ReactionMechanismGenerator / ARC

ARC - Automatic Rate Calculator
https://reactionmechanismgenerator.github.io/ARC/index.html
MIT License
42 stars 21 forks source link

Hindered rotors can not be identified correctly #482

Open wangdu817 opened 2 years ago

wangdu817 commented 2 years ago

Describe what you are trying to achieve A clear and concise description of what you want to achieve. I tried to run calculate rate constants using ARC and included the rotors job. Even though the rotors scan job was finished and relevant information has been written to the restart files, but the 'SUCCESS' state is null, so that rotors information did not correctly write to the Arkane input files, and the rate constants and thermal properties were calculated without hindered rotors considered. It seems automatic troubleshooting for rotor scan also did not run. Describe methods you have considered I try to manually replace 'null' with 'True', and all the rotors are writing to the Arkane input files, but I'm not sure are there any problems. By the way, is this project still maintained and will be updated?

alongd commented 2 years ago

In case there are problems with the 1D scans, ARC prefers not to use them at all than using "bad" results. You are right, in this case (if all rotors fail) ARC is only using the RRHO model. If you add more information (e.g., ARC's log file) we may be able to assist a bit more regarding the specific system. Also, you can try visualizing the actual 1D scans using, e.g., GaussView to assess their quality

wangdu817 commented 2 years ago

Thank you for your reply, I have checked the scan and found that the scan broke the Transition state, making the peak and valley not equal. And the energy is not equal after the 360-degree scan. I tried to use the preserved_parameters_in_scan to keep the structures, but it did not work, because the Gaussian input file did not add any keyword to freeze the atoms.

In case there are problems with the 1D scans, ARC prefers not to use them at all than using "bad" results. You are right, in this case (if all rotors fail) ARC is only using the RRHO model. If you add more information (e.g., ARC's log file) we may be able to assist a bit more regarding the specific system. Also, you can try visualizing the actual 1D scans using, e.g., GaussView to assess their quality