the frequencies is generated in the task branch.
Only the projected frequencies are interpreted. Again, maneuvering through all this is most easily done using the AtomSetChooser.
Additional information interpreted with each vibration:
- IR Intensity
When a Mulliken analysis of the total density is encountered, the charge on the atoms are assigned too.