Halogens kinetics

[davidfarinajr]

This PR adds halogens to the kinetics database
Note: the database tests will not pass until ReactionMechanismGenerator/RMG-Py#1997 is merged since the PR depends on the R!H!Val7 atomtype

FAMILIES

New Families - F_Abstraction, Cl_Abstraction, Br_Abstraction, Disproportionation-Y, XY_Addition_MulitpleBond, 1,2_XY_interchange

The families are put in a new halogens set in recommended.py acb9b8f

Existing Families -

The strategy for adding halogens to existing families was primarily to add the Val7 atomtype to H atom nodes

H_Abstraction - new groups are machine written, *3 halogen atom abstractors, added training reactions
Disproportionation -new groups are machine written, *1 halogen atom abstractors, added training reactions
R_Addition_MulitpleBond - new groups are machine written, *3 halogen atoms can add, added training reactions
1,2_Insertion_carbene - added halogens and training reactions
1+2_Cycloaddition - H -> [H,Val7] for unlabeled atoms
1,2-Birad_to_alkene - H -> [H,Val7] for unlabeled atoms
1,2_Insertion_CO - *3 H -> [H,Val7]
1,3_Insertion_CO2 - *3 halogens
1,3_Insertion_ROR - - *3 H -> [H,Val7]
Birad_R_Recombination - machine written, *1 [H,Val7]
Intra_H_migration - *3 H -> [H,Val7] (halogens can migrate). Plan to make new families (intra_X_migration) when we have more kinetic data
Singlet_Carbene_Intra_Dispropotionation -*3 H -> *3 [H,Val7]
CO_Dispropotionation - machine written H -> [H,Val7]
R_Recombination - new halogen groups added, new training reactions, this family should be reconstructed with automated tree gen
Intra_Disproportionation - *4 H -> [H,Val7]
Intra_ene_reaction - *6 H -> *6 [H,Val7]
Concerted Intra Diels alder monocyclic 1,2 shiftH - *7 H -> [H,Val7]
2+2_cycloaddition_CCO - H -> [H,Val7]
2+2_cycloaddition_Cd - H -> [H,Val7]
2+2_cycloaddition_CO - H -> [H,Val7]
Intra_RH_Add_Endocyclic - H -> [H,Val7]
Intra_RH_Add_Exocyclic - H -> [H,Val7]
ketoenol groups - H -> [H,Val7]
R_Addition_COm - H -> [H,Val7]

LIBRARIES

• HFC mech (NIST)

• CH2F2 (NIST)

• YF

• 2-BTP/CF3Br (NIST)

• DTU mech CH3Cl

[rwest]

is this change from kekulized to aromatic intentional and beneficial?

[davidfarinajr]

Now simultaneous with PR ReactionMechanismGenerator/RMG-Py#2135 because we need to swap labels for new own reverse families

[davidfarinajr]

======================================================================
1745
FAIL: test that the right number of reactions are in output network
1746
----------------------------------------------------------------------
1747
Traceback (most recent call last):
1748
  File "/home/runner/work/RMG-database/RMG-Py/arkane/explorerTest.py", line 81, in test_reactions
1749
    self.assertEqual(len(self.explorer_job.networks[0].path_reactions), 7)
1750
AssertionError: 6 != 7
1751

Lol, we just changed this from 6 to 7 when CO was added to primaryThermoLibrary 😂. I add a bunch of training reactions in this PR, so I must have altered the kinetics so we get 6 now instead of 7.

[davidfarinajr]

Non-identical kinetics!
3077original:
3078rxn: H(4) + CH2CHCHCH(15) <=> butadiene13(17)		origin: R_Recombination
3079tested:
3080rxn: butadiene13(17) <=> H(4) + CH2CHCHCH(15)		origin: R_Recombination
3081k(1bar)|300K   |400K   |500K   |600K   |800K   |1000K  |1500K  |2000K  
3082
3083k(T):  |   9.21|   9.03|   8.88|   8.76|   8.58|   8.43|   8.17|   7.98
3084k(T):  | -66.67| -45.24| -32.57| -24.25| -14.07|  -8.15|  -0.66|   2.76
3085
3086Kinetics: Arrhenius(A=(8.15666e+18,'cm^3/(mol*s)'), n=-1.493, Ea=(0,'kcal/mol'), T0=(1,'K'), comment="""BM rule fitted to 2 training reactions at node Root_1R->H_N-2R->S_N-2CHNO->H_N-2CNO-inRing_Ext-2CNO-R_Sp-3R!H=2CCNNOO_N-3R!H->O Total Standard Deviation in ln(k): 7.13613102162
3087Kinetics: Arrhenius(A=(5.3e+44,'s^-1'), n=-8.62, Ea=(123.608,'kcal/mol'), T0=(1,'K'), comment="""Matched reaction 188 C4H6-8 <=> C4H5-3 + H in R_Recombination/training
3088kinetics: BM rule fitted to 2 training reactions at node Root_1R->H_N-2R->S_N-2CHNO->H_N-2CNO-inRing_Ext-2CNO-R_Sp-3R!H=2CCNNOO_N-3R!H->O Total Standard Deviation in ln(k): 7.13613102162
3089kinetics: Matched reaction 188 C4H6-8 <=> C4H5-3 + H in R_Recombination/training

looks like the kinetics are being compared in opposite directions 🤔

[davidfarinajr]

Change in observables:

Observables Test Case: Aromatics Comparison
4069================
4070
4071All Observables varied by less than 0.500 on average between old model and new model in all conditions!

Observables Test Case: NC Comparison
5246================
5247
5248All Observables varied by less than 0.200 on average between old model and new model in all conditions!

Observables Test Case: SO2 Comparison
5572================
5573
5574The following observables did not match:
5575
5576Observable species O=S=O varied by more than 0.100 on average between old model SO2(15) and new model SO2(15) in condition 1.
5577
5578

Observables Test Case: EG1 Comparison
3529================
3530
3531All Observables varied by less than 0.050 on average between old model and new model in all conditions!
3532
3533

Observables Test Case: Oxidation Comparison
12394================
12395
12396The following observables did not match:
12397
12398Observable species [OH] varied by more than 0.500 on average between old model OH(5) and new model OH(5) in condition 1.
12399
12400

Observables Test Case: liquid_oxidation Comparison
16195================
16196
16197All Observables varied by less than 0.100 on average between old model and new model in all conditions!