[molpro-user] Convergence problem in CPHF

[Moilanen Jani]

Hello

I try to optimize the geometry of the copper complex of selenium diimide at the QCISD/def2-TZVP (def2-TZVPP for Cu) and QCISD(T)/def2-TZVP (def2-TZVPP for Cu) levels of theory. If I use the largest plausible noble gas core in calculations, they convergence properly. However, if I try to include 3s and 3p electrons of copper atoms into to the correlation space, i.e. I use slightly smaller frozen core, the CPHF does not convergence anymore (input below). Does anyone know where this problem originates from? To my best knowledge, this problem should not arise from the Def2-TZVPP basis set since it is all electron basis set in case of copper.

Best wishes,

Jani Moilanen

ipnut:
***,qsicd,def2-TZVP
memory,465,m
gprint,orbitals
geomtyp=xyz
geometry={
 N          2.3293322982       -0.3193382070        0.0000000000
 Se         2.2251791788       -2.0236218093        0.0000000000
 N          0.5383325792       -2.2881873050        0.0000000000
 Cu        -0.8987265307       -0.9879674984        0.0000000000
 N         -2.3293322982        0.3193382070        0.0000000000
 Se        -2.2251791788        2.0236218093        0.0000000000
 N         -0.5383325792        2.2881873050        0.0000000000
 Cu         0.8987265307        0.9879674984        0.0000000000
 H         -0.3497229432        3.3009231890        0.0000000000
 H          3.3197150569       -0.0358742249        0.0000000000
 H          0.3497229432       -3.3009231890        0.0000000000
 H         -3.3197150569        0.0358742249        0.0000000000

}

basis={

default=def2-TZVP,Cu=def2-TZVPP

}

{hf;wf,156,1,0;orbprint,2}

{qcisd,maxit=80;core,13,3,13,3;natorb,5000.2,10}

optg

{put,molden,sc_qcisd.molden;orbital,5000.2}

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20130902/606d8905/attachment.html>