Molcas Forum
Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
- Topics: Active | Unanswered
Announcement
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".#1 2022-08-04 21:29:57
- DAdrion
- Member
- Registered: 2022-08-04
- Posts: 4
[SOLVED] Error in non-adiabatic coupling calculation
Hello,
I am running some NAMD calculations and trying to compute NACs at every step. I am however getting stability issues with all of the dynamics trajectories that I am trying to run. Through the first few steps of the simulations (femtosecond time scale), all of the bonds in the molecule break apart. I have attached my input here (from the RASSCF section onwards), I'm not sure if running the NAC calculation at every steps significantly affects the stability, as I haven't done it in the past. Let me know if anything is unclear or if there are any additional questions. Thanks so much in advance!
&RASSCF
fileorb=$MOLCAS_PROJECT.StrOrb
Spin=1
Nactel=10 0 0
Charge=0
Ras1=0
Ras2=9
Ras3=0
ITERATIONS=200,100
CIRoot=4 4 1
MDRLXR=2
&ALASKA
NAC = 1 2
&Surfacehop
tully
decoherence = 0.1
&Dynamix
velver
dt = 20
velo = 1
ther= 2
temp = 300
>> End do- Dan
Offline
#2 2022-08-05 07:59:22
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,085
Re: [SOLVED] Error in non-adiabatic coupling calculation
I believe SURFACEHOP needs a gradient calculation (not a NAC) before, it's probably just taking the NAC as if it was a gradient, and hell breaks loose. Besides, I don't think it will use the NAC at all to compute the transition probabilities (it uses overlaps or CI vectors), so the computed NAC would only be for your information. Try adding another &ALASKA just before &SURFACEHOP.
Offline
#3 2022-08-08 18:43:38
- DAdrion
- Member
- Registered: 2022-08-04
- Posts: 4
Re: [SOLVED] Error in non-adiabatic coupling calculation
Thanks so much for the help! It looks like that worked and the job ran properly. I just wanted to post a screenshot here of the working input file that I used. It seems like the order in which NAC and gradients are calculated seems to also matter. If gradients are done first and then NACs, then the surface hopping module stills seems to think that the NACs are the gradients, and there are issues with the structure. Because of this, NACs have to be run first, and then gradients. An example is given below.
&ALASKA
NAC = 1 2
&ALASKA
ROOT=1
&ALASKA
ROOT=2
&Surfacehop
tully
decoherence = 0.1Offline