[molpro-user] Convergence Problem in MCSCF Calculations

[Sunil]

Dear Sir,

Thank you for your helping hand. I am getting convergence by choosing
distances above 0.9 Ang. However, there is convergence problem for distances
less than or equal to 0.8 Ang. I think, 0.8 or 0.7 Ang is not too small to
result in "no convergence". Can the convergence problem at small distances
be eliminated by using cards line WF and CON?

Thank you

Sunil

On Mon, Jan 26, 2009 at 4:04 PM, Peter Knowles <KnowlesPJ at cardiff.ac.uk>wrote:

> It looks to me as if your calculations nearly converge, and would do so
> with just a few more iterations in the MCSCF equation solver, which can be
> specified using the MAXITER command - see
> http://www.molpro.net/info/current/doc/manual/node252.htmlOf course, 0.4
> Angstrom is a very short bond length giving very high ground-state energies,
> and one can expect all kinds of challenges from low-lying states that might
> affect the choice of active orbital space.
> Peter
> On 26 Jan 2009, at 05:46, Sunil wrote:
>
> Dear Sir,
>
> I am facing a convergence problem when using Molpro to calculate potential
> energy surfaces of CO2 molecule. The input of the program is as follows:
>
>
> *** CO2 PE curve                        ! Title: PE Curve for CO2
> memory,32,m
> maxiter=100
> geomtyp=zmat
> geometry
> ang                                     ! ang is used if we intend to
> provide distance in Angstrom; default for Z-matrix is au
> C
> O1 C rco1(i)
> O2 C rco2(i) O1 oco
> end
>
> r1(1) = 0.400000 Ang
> r2(1) = 0.400001 Ang
> dr = 0.1 Ang
> do i = 1,19
> r1(i+1) = r1(i)
> r2(i+1) = r2(i) + dr
> enddo
>
> oco = 180 Deg
>
> basis = vqz                             ! Basis set specification
>
> do i = 1,20
>
> rco1(i) = r1(i)
> rco2(i) = r2(i)
>
> hf;
> enhf(i)=energy
>
> mp2;
> enmp2(i)=energy
>
> multi;
> emcscf(i) = energy
>
> enddo
>
> table rco1,rco2,enhf,enmp2,emcscf               !emscf
> !,ecas,emrci
> head R1,R2,SCF,MP2,MCSCF
> save co2_mcscf.tab
> Title Results for CO2 (Angle = 180 Deg)
> !sort 3,1,2
> ---
>
> It produces the output but without convergence for the MCSCF calculations.
>
> Thank you in advance for any kind of help.
>
> regards
> Sunil
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
>
>
> --
> Prof. Peter J. Knowles
> School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff
> CF10 3AT, UK
> Telephone: +44 29208 79182 Fax: +44 2920874030 Email
> KnowlesPJ at Cardiff.ac.uk
> WWW
> http://www.cardiff.ac.uk/chemy/contactsandpeople/academicstaff/knowles.html
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20090127/13b17ec8/attachment.html>