Author Topic: Tremolox using multiple tags for the same particle  (Read 1379 times)

0 Members and 1 Guest are viewing this topic.

Offline matsiv

  • Regular QuantumATK user
  • **
  • Posts: 8
  • Country: it
  • Reputation: 0
    • View Profile
I'm noticing an issue I have with the tremolox potential builder: apparently I cannot have a particle with more than one tag. If I do I get the errror "the potential set is missing support for particle "x" with tags ['y','z'].
Now, why would there be such a problem?
Basically I have a ferrocene molecule. I used a unique tag for all the carbon atoms involved in the two ciclopentadiene rings. All the ring atoms are bonded to a dummy hydrogen atom in the center and to the adjacent carbon atom
However then I need to set the dihedral to constrain the rotation of the rings parallel to each other. To do that I need to select two specific atoms of the ring and fix the dihedral [carbon,dummy atom, dummy atom, carbon].
However the only way I see to select a specific atom for a force field term is to assign a separate tag to it and reference the (atom type, tag) combination. If an atom can only have one tag the process of setting all interactions I need becomes extremely tedious because then I have to respecify separately all force field terms involving the new atom so I'd assume I'm missing on something

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5394
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: Tremolox using multiple tags for the same particle
« Reply #1 on: July 29, 2021, 08:55 »
This is not my area of expertise and it's quite complex, but since a lot of our support staff is on summer vacation I figured I at least answer what I can.

I think you are both right and wrong :-) You can certainly have more than 2 tags on atoms, in fact it's often necessary for polymers. The error message is just written like this to be as informative as possible. However, a particular atom cannot belong to more than one potential set, I believe. Otherwise we don't know which one to use to compute the forces on it...

Can't the potential terms be combined to a single parameter set? Maybe if the dummy atom is not H but, say He?