Molecules breaking during geometry scan


Just Got Here
Hello -

I have been attempting to use nwchem to calculate the energy gradient as a solvent molecule is moved towards an ion, using the repeating input block below:

Quote:
geometry adjust
zcoord
bond 1 2 1.4 oyb
end
end

set int:acc_std 1D-16
charge 3
dft
odft
mult 8
grid xfine
iterations 800
vectors input Gd-meoh_1.6.movecs output Gd-meoh_1.4.movecs
xc b3lyp
decomp
end
task dft gradient
geometry adjust
zcoord
bond 1 2 1.2 oyb
end
end


However, I have been encountering problems when the molecule is closer to the ion than its equilibrium distance. The oxygen atom will move towards the ion (that is the adjust bond 1 2 that I specify), but the molecule will break - the rest of the molecule (in this case, a methyl group and a proton) will be left behind. Obviously this will not generate a useful energy gradient.

How can I fix the solvent molecule so that it will not break? Thanks!


Forum >> NWChem's corner >> General Topics