Author Topic: eht  (Read 8211 times)

Kiwinator

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
eht
« on: December 05, 2007, 11:18:07 AM »
Hi

I can't manage a eht guess for ecp's for ten electrons for Ca, for example the LANL2-DZ-ECP?

What might be the reason??

christof.haettig

  • Global Moderator
  • Sr. Member
  • *****
  • Posts: 291
  • Karma: +0/-0
    • Hattig's Group at the RUB
Re: eht
« Reply #1 on: January 02, 2008, 11:49:25 AM »
Hi,

LANL2DZ uses the Hay/Wadt 10 electron ECP. For this you need a EHT defintion prepared for a 10 electron ECP,
which is not (yet) included in TURBOMOLEs basis set library. But you can add this yourself to the file $TURBODIR/basen/ca:

1.) replace in this file the $ehtdef section by:

$ehtdef         definitions=3
1S(DZ)                                    basis=ca DZ
ECP-10               ncore=10  basis=ca LANL2DZ ECP
1S(eht,ecp-18)  ncore=18  basis=ca ecp-18 dz hay & wadt

2.)  add to the $ehtao section the following lines:

ECP-10   ehtdata   (4d20.14)
# SCF energy is      -36.2139710757 a.u.  (virial theorem =  1.361646110)
     1  s      eigenvalue=-.22319791919905D+01   nsaos=3
0.99633015174826D+00-.73944165495912D-02-.77273128854640D-02
     2  s      eigenvalue=-.19613367930300D+00   nsaos=3
-.18607984242445D-030.37381128351125D+000.91740592073086D+00
     1  p      eigenvalue=-.13519016695886D+01   nsaos=3
0.99552403492089D+000.13900062855501D-01-.27930273950130D-03
*

Christof

Kiwinator

  • Newbie
  • *
  • Posts: 5
  • Karma: +0/-0
Re: eht
« Reply #2 on: January 17, 2008, 02:30:34 PM »
Thanks for the extensive answer!

Kiwinator