Hi all,
While building an RMG model of PFOA + Air (on my RMG-database branch that includes PFAS chemistry that I'm currently developing), I noticed my model is highly sensitive to the pressure dependent reaction O(25) + CO2(6) (+ M) <=> [O][C]1OO1(1495) (+ M):
This reaction comes out of pressure dependent network where the first reaction of the network is
reaction(
label = 'reaction1',
reactants = ['O(25)', 'CO2(6)'],
products = ['[O]C([O])=O(839)'],
transitionState = 'TS1',
kinetics = Arrhenius(A=(4.24329e-06,'m^3/(mol*s)'), n=3.39993, Ea=(99.783,'kJ/mol'), T0=(1,'K'), Tmin=(300,'K'), Tmax=(2000,'K'), comment="""Estimated from node Root_N-3R-inRing_N-3R->C_N-1R!H->N in family R_Addition_MultipleBond.
Ea raised from 97.6 to 99.8 kJ/mol to match endothermicity of reaction."""),
)
This reactant channel is coming from R_Addition_MultipleBond (a family I recently retrained to include some PFAS degradation training reactions). If anyone is curious, my group definition at this node looks like this:
Looks like a lot of the kinetic libraries in RMG include atomic oxygen reacting with CO2, but only report products as O2 + CO. The 2025 ThInK 1.0 mech (for modeling C1 - C3 small molecule combustion chemistry) also doesn't include O + CO2 <=> CO3, so I'm guessing this reaction shouldn't be making its way into the core.
Even on the main RMG-database branch, R_Addition_MultipleBond will still propose this reaction. I've compared the log(k)'s of this reaction using main's database (on commit from July 2025, e883a4c) and my database with PFAS chemistry.
-
Is this reaction causing issues in anyone else's models? Or even included? If I look back at a CH3F model (from probably 4-5 years ago, generated by David Farina), the model includes pressure dependent reactions of CO2 + O <=> CO3, so I think it is making its way into other people's models besides myself.
-
Should we be forbidding the reaction of O + CO2 in R_Addition_MultipleBond? Looks like the authors of many kinetic libraries have considered the reaction of O + CO2, so maybe they're purposefully excluding CO2 + O <=> CO3 because it's insignificant?
Any help/advice is appreciated :) Thanks!
Hi all,
While building an RMG model of PFOA + Air (on my RMG-database branch that includes PFAS chemistry that I'm currently developing), I noticed my model is highly sensitive to the pressure dependent reaction
O(25) + CO2(6) (+ M) <=> [O][C]1OO1(1495) (+ M):This reaction comes out of pressure dependent network where the first reaction of the network is
This reactant channel is coming from R_Addition_MultipleBond (a family I recently retrained to include some PFAS degradation training reactions). If anyone is curious, my group definition at this node looks like this:
Looks like a lot of the kinetic libraries in RMG include atomic oxygen reacting with CO2, but only report products as O2 + CO. The 2025 ThInK 1.0 mech (for modeling C1 - C3 small molecule combustion chemistry) also doesn't include O + CO2 <=> CO3, so I'm guessing this reaction shouldn't be making its way into the core.
Even on the main RMG-database branch, R_Addition_MultipleBond will still propose this reaction. I've compared the log(k)'s of this reaction using main's database (on commit from July 2025, e883a4c) and my database with PFAS chemistry.
Is this reaction causing issues in anyone else's models? Or even included? If I look back at a CH3F model (from probably 4-5 years ago, generated by David Farina), the model includes pressure dependent reactions of CO2 + O <=> CO3, so I think it is making its way into other people's models besides myself.
Should we be forbidding the reaction of O + CO2 in R_Addition_MultipleBond? Looks like the authors of many kinetic libraries have considered the reaction of O + CO2, so maybe they're purposefully excluding CO2 + O <=> CO3 because it's insignificant?
Any help/advice is appreciated :) Thanks!