Zero adsorption when intra vdW and Coulomb are on

Started by ae213, December 10, 2020, 07:12:27 PM

Previous topic - Next topic

ae213

Dear Prof. Dubbeldam and RASPA community,

I am trying to reproduce adsorption of sarin on IRMOF-6 (http://aam.ceb.cam.ac.uk/mof-explorer/CWACapture/).
I use GCMC ensemble and the molecule is flexible. The MOF charges are taken from CoRE MOF DDEC database.
I was able to get a reasonable value for adsorption when intra vdW and Coulomb interactions are turned off.
However, I believe it should be taken into account for a correct model. When I turn it on (same case for 1-4 and 1-5 or only 1-5 neighbors), the resulted adsorption is zero.

I did the following to make sure only this factor gives such a result:
- I tested the model of the molecule with bulk NPT (with intra interactions)  and the result was reasonable and there was nothing strange with the system
- On the same model of IRMOF-6 I exactly reproduced adsorption of CO2 from literature
- Sarin with intra vdW and Coulomb did not give a zero adsorption on a zeolite (NaX)

Do you know what might cause the problem and is there a way to resolve it?

Thank you very much,
Alina


David Dubbeldam

Did you set the 'IdealGasRosenbluthWeight'?

Since every flexible model changes the reference chemical potential, the IdealGasRosenbluthWeight needs to be explicitly computed using a simulation of a single molecule in an empty box, and then used as the reference in the adsorption simulation.

ae213

Dear Prof. Dubbeldam,

Thank you very much for your reply.
I computed and specified the IdealGasRosenbluthWeight but it did not give any improvement. I also noticed that, when intra vdW is on and intra Coulomb is off, adsorption is not zero, but still much lower (and close to zero) than the case with internal non-bonded interactions turned off (and much lower than what is given here http://aam.ceb.cam.ac.uk/mof-explorer/CWACapture/).
For example, for IRMOF-6, I was only able to get adsorption close to what is given in the data base with turned off internal non-bonded interactions.
Interestingly, it happens for all the MOFs from RASPA library (I used GenericMOFs force field for all of them) and does not happen for any of the zeolites.


David Dubbeldam

Note that force fields like 'GenericMOFs' are only examples (just like any molecule or structure file), just for the example cases in the code to work. For any real work you need to create and develop your own force fields, or take them from from the literature, and carefully construct, check and validate them.