VMD-L Mailing List

From: Vardan Vardanyan (vv25_at_rice.edu)
Date: Thu Aug 15 2024 - 11:11:19 CDT

Dear Comunity

I figured out a few things regarding the question I have asked.
So, my first question needed to be corrected. The orca/vmd uses the
coordinates from geometry optimization at level MP2.
The orca file has two jobs. The first job is to calculate and print Hessian
in redundant internal coordinates. The second job is importing hessian
redundant internal coordinates and doing the same job.

In contrast, Gaussian can print the job script in redundant internal
coordinates; thus, there is no need for a second job.

The error only occurs during the second job.
Regardless, the orca output can be imported into VMD and perform
optimization (and I still don't know why).

I compared optimized results (for the molecule from the tutorial) in the
case of ORCA and Gaussian, and they are the same (they are not exactly the
same, but they can't be exactly the same).

In conclusion,

1. During bond and angle optimization, fftk uses the same idea for Gaussian
and ORCA.
2. ORCA gives an error at the end. However, the output file can be uploaded
into fftk for the next step to optimize the bond angle parameters.

Best,
VV.

On Thu, Jul 25, 2024 at 1:05 PM Joel Subach <mjsubach_at_alumni.ncsu.edu>
wrote:

> Hi VV the ORCA Forum would probably function for your 2nd part:
>
> *
> https://urldefense.com/v3/__https://orcaforum.kofo.mpg.de/viewforum.php?f=8&sid=b1dd01cddbe8cb6d9f821a49b30ac248__;!!DZ3fjg!-nJQm-FXAUDb-znAXTbY6gyb7NVaWbdX9DptTqN3TLI6AGtVltFs4OfAUsQxDTVbeCfxggWk2b9XqVxbYTekGFruqg
> $
> <
> https://urldefense.com/v3/__https://orcaforum.kofo.mpg.de/viewforum.php?f=8&sid=b1dd01cddbe8cb6d9f821a49b30ac248__;!!DZ3fjg!-nJQm-FXAUDb-znAXTbY6gyb7NVaWbdX9DptTqN3TLI6AGtVltFs4OfAUsQxDTVbeCfxggWk2b9XqVxbYTekGFruqg$
> >*
>
> Try `Common Discussions´ Heading and feel free to contact me since I am on
> the same steps as you i.e. ffTK Charge Optimization ORCA step.
>
> Best,
> Joel
>
> On Thu, Jul 25, 2024 at 7:48 PM Gumbart, JC <gumbart_at_physics.gatech.edu>
> wrote:
>
>> This is a good question. I wasn’t involved in the addition of support
>> for ORCA, so I can’t comment on the logic behind it. I suspect it would be
>> fine to remove the opt keyword and instead just make sure to use the
>> optimized geometry as input.
>>
>> Best,
>> JC
>>
>> On Jul 23, 2024, at 4:29 PM, Vardan Vardanyan <vv25_at_rice.edu> wrote:
>>
>> You don't often get email from vv25_at_rice.edu. Learn why this is important
>> <https://urldefense.com/v3/__https://aka.ms/LearnAboutSenderIdentification__;!!DZ3fjg!-FxcmVd8JkY9-AtmNw7T10un7oHth4sz-kuRDMAxv2byM6o1UBQdQwC4QwLFMPn56cvDzA4cPhaFQcE5s2O8AMim2Ws$>
>> Dear Comunity
>>
>> I am using fftk to define the parameters of my ligands (they are big
>> cyanin dyes, and I am using the framgmetion technique).
>>
>> For the quantum calculations, I chose ORCA. In step 4, for bond and angle
>> optimization, where I have to reconstruct the potential energy surface, the
>> fftk writes hess.inp file.
>> It gives the error:
>>
>>
>> ----------------------------------------------------------------------------
>> ERROR !!!
>> The optimization did not converge but reached the maximum
>> number of optimization cycles.
>> As a subsequent Frequencies calculation has been requested
>> ORCA will abort at this point of the run.
>> Please restart the calculation with the lowest energy geometry
>> and/or
>> a larger maxiter for the geometry optimization.
>>
>> ----------------------------------------------------------------------------
>>
>> I changed the theory level and basis set, but nothing helped. I also
>> increased the maxiter for the geometry optimization.
>>
>> So I have two questions:
>>
>> First, why do we need geometry optimization at this stage if we have
>> already done it in the beginning (according to the tutorial)? Why can't we
>> import coordinates from the initial structure obtained by geometry
>> optimization ( at the MP2 level)?
>>
>> Second, how to resolve the error from Hess.inp.
>>
>> Thanks,
>> VV.
>>
>>
>>