Limited support for analytic Hessians in Q-Chem

Questions
#1

Hi friends,
I have a question about the limited support for analytic Hessians in Q-Chem. Why is Q-Chem so limited in this area compared to other popular programs like Gaussian and ADF?

#2

If you are interested in becoming a Q-Chem developer to help us implement more analytic Hessians, please contact Q-Chem support.

From my personal point of view, finite-difference 2nd derivatives are just fine and they eliminate a major memory bottleneck in the analytic approach.
https://doi.org/10.1002/jcc.24811

Meanwhile, Q-Chem does have some 2nd derivatives that are not available elsewhere, e.g., for VV10-containing functionals:

1 Like
#3

Hello farshadshiri,

Thank you for bringing it up. If you have an interest in analytic Hessians of some functionals, you may send a feature request to our support email. (support@q-chem.com) We are more than happy to put it on our TODO list.

Best,

Kuan-Yu

#4

Dear Kuan-Yu and John,

Thank you for your comments and help. I have sent a feature request to the Q-Chem developers for the addition of analytical Hessians to the program.

#5

Hello @farshadshiri ,

Please see Table 5.1 which provides a summary, arranged according to Jacob’s Ladder, of which categories of functionals are available with analytic first derivatives (for geometry optimizations) or second derivatives (for vibrational frequency calculations). Q-Chem contains over 150 functionals which are listed in the rest of Ch 5, if there is a specific functional that you wish to have analytical Hessian within Q-Chem, please inform support of the feature request for that specific functional.

#6

Hello Peter,
I appreciate your assistance.
You’re absolutely right; Q-Chem does provide support for analytic energy gradients and Hessians for various functionals. However, it has limitations when it comes to implicit solvent models. Currently, analytic Hessians are restricted to the C-PCM model, and it would be advantageous to extend this support to SMD, as it offers improved accuracy and generality.

#7

Analytic Hessians for SMD are under development in my group.

#8

Hello John,
I’m glad to hear that. Analytic Hessians for SMD are much needed and valuable. How long do you think it will take you to complete the development?

#9

Aiming for Q-Chem 6.2, if not that then a bug-fix release (6.2.x) later next year.