search

peteboyd / lammps_interface

automatic generation of LAMMPS input files for molecular dynamics simulations of MOFs
MIT License
125 stars 63 forks source link

Unable to generate lammps file for UiO-66 with BTW_FF, Dubbeldam and Dreiding #13

Open ml4132 opened 5 years ago

ml4132 commented 5 years ago

Dear Pete,

First of all I would like to thank you for sharing your code with the scientific community.

I recently started testing it on a structure of UiO-66 I optimized in VASP. Things went smoothly with UFF and UFF4MOF and I obtained correct lamps input files. I then tested the same structure for the Dubbeldam, Dreiding, and BTW_FF force fields using the same command line (e.g., python lammps_interface.py --force_field Dubbeldam UiO-66.cif) but I experienced some issues.

In the case of Dreiding I get an error message that the force field for atom 433 with element 'Zr' is missing and for Dubbeldam and BTW_FF the script just prints "Detecting inorganic clusters" and then hangs forever never printing more output.

I would appreciate any help on how to solve this issue and would be happy to provide you with my cif file and any additional information you might find helpful.

Thanks in advance.

Kind Regards, Martina

-- Martina Lessio, Ph.D. University of Sydney Fellow, Postdoctoral Research Associate School of Chemistry University of Sydney E: martina.lessio@sydney.edu.au

ltalirz commented 4 years ago

@peteboyd Do you have any idea what the issue may be and whether it still exists?

peteboyd commented 4 years ago

Hi Dr. Lessio,

Your first issue regarding Dreiding and missing 'Zr' is due to the fact that the Dreiding force field does not have parameters for Zr atoms. There is not much we can do about that, and you will have to forget any analysis you were planning to do with UiO-66 and Dreiding.

The 'Detecting inorganic clusters' issue is a result of an extremely inefficient code that tries to find the Zr4O cluster with 3D pattern recognition. There are better ways to discover chemical patterns in molecules, something I've wanted to implement if I had the time. However if you used a unit cell of UiO-66 and were extremely patient, the code should provide you with LAMMPS input files.

I will leave this issue open for now, to serve as a reminder that the code should be improved.

Best, Pete

ml4132 commented 4 years ago

Dear Pete,

Thank you very much for your response. That makes perfect sense. I did figure out the issue with Dreiding myself but this was after posting. Apologies for that, I should have looked into that earlier.

Thanks again for sharing the code, Martina