The CPMCSCF is best used when you are dealing with state-averaged orbitals, like it says in the manual. You seem to be only doing a ground state calculation, as far as I can see, and CPMCSCF seems unnecessary. You might want to compare the geometries, basis sets and rotational constants closely between the molpro and gaussian input to make sure you are looking at the same calculation. Often in CASSCF, you want to be careful about exactly what the active space actually is between the two inputs.<br>
Cheers,<br>Jayashree<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<br>
Today's Topics:<br>
<br>
1. two questions: freq for CASSCF and gradients for QCISD(T)<br>
(<a href="mailto:eurisco1@pochta.ru">eurisco1@pochta.ru</a>)<br>
2. Re: wrong PBE0 energies (Fred Manby)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Sun, 22 Nov 2009 15:14:09 +0300<br>
From: <a href="mailto:eurisco1@pochta.ru">eurisco1@pochta.ru</a><br>
Subject: [molpro-user] two questions: freq for CASSCF and gradients<br>
for QCISD(T)<br>
To: <a href="mailto:molpro-user@molpro.net">molpro-user@molpro.net</a><br>
Message-ID: <<a href="mailto:1f8acfcaf3515a5f32861e7e19cc5370771791d8@www.pochta.ru">1f8acfcaf3515a5f32861e7e19cc5370771791d8@www.pochta.ru</a>><br>
Content-Type: text/plain; charset="koi8-r"<br>
<br>
<br>
<br>
<br>
<br>
<br>
Dear Molpro Users and Developers,<br>
<br>
<br>
Would you allow me to address to you<br>
necessary questions:<br>
<br>
<br>
<br>
I have calculated frequencies<br>
(numerically) as well as analytically by using routes on Version<br>
2008.1 linked 05 Sep 2008 (trial) at the level CAS(10,10)/cc-pVDZ<br>
<br>
<br>
<br>
First input:<br>
<br>
<br>
geomtyp=xyz<br>
<br>
<br>
geometry={<br>
<br>
6<br>
<br>
<br>
<br>
<br>
<br>
my molecule _xyz<br>
<br>
}<br>
<br>
<br>
basis=VdZ<br>
<br>
<br>
{hf;wf,46,1,0,0}<br>
<br>
{multi;occ,28;closed,18;wf,46,1,0,0;cpmcscf,hess}<br>
<br>
{frequencies} !do<br>
<br>
<br>
---<br>
<br>
<br>
Second input:<br>
<br>
geomtyp=xyz<br>
<br>
<br>
geometry={<br>
<br>
6<br>
<br>
<br>
<br>
<br>
<br>
my molecule _xyz<br>
<br>
}<br>
<br>
<br>
basis=VdZ<br>
<br>
<br>
{hf;wf,46,1,0,0}<br>
<br>
{multi;occ,28;closed,18;wf,46,1,0,0}<br>
<br>
{frequencies;<br>
<br>
<br>
print,low,img;} !do<br>
<br>
and I have always observed only real<br>
frequencies (which were slightly different). But if I used Gaussian<br>
(both 03 and 09) to calculate (numerically) frequencies I observed<br>
only one imag. (~100i) frequency (I performed IRC and achieved both<br>
valleys).<br>
<br>
<br>
<br>
<br>
Why I got two different results? How<br>
can I treat this situation?<br>
<br>
<br>
<br>
In the Molpro manual states that<br>
analytical gradients are available<br>
for qcisd(t). I have tested simple input<br>
<br>
<br>
<br>
! $Revision: 2008.1<br>
<br>
<br>
***,h2o !A title<br>
<br>
memory,150,m<br>
<br>
r=2.5,theta=120<br>
<br>
geometry={o;<br>
<br>
h1,O,r;<br>
<br>
h2,O,r,H1,theta<br>
<br>
}<br>
<br>
<br>
basis=vdz<br>
<br>
<br>
rhf<br>
<br>
rqcisd(t)<br>
<br>
<br>
optg<br>
<br>
<br>
<br>
<br>
<br>
but I have observed only messages about<br>
numerical calculations of gradients<br>
<br>
Numerical gradient completed.<br>
CPU-time: 0.75 sec, Elapsed: 2.50 sec<br>
<br>
<br>
What is wrong in my input? Are<br>
analytical gradients available for geometry optimizations if qcisd(t)<br>
method is used for calculation of energy?<br>
<br>
Sincerely,<br>
<br>
<br>
PhD student Oleg B. Gadzhiev<br>
<br>
<br>
ps: I have downloaded molpro 5 days<br>
ago.<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://www.molpro.net/pipermail/molpro-user/attachments/20091122/8ee019f7/attachment.html" target="_blank">http://www.molpro.net/pipermail/molpro-user/attachments/20091122/8ee019f7/attachment.html</a><br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Mon, 23 Nov 2009 10:10:22 +0000<br>
From: Fred Manby <<a href="mailto:fred.manby@bristol.ac.uk">fred.manby@bristol.ac.uk</a>><br>
Subject: Re: [molpro-user] wrong PBE0 energies<br>
To: <a href="mailto:molpro-user@molpro.net">molpro-user@molpro.net</a><br>
Message-ID: <<a href="mailto:B8EE3409-7FB1-407A-ADFC-3093629F6B6F@bris.ac.uk">B8EE3409-7FB1-407A-ADFC-3093629F6B6F@bris.ac.uk</a>><br>
Content-Type: text/plain; charset="utf-8"<br>
<br>
Dear Wenli<br>
<br>
Many thanks for alerting us to this discrepancy. I have reproduced the<br>
problem (using yet another code, Turbomole) and will fix the problem.<br>
<br>
Best wishes<br>
<br>
Fred<br>
<br>
On 13 Nov 2009, at 17:30, Zork Zou wrote:<br>
<br>
> Dear all,<br>
><br>
> I performed some DFT calculations, and found that the PBE0 energies<br>
> are different from the results of other programs (Gaussian, Gamess,<br>
> Dalton, ?). For example,<br>
><br>
> Ne atom: STO-3G<br>
> -126.88397 (Molpro)<br>
> -126.85545 (other)<br>
><br>
> He atom: 6-31G<br>
> -2.89155 (Molpro)<br>
> -2.88769 (other)<br>
><br>
> I also checked the BLYP and B3LYP functionals. The energies agree<br>
> very well. So there must be something different in Molpro?s PBE0<br>
> from the other programs.<br>
><br>
> Wenli<br>
><br>
><br>
> ?????????????????<br>
> _______________________________________________<br>
> Molpro-user mailing list<br>
> <a href="mailto:Molpro-user@molpro.net">Molpro-user@molpro.net</a><br>
> <a href="http://www.molpro.net/mailman/listinfo/molpro-user" target="_blank">http://www.molpro.net/mailman/listinfo/molpro-user</a><br>
<br>
--<br>
Dr Fred Manby<br>
Centre for Computational Chemistry<br>
School of Chemistry, University of Bristol<br>
Bristol BS8 1TS, UK<br>
+44 117 92 (87664)<br>
<a href="http://www.chm.bris.ac.uk/pt/manby/" target="_blank">http://www.chm.bris.ac.uk/pt/manby/</a><br>
<br>
Computational Molecular Science 2010: <a href="http://www.chm.bris.ac.uk/cms" target="_blank">http://www.chm.bris.ac.uk/cms</a><br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://www.molpro.net/pipermail/molpro-user/attachments/20091123/107d7227/attachment-0001.htm" target="_blank">http://www.molpro.net/pipermail/molpro-user/attachments/20091123/107d7227/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
Molpro-user mailing list<br>
<a href="mailto:Molpro-user@molpro.net">Molpro-user@molpro.net</a><br>
<a href="http://www.molpro.net/mailman/listinfo/molpro-user" target="_blank">http://www.molpro.net/mailman/listinfo/molpro-user</a><br>
<br>
<br>
End of Molpro-user Digest, Vol 16, Issue 19<br>
*******************************************<br>
</blockquote></div><br><br clear="all"><br>-- <br>Jayashree<br>