Molecules moving far apart during metadynamics

[Aditya Goyal]

Dear All,

I am performing metadynamics simulation on my system having 2 molecules to perform a reaction with coordination numbers as 2 collective variables. Unfortunately, a molecule is moving far apart from the other molecule (fixed) during simulation. So, I wanted to apply a boundary to restrict the motion of moving molecule within a certain distance (d) from the fixed molecule,  to maintain enough interactions to perform this reaction. I've seen some answers suggesting restraining the distance using a collective variable but that would fix the moving molecule to distance 'd '  in space and restrict its motion for distances ≤ d .  Ideally, the moving molecule should stay within distance : (0 to d ) for a good metadynamics run. Any suggestions would be appreciated.  

Regards

Aditya

[Matt Watkins]

Hello,

sorry I don't know the metadynamics code well.

Maybe if you explain very clearly what you want to achieve and include your input / output you can get more assistance.

Matt

[Aditya Goyal]

Dear Matt , 

I have attached an output movie file with this message. In that file, first image is the equilibrated structure of both molecules when placed to closed to each other. Then, After running a meta-dynamics run for few steps the water molecule moves far apart from the parent (fixed) molecule. As i mentioned in the original message : I wanted to restrict the water molecule within a certain distance ( not fixed ) during meta-dynamics run, so as to maintain enough interactions between molecules to perform this reaction. Any suggestions would be appreciated. 

Regards

Aditya

  

[Aditya Goyal]

[Marcella Iannuzzi]

Dear Aditya

You can use either a WALL associated to some collective variable, or an external potential confining the coordinates within a certain region of space. 

Regards

Marcella

[Aditya Goyal]

Dear Marcella

Thanks for your reply. I did apply WALL  associated with collective variables (coordination numbers of C  and O in parent molecule) for my system. I applied a lower limit (found using an unbiased MD run) of coordination numbers using WALL_MINUS.

But the water molecule was still moving away inside the system during metaD. run, probably because the system could still maintain the wall limit with/without water.  Could you please suggest any different way to apply WALLS for this system? Also, you suggested using an external potential confining the coordinates within a certain region of space. Could you please explain the keywords of CP2K  to apply external potential for any system? 

Regards

Aditya