tce/qk style about sparta

[stanmoore1]

@gokulbr4 do you have a minimal working example that reproduces the issue that we can run? Sorry for the basic question, but are you sure that your simulation conditions are such that would produce reactions? E.g. temperature, density, etc.

[knstmrd]

@gokulbr4 I took the input files from the email you sent to the mailing list, could reproduce the issue with no reactions when using the "tce/qk" hybrid style (switching to either "pure" tce or qk leads to reactions taking place).

There is indeed a bug in the react_tce_qk.cpp file: in line 126, a check is performed if (e_excess > 0.0) return 0; but it is unphysical (and the correct condition that e_excess >= 0.0 is already being checked for in the base attempt function).

There also seem to be an inconsistency in the code:
the base function attempt checks that the total energy of the colliding partners is larger than the threshold energy of the reaction. But the QK and TCE styles themselves check for translational+fraction of rotational energy > threshold energy. So the base check (from attempt) can't be re-used, one has to check twice. Related to #279

I am also not 100% sure about the implementation of the reaction loop: in the "pure" TCE/QK styles, a random number is sampled once, and the reactions are looped over, each one adding to the reaction probability CDF. In the hybrid style, each reaction is checked against a fresh random number – not fully sure if this is consistent with the basic TCE/QK styles? Not entirely sure how the react command works together with the collide_vss style, so am definitely not the best judge on this.

[gokulbr4]

Thankyou so much for sorting it out and it is working.