Start from seed fails under certain conditions

[sevyharris]

Bug Description

RMG can't start from a seed mechanism under certain conditions. When I include my library of DFT calculations as a library but not as part of the seed, the RMG run crashes with the following error:

Making seed mechanism...
Conducting simulation of reaction system 1...
At time 0.0000e+00 s, species [CH2]CCC(24) at 76545039.70987049 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species C[CH]CC(23) at 2872877563043.813 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species [CH2](27) at 209469.38874040215 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species CCC(28) at 209469.38874040215 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species C[CH2](29) at 892271302189.8767 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species [CH3](31) at 59248654183.99109 exceeded the minimum rate for simulation interruption of 1.0
At time 0.0000e+00 s, species [CH2]CC(30) at 59248654183.99109 exceeded the minimum rate for simulation interruption of 1.0
Exceeded max number of objects...removing excess objects
Reached max number of objects...preparing to terminate
At time 0.0000e+00 s, species C[CH]CC(23) at rate ratio 2872877563043.813 exceeded the minimum rate for moving to model core of 0.1
terminating simulation due to interrupt...


Adding species C[CH]CC(23) to model core
Traceback (most recent call last):
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/site-packages/julia/pseudo_python_cli.py", line 308, in main
    python(**vars(ns))
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/site-packages/julia/pseudo_python_cli.py", line 59, in python
    scope = runpy.run_path(script, run_name="__main__")
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 263, in run_path
    pkg_name=pkg_name, script_name=fname)
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 96, in _run_module_code
    mod_name, mod_spec, pkg_name, script_name)
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/sevy/rmg/RMG-Py/rmg.py", line 118, in <module>
    main()
  File "/home/sevy/rmg/RMG-Py/rmg.py", line 112, in main
    rmg.execute(**kwargs)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/main.py", line 1029, in execute
    self.reaction_model.enlarge(objectToEnlarge)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/model.py", line 644, in enlarge
    reactions_moved_from_edge = self.add_species_to_core(new_species)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/model.py", line 1144, in add_species_to_core
    self.edge.phase_system.pass_species(spec.label, self.core.phase_system)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/reactors.py", line 144, in pass_species
    assert spc.name not in phasesys.phases[phase_label].names, spc.name
AttributeError: 'NoneType' object has no attribute 'name'

What's very strange is that it runs just fine if I include my DFT library as part of the seed.

How To Reproduce

My example input file.
input.txt

I was able to reproduce this with a very minimal version of my kinetics library.
Here's reactions.py:

#!/usr/bin/env python
# encoding: utf-8

name = "harris_butane"
shortDesc = ""
longDesc = """

"""
autoGenerated=False
entry(
    index = 288,
    label = "OH(15) + butane(1) <=> H2O(8) + [CH2]CCC(24)",
    degeneracy = 1.0,
    elementary_high_p = True,
    kinetics = Arrhenius(A=(3.94746e+06,'cm^3/(mol*s)'), n=2.27144, Ea=(-0.405729,'kJ/mol'), T0=(1,'K'), Tmin=(298,'K'), Tmax=(2500,'K'), comment="""Fitted to 50 data points; dA = *|/ 1.06999, dn = +|- 0.00888755, dEa = +|- 0.0483657 kJ/mol"""),
)

and dictionary.txt:

OH(15)
multiplicity 2
1 O u1 p2 c0 {2,S}
2 H u0 p0 c0 {1,S}

butane(1)
1  C u0 p0 c0 {2,S} {3,S} {5,S} {6,S}
2  C u0 p0 c0 {1,S} {4,S} {7,S} {8,S}
3  C u0 p0 c0 {1,S} {9,S} {10,S} {11,S}
4  C u0 p0 c0 {2,S} {12,S} {13,S} {14,S}
5  H u0 p0 c0 {1,S}
6  H u0 p0 c0 {1,S}
7  H u0 p0 c0 {2,S}
8  H u0 p0 c0 {2,S}
9  H u0 p0 c0 {3,S}
10 H u0 p0 c0 {3,S}
11 H u0 p0 c0 {3,S}
12 H u0 p0 c0 {4,S}
13 H u0 p0 c0 {4,S}
14 H u0 p0 c0 {4,S}

H2O(8)
1 O u0 p2 c0 {2,S} {3,S}
2 H u0 p0 c0 {1,S}
3 H u0 p0 c0 {1,S}

[CH2]CCC(24)
multiplicity 2
1  C u0 p0 c0 {2,S} {3,S} {5,S} {6,S}
2  C u0 p0 c0 {1,S} {4,S} {7,S} {8,S}
3  C u0 p0 c0 {1,S} {9,S} {10,S} {11,S}
4  C u1 p0 c0 {2,S} {12,S} {13,S}
5  H u0 p0 c0 {1,S}
6  H u0 p0 c0 {1,S}
7  H u0 p0 c0 {2,S}
8  H u0 p0 c0 {2,S}
9  H u0 p0 c0 {3,S}
10 H u0 p0 c0 {3,S}
11 H u0 p0 c0 {3,S}
12 H u0 p0 c0 {4,S}
13 H u0 p0 c0 {4,S}

Expected Behavior

RMG should be able to use any of its kinetics libraries as seed mechanisms without crashing.

Installation Information

Describe your installation method and system information.

• OS: Ubuntu 23.10
• Installation method: source
• RMG version information:
  • RMG-Py: b041507
  • RMG-database: f48dae1abf4b50cfbe5fa90df0e4ebe32257afcb

Additional Context

This problem has been ongoing for a while and is related to #2382

[sevyharris]

Interesting... it seems that getting rid of the (1) suffix that comes after the species label in the dictionary is enough to get this running.

[sevyharris]

Okay, this is super weird. I'm trying to recreate Hao-Wei's description of the bug in a simple RMS reactor with pressure dependence. I ran the ConstantVIdealGasReactor with a few extra libraries for seeds/kinetics, and it runs fine, but as soon as I add the pressure dependence block

pressureDependence(
    method='modified strong collision',
    maximumGrainSize=(0.5,'kcal/mol'),
    minimumNumberOfGrains=250,
    temperatures=(300,2000,'K',8),#due to thermo library max pdep=2500K cannot go to 3000K
    pressures=(0.01,100,'bar',5),
    interpolation=('Chebyshev', 6, 4),
    maximumAtoms=16,
)

I get this error:

1000
Traceback (most recent call last):
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/site-packages/julia/pseudo_python_cli.py", line 308, in main
    python(**vars(ns))
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/site-packages/julia/pseudo_python_cli.py", line 59, in python
    scope = runpy.run_path(script, run_name="__main__")
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 263, in run_path
    pkg_name=pkg_name, script_name=fname)
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 96, in _run_module_code
    mod_name, mod_spec, pkg_name, script_name)
  File "/home/sevy/miniconda3/envs/rmg_env/lib/python3.7/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/home/sevy/rmg/RMG-Py/rmg.py", line 118, in <module>
    main()
  File "/home/sevy/rmg/RMG-Py/rmg.py", line 112, in main
    rmg.execute(**kwargs)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/main.py", line 785, in execute
    self.initialize(**kwargs)
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/main.py", line 523, in initialize
    self.check_input()
  File "/home/sevy/rmg/RMG-Py/rmgpy/rmg/main.py", line 342, in check_input
    [not s.reactive for s in reaction_system.initial_mole_fractions.keys()]
AttributeError: 'ConstantVIdealGasReactor' object has no attribute 'initial_mole_fractions'

It's just an RMS/RMG compatibility thing. RMG expects all of its reactors to instantiate with the initial_mole_fractions attribute, but RMS never got the memo, which I guess means nobody's trying to run the ConstantVIdealGasReactor with pressure dependence. Hao-Wei was probably using a different system.

[sevyharris]

I checked, and renaming butane to my_butane doesn't cause problems. I think it just doesn't like the parentheses in the library dictionary?

• CurranPentane has some () in the dictionary due to using smiles strings and it hasn't caused problems for me.
• JetSurf2.0 has this thing: C5H5O(1,3) which is kind of like my (1) suffix. I can see if that causes problems
• KlippensteinGlarborg also has this: CH2(S) and it didn't cause problems.

In the case of the JetSurF2.0 one, maybe the model has to try to add the species and the library has to have () for it to be a problem:

C5H5O(1,3)
multiplicity 2
1  C u0 p0 c0 {2,S} {3,S} {6,S} {7,S}
2  C u0 p0 c0 {1,S} {4,S} {8,D}
3  C u1 p0 c0 {1,S} {5,S} {9,S}
4  C u0 p0 c0 {2,S} {5,D} {10,S}
5  C u0 p0 c0 {3,S} {4,D} {11,S}
6  H u0 p0 c0 {1,S}
7  H u0 p0 c0 {1,S}
8  O u0 p2 c0 {2,D}
9  H u0 p0 c0 {3,S}
10 H u0 p0 c0 {4,S}
11 H u0 p0 c0 {5,S}

JetSurF2.0 by itself as a library doesn't cause problems. But let's try adding the species that has parentheses in the name and see if that changes things. Nope. My experimental library is still the only place I've seen this bug.

[sevyharris]

Trying to understand the rules of this failure:

• RMG runs when only OH(15) is labeled with the () suffix
• RMG runs when only butane(1) is labeled with the () suffix
• RMG crashes when only [CH2]CCC(24) is labeled with the () suffix
• RMG runs when only [CH2]CCC(24) is labeled with the () suffix but I add my library as a seed and not just a kinetics library

[sevyharris]

All libraries are added to the edge by default in the initialization, which explains why my experimental library may be tripping things up.

[sevyharris]

I tried sanitizing my experimental library. It had a bunch of names with [ ] in it, so I got rid of these and it still crashed with the same error.

[sevyharris]

This line is writing to spec.label:

https://github.com/ReactionMechanismGenerator/RMG-Py/blob/886329862a1dbad152a7445319fd59c3d87ace99/rmgpy/rmg/output.py#L90C1-L90C46

Can we rewrite the HTML writer without overwriting spec.label?

[github-actions]

This issue is being automatically marked as stale because it has not received any interaction in the last 90 days. Please leave a comment if this is still a relevant issue, otherwise it will automatically be closed in 30 days.