| 7:02:27 AM PDT - Tue, Apr 28th 2015 |
|
Thank you for inspecting the patch. Yes, comparing this density to the output from another QM code seems OK, as well as to the output from a dplot calculation (which is not supported via the python interface):
dplot
GAUSSIAN
TITLE Density
LimitXYZ
-4.611 4.718 107
-5.058 4.295 108
-4.042 4.042 93
spin total
output density.cube
end
Visual inspection via VMD also seems OK.
A little background on why I'm doing this. I'm essentially carrying out multiple MEP calculations on the same molecule at different orientations to average discretization effects from the projection onto a grid. I am using the MEP at a specific isosurface at which the electron density has a value of 0.002 (hence the need for these two cubes). This MEP isosurface is then used for a "coarse grained" abstraction reducing the molecule down to a set of Surface Site Interaction Points (SSIPs). More about this approach can be found here: http://dx.doi.org/10.1039/C3SC22124E
I am thinking perhaps of perhaps submitting a patch so that NWChem code so that for the property esp, one can pass an option to only calculate the MEP values at a specific electron density isovalue (within a tolerance), but this may require a bit more work than the patch above to write just the electron density.
As an aside, the python interface is extremely useful for bespoke and non-standard calculations; I really appreciate this functionality.
|