[Pw_forum] Question About Lowdin Charges

Thu Oct 23 08:32:32 CEST 2014

Dear Victor,

I've recently run into the same problem with the Lowdin
projection. Similarly to your case, I also found that some species that
were supposed to be charged a certain way were in fact charged in the
opposite sense (and also the total charge on some gas phase molecules to
not add up to what they are supposed to be in spite of a tiny spilling
parameter). I suggest that you do Bader charge analysis instead off of a
.cube file that pp.x generates. Lowdin anyway tends to underestimate
partial charges. Beware though that since you are using non-PAW
pseudopotentials, some of your Bader charges may come out to be zero. This
results from the fact that sometimes the Bader decomposition of space into
atomic volumes does not manage to find a good extremum in the charge
density due to the lack of enough semicore states. You may then need more
semicore states in your pseudopotentials.

Hope this helps.
Hande Toffoli

On Wed, 22 Oct 2014, Vic Bermudez wrote:

> Greetings All,
>
> I'm doing a two-dimensionally-periodic slab calculation for the GaSb (001)
> surface, and I'm trying to make sense out of the Lowdin charge results from
> projwfc.x.  This is a segment of the relevant output (slightly edited for
> clarity) for atoms in the "bulk" of the slab:
>
> Sb
>     Atom #  17: total charge =  14.9047, s =  1.6852, p =  3.2209, d =
> 9.9986,
>     Atom #  18: total charge =  14.8065, s =  1.4984, p =  3.3097, d =
> 9.9985,
>     Atom #  19: total charge =  14.7370, s =  1.5752, p =  3.1631, d =
> 9.9988,
>     Atom #  20: total charge =  14.9047, s =  1.6852, p =  3.2209, d =
> 9.9986,
>
> Ga
>     Atom #  21: total charge =  13.2248, s =  1.1862, p =  2.0448, d =
> 9.9938,
>     Atom #  22: total charge =  13.2300, s =  1.1914, p =  2.0447, d =
> 9.9939,
>     Atom #  23: total charge =  13.2174, s =  1.2033, p =  2.0194, d =
> 9.9947,
>     Atom #  24: total charge =  13.2248, s =  1.1862, p =  2.0448, d =
> 9.9938,
>
> The free Sb atom has a valence occupancy of d(10), s(2), p(3), and the Ga is
> d(10), s(2), p(1). Since the "total charge" reported by projwfc.x is the
> total electron occupancy, what I think I'm seeing is an Sb charge of about
> 15 - 14.8 = +0.2 and a Ga charge of about 13 - 13.2 = -0.2. This is the
> reverse of what it should be. Sb is the anion, which is why the material is
> called "gallium antimonide" and not "antimony gallide". I've looked at
> previous forum posts on this subject, and I believe that my interpretation
> of the Lowdin numbers is consistent with these discussions.
>
> Any advice would be much appreciated.
>
> Thanks,
> Vic Bermudez
>
> Victor M. Bermudez
> Code 6877
> Naval Research Laboratory
> 4555 Overlook Ave., S.W.
> Washington, DC 20375-5347
>
> victor.bermudez at nrl.navy.mil
> Phone: 202-767-6728
> FAX: 202-767-1165
>
> P.S.: In case it helps, here is the input (with a lot of atom positions
> deleted) for the nscf run that generated the input used by the projwfc.x
> run:
>
> &CONTROL
> calculation='nscf',
> restart_mode='restart',
> title='prepare DOS input for bare surface; #62617',
> pseudo_dir='/lustre/cmf/scratch/b/bermudez/QE_PP/',
> outdir='/lustre/cmf/scratch/b/bermudez/SurfDOS',
> wf_collect=.TRUE.,
> verbosity='default',
> tefield=.TRUE.,
> dipfield=.TRUE.
> /
>
> &SYSTEM
> ibrav=8,
> a=17.1119841047, b=12.8339880785, c=44.0,
> nat=128,
> ntyp=3,
> ecutwfc=30.0,
> ecutrho=300.0,
> london=.TRUE.,
> nosym=.TRUE.,
> occupations='tetrahedra',
> edir=3,
> eamp=0.0,
> emaxpos=0.5,
> eopreg=0.1,
> input_dft='pw91'
> /
>
> &ELECTRONS
> electron_maxstep=2000,
> conv_thr=1.0D-9,
> mixing_mode='local-TF'
> /
>
> ATOMIC_SPECIES
> Ga   69.72   Ga.pw91-n-van.UPF
> Sb  121.75   Sb.pw91-n-van.UPF
> H5    1.25   H.pz-vbc_125.UPF
>
> ATOMIC_POSITIONS angstrom
> Sb       2.138997330   5.022847724   6.773542341
> Ga       2.138997871  -5.172774747   5.920904901
> Ga       6.303916687   1.112378442   5.961021454
> Sb       6.422475638   3.685129656   6.784913634
> Ga      -6.416994681   1.257691244   6.085010916
> Sb      -6.416994193   3.824801392   6.836466153
> Ga      -2.025921740   1.112378091   5.961021393
> Sb      -2.144480532   3.685129427   6.784912981
> Sb       0.309752603   0.016789605   5.583506326
> Sb       0.021948893   4.297405309   5.001534742
> Sb      -0.337617879  -4.315074540   5.610294615
> .
> .
> .
> H5       5.198446115   0.000002191  -6.285644236    0   0   0
> H5       5.198445769   4.277998300  -6.285643824    0   0   0
> H5       5.198445582  -4.278000175  -6.285643213    0   0   0
> H5      -7.635543283  -0.000001494  -6.285642838    0   0   0
> H5      -7.635543358   4.277997263  -6.285642702    0   0   0
> H5      -7.635543780  -4.277997026  -6.285642281    0   0   0
> H5      -3.357548712  -0.000000728  -6.285641379    0   0   0
> H5      -3.357548071   4.277997544  -6.285641893    0   0   0
> H5      -3.357547485  -4.277997519  -6.285642688    0   0   0
>
> K_POINTS automatic
> 6 8 1 0 0 0
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>

--
Hande Toffoli
Department of Physics
Office 439
Middle East Technical University
Ankara 06531, Turkey
Tel : +90 312 210 3264
http://www.physics.metu.edu.tr/~hande