Hi again,
I just wanted to confirm that constraining QM region close to the center of the box during the heating/equilibration runs was indeed a success! I was even able to set this up with a slightly smaller box size.
Thanks for your help!
Best regards,
Anthony
________________________________
From: Goetz, Andreas <awgoetz.ucsd.edu>
Sent: Tuesday, October 7, 2025 3:07 PM
To: Anthony Dominic Iii <Anthony.DominicIii.colorado.edu>; AMBER Mailing List <amber.ambermd.org>
Subject: Re: [AMBER] QM/MM Reaction help
[External email - use caution]
Hi Anthony,
Steered MD is not the issue here.
The simulation box needs to be large enough to accomodate size of QM region + 2x qmcut. In your case it should fit easily int your ~34.5 A box.
Which version of Amber are you using? Older versions of Amber had a bug in computing QM region size if adjacent molecules would be wrapped to either side of a box (e.g. one molecule at the box edge, one beyond).
You can fix that by keeping your QM region close to the box center, for instance with restraints during equilibration.
All the best,
Andy
—
Dr. Andreas W. Goetz
Associate Research Scientist
San Diego Supercomputer Center
Tel: +1-858-822-4771
Email: agoetz.sdsc.edu
Web: www.awgoetz.de
On Oct 7, 2025, at 12:55 PM, Anthony Dominic Iii via AMBER <amber.ambermd.org> wrote:
Hi everyone,
I am reaching out for help with using QM/MM to study an SN2 reaction from
https://urldefense.com/v3/__https://doi.org/10.1021/acs.jpcb.4c05926__;!!Mih3wA!HCbCkNI9VvkwBQomz3Uw55lDF-CGMYwxZ0OWxxg0QG5jM1xevovoFK_YqBGK3yjvKhZzH8a0BaQqyA$ . I follow their setup pretty closely, except that I am using Amber rather than Charmm for the MD simulations.
To summarize, my system contains NH3 and ClCH3 solvated with the TIP3P water model, and the box side lengths are ~34.5 angstroms after equilibration (this is the box side length used in the above paper). I am trying to reproduce the Menshutkin reaction from the linked paper: NH3 + ClCH3 --> Cl(-) + H3C-NH3(+).
Prior to the QM/MM attempts, I set up my simulation following the usual protocol (minimization, heating, NPT equilibration, NVT equilibration). I then use steered MD (still with classical force field) to bring NH3 to the first solvation shell of ClCH3. It is worth noting that after these steps, the system is visually well-behaved, and the density, pressure, and temperature are all converged.
To get reactivity, I have tried two approaches with QM/MM: (1) Steered MD (SMD) & (2) Umbrella sampling (US). Ideally, I would like to first do SMD to generate initial configurations for US along the reaction coordinate similar to the alanine tetrapeptide
When performing QM/MM with SMD or US, I get errors saying: " SANDER BOMB in subroutine QM_CHECK_PERIODIC<qm_mm.f> QM region + cutoff larger than box cannot continue, need larger box." In my attachment, I include the script that gave this error, where I try SMD (with a LCOD collective variable) to get the reactants closer together (LCOD goes from -3 angstroms to -2 angstroms), without reactivity using a force constant of 80 kcal/mol.
I tried the following solutions with no luck:
1.
Increasing the box size to L=40 angstroms
2.
Increasing the box size to L=45 angstroms
3.
Reducing the value of "qmcut"
4.
Reducing the value of both "qmcut" and "cut"
5.
Changing from AM1 to PM3 (AM1 is what the paper used)
Could someone please let me know if they have overcome a problem like this before? Is there anything that I am missing? Perhaps I am misunderstanding the format of the cv_LCOD.in file?
Thanks for your help in advance!
Best regards,
Anthony Dominic
<cv_LCOD.in><06_QMMM (1).in>_______________________________________________
AMBER mailing list
AMBER.ambermd.org
https://urldefense.com/v3/__http://lists.ambermd.org/mailman/listinfo/amber__;!!Mih3wA!HCbCkNI9VvkwBQomz3Uw55lDF-CGMYwxZ0OWxxg0QG5jM1xevovoFK_YqBGK3yjvKhZzH8Zrff3cmg$
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 13 2025 - 08:30:02 PDT
