[Pw_forum] Some efg.x confusions cleared up

[Josef Zwanziger]

Dear Colleagues--

last month I posted some questions and concerns about the efg.x code.
Here's an update--first, concerning units of the quadrupole moment, I
found that the expressions in efg.x are correct after all, provided that
you interpret the input unit as 10E-30 m^2 (the documentation is a
little unclear on this, but no matter). Therefore, for oxygen-17 the
correct input is 2.58, as example24 uses. I was able to trace the issue
back to faulty data published on www.webelements.com, where there is in
fact a confusion about the definition of the unit of barns (a little
disturbing, as this is probably the best known web site for beginning
chemistry students to assemble data).

Second, concerning the pseudopotentials used in example24, I was finally
able to reconstruct the appropriate input files for oxygen, following
those for silicon in atomic_doc/pseudo-gen, so that using ld1.x I can
make both Si, O, and .recon files that reproduce the example24 data.
However, life is quite simple yet--if I change the pseudopotentials in
any way, say by using lloc=2 instead of lloc=1, or using 2 projectors in
each channel, the computed Cq's for oxygen in SiO2 change by orders of
magnitude. I'm still trying to understand why this works this way.

Best,

Joe Zwanziger

Josef W. Zwanziger                                
Prof. of Chemistry
Canada Research Chair in NMR Studies of Materials
Director, Atlantic Region Magnetic Resonance Centre
Dalhousie University
Halifax, NS B3H 4J3=20
Canada

tel: 902.494.1960     net: jzwanzig at dal.ca
fax: 902.494.1867     web: http://jwz.chem.dal.ca