# [molpro-user] CASPT2 optg : zmatrix.f:zmat_orient problem

Werner Győrffy gyorffy at theochem.uni-stuttgart.de
Mon Sep 22 15:21:38 BST 2014

```Dear Dani,

Your system has 6 symmetry unique internal coordinates (in C2h
symmetry). The F-P-P-F dihedral is 180 degree which you can keep fixed.
As soon as you do so, your system stays in C2h symmetry during the
entire geometry optimization with numerical differentiation. I have
cleaned up your geometry input as follows

geometry={
p
f    1 fp
p    1 pp    2 ppf
f    3 fp    1 ppf   2  180.0
h    1 hp    3 hpp   4  dih
h    3 hp    1 hpp   2  dih
h    1 hp    3 hpp   4 -dih
h    3 hp    1 hpp   2 -dih
}
fp   =   1.631683
pp   =   2.461103
ppf  = 163.330
hp   =   1.407364
hpp  =  92.933
dih  = -48.581

I hope it helps.

Regards,

Werner.

On 09/18/2014 05:18 PM, Dani Setiawan wrote:
> Dear Werner,
>
> Yes I did. I put same variable for all equivalent coord related by symmetry.
>
> But this is not the case, as it works well with single point CASPT2, and the program correctly determined its symmetry. The problem comes when it enters CASPT2 optg.
>
> Best,
> Dani
>
>
> On Sep 18, 2014, at 6:00 AM, molpro-user-request at molpro.net wrote:
>
>> Today's Topics:
>>
>>    1. Re: CASPT2 optg :  zmatrix.f:zmat_orient problem (Werner Gy?rffy)
>>
>>
>> ----------------------------------------------------------------------
>>
>> Message: 1
>> Date: Wed, 17 Sep 2014 11:23:09 +0200
>> From: Werner Gy?rffy  <gyorffy at theochem.uni-stuttgart.de>
>> To: molpro-user at molpro.net
>> Subject: Re: [molpro-user] CASPT2 optg :  zmatrix.f:zmat_orient
>> 	problem
>> Message-ID: <541952FD.4070908 at theochem.uni-stuttgart.de>
>> Content-Type: text/plain; charset=UTF-8; format=flowed
>>
>> Dear Dani,
>>
>> Have you used the same variable for internal coordinates which are
>> related by symmetry? Unless you want to lower the symmetry used in the
>> calculation, you can let the program find out the symmetry of your
>> molecule.
>>
>> Regards,
>>
>> Werner Gyorffy.
>>
>> On 09/17/2014 07:22 AM, Dani Setiawan wrote:
>>> Dear molpro users,
>>>
>>> I got this error when doing geometry optimization with CASPT2:
>>> /
>>>   Starting numerical gradient for RS2C//
>
>
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user
>

```