iOpenShell » Technical questions » using PBE0 in G03

using PBE0 in G03

Moderators: kadir, krylov, piotr.

Page: 1

Author Post
Member
Registered: Sep 2007
Posts: 11
I'm looking to do some PBE0 calculations with G03 (for later use in Stone's GDMA program), but G03 doesn't support the PBE0 functional it looks like. So I've tried to "build" the functional according to G03's online manual. As best as I can figure out, it should look something like this:

# PBEPBE IOp(3/76=100002500) IOp(3/77=075000000)

(For reference, PBE0 is just the PBE functional with 25% HF exchange mixed in.)

However, when I ran this, I got the following error:

PBE exchange requires equal scale factors.

I was wondering if anyone has any advice or has done this in the past?
Thanks!
Glen
Administrator
Registered: Aug 2007
Posts: 200
Hi Glen!
Did you check "Jaguar". I remember long time ago I played a bit with it to construct DFT functionals that I couldn't construct with other programs. Jaguar used to give the flexibility of building whatever you like.
Member
Registered: Sep 2007
Posts: 11
I'll give that a shot. Thanks!
Member
Registered: Dec 2007
Posts: 14
Location: Tehran
I think PBE0 is implementeable with G03, although some other packages may include broader variety of functionals. Also, the iop - IOp(3/76)- seems to be correct at the first sight, but, I suggest looking at the original publications of the method developers. In this way you can make sure of others essential options for calculations. Generally some other options such as requesting tight convergence for SCF calculations is needed for this kind of DFT methods, i.e. hybrid DFT methods. One another option is increasing the numbers of grip point in the integrations. By the way, you can email to the users of this functional as a last resort, too.!! ;) Please let me know if you get it. :mrgreen:
Member
Registered: Sep 2007
Posts: 8
glen wrote
I'm looking to do some PBE0 calculations with G03 (for later use in Stone's GDMA program), but G03 doesn't support the PBE0 functional it looks like. So I've tried to "build" the functional according to G03's online manual. As best as I can figure out, it should look something like this:

# PBEPBE IOp(3/76=100002500) IOp(3/77=075000000)

(For reference, PBE0 is just the PBE functional with 25% HF exchange mixed in.)

However, when I ran this, I got the following error:

PBE exchange requires equal scale factors.

I was wondering if anyone has any advice or has done this in the past?
Thanks!
Glen


Hello Glen!
in Gaussian PBE1PBE = PBE0
Just google it. ;) You will find evidences and correct scitation.
Member
Registered: Oct 2008
Posts: 3
I suggest looking at the original publications of the method developers. In this way you can make sure of others essential options for calculations. Generally some other options such as requesting tight convergence for SCF calculations is needed for this kind of DFT methods, i.e. hybrid DFT methods.
« Last edit by iopenshell on Thu Oct 23, 2008 5:02 pm. »
Member
Registered: Oct 2008
Posts: 3
Nobody is replying is there any problem?

Page: 1

iOpenShell » Technical questions » using PBE0 in G03