mcscf geom. optimization

Maciej Bobrowski mate at chemik.chem.univ.gda.pl
Tue Nov 27 11:07:20 GMT 2001




Hi,


I have the following problem:
I want to optimize the geometry at the mcscf level. the fragment of my
input file is as follows:

........
spin=1;
IMPORT,2100.2,/homec/mate/molpro/rohf_test.wfu
basis=6-31+G*
multi;
MAXITER,40
core,10; occ,18;  closed,8;
CONFIG,DET;
wf,33,1,1;
start,2100.2;
rotate,14.1,10.1
rotate,13.1,9.1
rotate,12.1,8.1
ITERATIONS;
DO,NEWTON,4,TO,40
DONT,WERNER,4,TO,40
END;

GRADTYP=CADPAC
optg;
POP; INDIVIDUAL;
PUT,MOLDEN,mcscf_test.molden;

and after the mcscf iterations (at the beginning of optimization program):
SETTING GRADTYP        =    CADPAC
STARTING GEOMETRY OPTIMIZATION FOR MULTI
 CONVERGENCE THRESHOLDS:  .300D-03 (STEP)  .300D-03 (GRADIENT)  .100D-05
(ENERGY)
 MAX. NUMBER OF STEPS:          50
 *** Long output written to logfile /homec/mate/molpro/mcscf_test.log ***

 CORE ORBITAL RECORD NOT SPECIFIED
 ...and the job was finished :(

How can I specified the "CORE ORBITAL RECORD" ??

and by the way, I wanted to change the ordering of some orbitals and
used the "rotate" command, but in the output file I obtained something
like that:

 ROTATION OF ORBITALS  14.1   10.1       90.00 DEGREES
 ROTATION OF ORBITALS  13.1    9.1       90.00 DEGREES
 ROTATION OF ORBITALS  12.1    8.1       90.00 DEGREES

I didn't want to rotate them in 90 deg. I just wanted to change the
ordering of some of them. What's the problem, what is wrong in my input
file??

Maciej

   ......................................................................
    Maciej Bobrowski

          Faculty of Chemistry        Academic Computer Center
          University of Gdansk        TASK
                Sobieskiego 18        Narutowicza 11/12
         80-952 Gdansk, Poland        80-952 Gdansk, Poland
        phone: +48 58 345-0361        phone: +48 58 347-2411
         fax: +48 58 341-0357         fax: +48 58 341-5821
   e-mail: mate at chemik.chem.univ.gda.pl   e-mail: mate at kliper.task.gda.pl
   ......................................................................




More information about the Molpro-user mailing list