Hello,
I am having some trouble running basis set superposition error
corrections in a slab + adsorbate geometry with synthetic atoms
saturating bottom of the slab.
I have followed the helpful instructions from Abraham posted regarding
setting up BSSE corrections calculations:
http://www.mail-archive.com/[email protected]/msg02916.html
However, I have run into trouble when attempting to run the calculation
with the adsorbate and ghost atoms substituted for the surface. As was
posted previously
(http://www.mail-archive.com/[email protected]/msg01507.html), the program
exits if Z < -200 with the error: "QVLOFZ: ERROR: Z out of range" so I
can't run with the same synthetic atoms pseudopotentials as the standard
surface calculations.
My synthetic atoms are merely 1.25 charged H, so I attempted replacing
the synthetic pseudopotential with a standard H.psf. I also tried
editing the fractional H psf file to only have 1.00 occupation instead
of 1.25 (this shouldn't matter since it's just the basis set being used,
right?). Both of these calculations ran successfully, but resulted in
energies vastly different than the adsorbate alone (-506 eV vs. -479 eV
for the water adsorbate). The surface error correction (E of surface &
ghost adsorbate - E of surface alone) came out to a rational -0.22 eV.
Here are my questions:
1) Is there something I can edit in the source code which would let me
use the proper synthetic atoms? I traced the error message to
src/periodic_table.f but am unclear how to connect that to synthetic
atom definitions.
2) How much of the BSSE correction am I loosing by not including the
correction for the adsorbate (a single water molecule)? Will this be the
same order of magnitude as the surface BSSE? And should it be the same
sign (i.e. raise the magnitude of the correction)?
3) Is it rational to calculate the BSSE correction by keeping the bottom
layers of the slab (as well as the synthetic H), which are kept fixed in
all these calculations anyway, fully described in all calculations? I.e.
my BSSE calculations would consist of
1) SCF of surface only with position fixed to adsorbate geometry
2) SCF of adsorbate and bottom layers of surface within ads geometry.
3) SCF of surface plus ghost states of adsorbate
4) SCF of adsorbate and bottom layers of slab w/ ghost states of top
layers of slab.
Any help would be appreciated,
--
Alejandro Suarez (Research Group of Tom Reinecke)
Postdoctoral Associate
Naval Research Laboratory
Washington, D.C. USA
