[Pw_forum] Pseudopotential problem about Mg

Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu
Fri Mar 2 06:09:16 CET 2007

On Fri, 2 Mar 2007, Lawrence Lee wrote:

LL> Dear all,

LL> I did a molecular dynamics run using pw.x on MgO, using
LL> pseudopotentials generated myself. The results turn out to be quite
LL> different from literature. I suspect that it's my pseudopotential's
LL> problem. Here below is the input file for ld1.x. Since I am green in
LL> generating pseudopotential, can anyone point out any problems in my
LL> input file, if there are any?

Mg is a tough customer in terms of pseudopotentials.
i've been trying to get a good Mg potential on and off
for quite a while now. so it is probably better you start 
with practicing on some easier elements.

also, for MgO you'll need to include semi-core states.
a minimal pseudo will essentially be stripped off almost
all its electron density and is likely to give bad results.


LL> Lawrence.
LL>  &input
LL>      title='Mg',
LL>      prefix='mg_lda_s1.5p0d0_n'
LL>      zed=12,
LL>      rel=1,
LL>      beta=0.2,
LL>      rlderiv=2.5,
LL>      eminld=-6.0,
LL>      emaxld=6.0,
LL>      deld=0.02d0,
LL>      nld=3,
LL>      config='[Ne] 3s1.5 3p0 3d0',
LL>      iswitch=3,
LL>      dft='LDA',
LL>  /
LL>  &inputp
LL>    pseudotype=2,
LL>    lloc=2,
LL>   !rcloc=1.00
LL>    file_pseudopw='Mg_lda_s1.5p0d0_n.upf',
LL>    nlcc=.true.,
LL>  /
LL> 3
LL> 3S  1  0  1.50  0.00  1.90 1.90
LL> 3P  2  1  0.00  0.00  1.90 1.90
LL> 3D  3  2  0.00  0.00  1.90 1.90
LL>  _______________________________________
LL>  YM - Â÷½u°T®§
LL>  ´Nºâ§A¨S¦³¤Wºô¡A§AªºªB¤Í¤´¥i¥H¯d¤U°T®§µ¹§A¡A·í§A¤Wºô®É´N¯à¥ß§Y¬Ý¨ì¡A¥ô¦ó»¡¸Ü³£ÉN¨«¥¢¡C
LL>  http://messenger.yahoo.com.hk

Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
If you make something idiot-proof, the universe creates a better idiot.

More information about the users mailing list