<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hey,<br><br>I indeed miss the point you mention in the documentation concerning ALASKA and state averaged MCSCF. I will so run my gradient calculation using rs2 code and try CADPAC gradient on a segmented basis to monitor the difference. <br><br> Thank you very much your advices.<br><br>Best Regards<br><br>Benjamin<br><br><div>> Date: Fri, 6 Jun 2014 21:30:31 +0200<br>> From: michael.banck@ch.tum.de<br>> To: molpro-user@molpro.net<br>> Subject: Re: [molpro-user] SA-CASSCF optimisation with ALASKA<br>> <br>> Hi,<br>> <br>> On Fri, Jun 06, 2014 at 03:04:46PM +0200, benjamin gonon wrote:<br>> > I'm actually encountering a problem in performing optimization with<br>> > state specific gradient including cpmcscf. If i well understand Molpro<br>> > Manual, this need to include a CPMCSCF card in MCSCF group and use<br>> > ALASKA gradient code (supposed to be the default) I think i did this<br>> > well:<br>> <br>> [...]<br>> <br>> The 2012.1 manual says this in section 44.1: "The CADPAC gradient<br>> program [...] works for [...] (state averaged) MCSCF" and further: "The<br>> ALASKA gradient program [...] allows the calculation of gradients of<br>> generally contracted basis functions for [...] MCSCF.[...] Gradients for<br>> state averaged MCSCF wave functions can be evaluated using the RS2<br>> gradient program".<br>> <br>> So ALASKA does *not* work for state-averaged MCSCF wave functions via<br>> the MULTI program.<br>> <br>> > Unfortunately this doesn't seem to work and get me this too relevant<br>> > error lines in output:<br>> > VARIABLE ALASKA UNDEFINED, ASSUMING 0<br>> > GENERAL BASIS CONTRACTION NOT IMPLEMENTED IN CADPAC GRADIENTS<br>> <br>> As mentioned above, CADPAC does not allow basis contraction.<br>> <br>> > I know the same gradient can be computed using the rs2 program and<br>> > that's probably what i will do in the end. But i would prefer to use<br>> > this straight method than use rs2 code, as i will need to compute<br>> > non-adiabatic coupling and optimize conical intersection.<br>> <br>> What I have done in the past when RS2 was not an option was to use the<br>> respective uncontracted basis set, by specifying the basis via the<br>> primitive set definition (see section 11.8 in the 2012.1 manual).<br>> <br>> I believe in your case it would look like this:<br>> <br>> basis={<br>> sp,h,vdz<br>> spd,c,vtz<br>> spd,o,vtz}<br>> <br>> Using this as basis, the CADPAC gradient is getting evaluated.<br>> <br>> Be aware that (i) I am not really sure this is a sound approach, (ii)<br>> the calculation will take longer and (iii) the energies will differ<br>> slightly.<br>> <br>> <br>> Best regards,<br>> <br>> Michael<br>> <br>> _______________________________________________<br>> Molpro-user mailing list<br>> Molpro-user@molpro.net<br>> http://www.molpro.net/mailman/listinfo/molpro-user<br></div> </div></body>
</html>