Is there an automatic way to generate conformers for radical like systems in general (a doublet spin multiplicity)? My current system for study is a substituted fullerene with a missing H atom from the functional group.
I have tried using different conformer generators for this purpose (Baloon, RDKit, Obabel, Fullmonte, CCDC's Conformer_generator) but none of them successfully yielded out any conformers. I do not know how to run MD of such a system for which there is no general force-filed available. I have also tried to go through the AMBER tutorial but was unsuccessful: http://ambermd.org/tutorials/basic/tutorial4b/
I would really appreciate if anyone can suggest something that actually works for the system in consideration.
Thanks in advance.
Hello Viki,
Although I'm not sure about your exact requirements, you can try few structure prediction software like CALYPSO(http://www.calypso.cn/) or USPEX(http://uspex-team.org/en/). As far I remember CALYPSO can be combined with QM softwares like VASP, GAUSSIAN, SIESTA etc. You'll get the global minima (with requested spin multiplicity) along with other conformers of higher energy. Search for other similar softwares in case these does not serve your purpose.
Most methods for conformer generations use molecular mechanics methods. Unfortunately, molecular mechanics do not deal with electron at all. So it is not possible to generate radicals using such classical softwares. This also includes Molecular Dynamics simulations. What you need for radical systems is to create all possible states and run full Quantum mechanical calculations. The softwares include but not limited to Gaussian, Gamess, Orca, and Siesta.
Turbasu Sengupta: Thanks for your answer. I will take a look into the softwares you mentioned.
Abdulkadir Kocak: Thanks for your answer. I understand that MM cannot model electrons but what I do not understand is what are the methods you suggest to generate the possible states you mentioned? As far my understanding, the QM softwares just do static calculations and cannot perform the structure generation task, which is what I want an answer for. QM softwares could be coupled with MD to run ab initio MD but that would be too expensive even for semi-empirical methods. I want a software which runs on some stochastic or genetic algorithm
- Badges
- Science topic
- Similar topics
- Chemistry
- Theoretical Chemistry