Molcas Forum

dkaliakin

Member

From: Northeastern University Boston

Registered: 2020-09-03

Posts: 7

QM/MM Geometry Optimization

Dear Molcas Experts,

I am currently trying to perform QM/MM geometry optimization using Molcas_19.11 / Tinker_6.3.3 interface, but I am having some difficulties with set up of input file. I am using example input file from Molcas manual as the test

https://www.molcas.org/documentation/manual/node79.html

My *.key and *.xyz files are identical to this example file and I only modify *.inp to change the run type from single point energy calculation to geometry optimization

My input file is

>  COPY  $PATH/$Project.xyz  $WorkDir/$Project.xyz
>  COPY  $PATH/$Project.key  $WorkDir/$Project.key
>  EXPORT TINKER=$PATH/Molcas/tinker/bin
>> EXPORT MOLCAS_MAXITER=10000

&Gateway
 Tinker
 Basis  =  STO-3G
 Group  =  Nosym

>>> Do while

&Seward

&Espf
 External  =  Tinker
 LAMorok

&SLAPAF
 cartesian
 rHidden = 4.0

>>> EndDo

The error message that I get is

     Gradient is translational variant!
     Gradient is rotational variant!
 ###############################################################################
 ###############################################################################
 ###                                                                         ###
 ###                                                                         ###
 ###    Location: get_dArray                                                 ###
 ###                                                                         ###
 ###                                                                         ###
 ###    Data not defined: Dipole moment                                      ###
 ###                                                                         ###
 ###                                                                         ###
 ###############################################################################
 ###############################################################################
--- Stop Module: slapaf at Thu Sep  3 17:42:19 2020 /rc=_RC_INTERNAL_ERROR_ ---
*** files: xmldump
    saved to directory /scratch/d.kaliakin/13723471

>>> END DO

.########################################################.
.# Some internal inconsistency of the code was detected #.
.########################################################.

Thank you very much for your help in advance.

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,012

Re: QM/MM Geometry Optimization

You are not computing any energy. For example, you could include before &SLAPAF:

&SCF
  Charge = -1

dkaliakin

Member

From: Northeastern University Boston

Registered: 2020-09-03

Posts: 7

Re: QM/MM Geometry Optimization

Thank you very much for your help! I really appreciate that.

Could you please help me with quick follow up question? I was wondering if there is a (proper) way to visualize the geometrical change (or at least the final geometry) of MM part during the QM/MM geometry optimization? From what I can see both *.geo.molden and *.Opt.xyz contain only QM geometry. Do I need to specify additional environmental variable in *key or *inp in order to be able to obtain the coordinates of optimized MM part? Ideally I would like to have the geometry for the combined QM/MM system.

I can achieve that (visualization of full QM/MM system) by simply increasing the number of MM atoms specified in *key, but I want to make sure that this does not lead to any potential issues in more complex production simulations.

Thank you very much for your help again.

niko

Member

From: Marseille

Registered: 2015-11-08

Posts: 59

Website

Re: QM/MM Geometry Optimization

Hi,

Only Tinker is aware of the full system, hence you need to rely on Tinker tools to achieve what you want. If you still has access to the temporary directory, you will find numbered xyz files that you can easily visualize, using eg Molden. Alternatively, you can use Tinker's utility archive to concatenate these files into a single "movie".

Hope this helps.

dkaliakin

Member

From: Northeastern University Boston

Registered: 2020-09-03

Posts: 7

Re: QM/MM Geometry Optimization

Great! Thank you very much! This was really helpful and I really appreciate that!