Singlet O=O vdW species in Surface_Adsorption_Abstraction_vdW

[Tingchenlee]

As we discussed in #458, the gas phase O2 will be represented as [O][O], which has the following adjaceny list notation:

1 O u1 p2 {2,S}
2 O u1 p2 {1,S}

Since we don’t use the singlet O=O for the gas phase oxygen, do we want the vdW O2 to have this form in Surface_Adsorption_Abstraction_vdW family?

entry(
index = 11,
label = "O=O",
group =
"""
multiplicity [1]
1 *1 Xv u0 p0 c0
2 *2 O u0 p2 c0 {3,D}
3 *3 O u0 p2 c0 {2,D}
""",
kinetics = None,
)

We didn’t add the forbidden entry O=O in Surface_Adsorption_Bidentate family as well, and the following reaction will be created:

For consistency, should we keep all of them or remove the O=O entry

[rwest]

Short term

as discussed 2021-04-12 on Northeastern/Brown call

• In all surface families treat all O2(g) as being the triplet [O][O].
• Ensure the triplet can react. So that it's made in the reverse direction.
  • might have to put this in a reaction library, because it's too special a case to fit in a family Forbid the singlet O=O
  • make sure people use the library!

Medium term.

Read papers like

• https://doi.org/10.1016/j.electacta.2020.137175
• https://doi.org/10.3389/fchem.2020.00605

[rwest]

Further thoughts...

In the Surface_Adsorption_Abstraction_vdW family...

if we block O=O from being *2=*3 (because we want O2 to be [O][O]) then it can't react in that family at all. We can't just add a single reaction into a library to replace it, because it could be reacting with any *4 and *5.

We could either

(a) add a new reaction family for adjacent biradical, i.e. probably just triplet oxygen,

or (b) say that when O2 is van-der-waals adsorbed it's represented as the singlet O=O, and hence i) block O2(T) from the Surface_Adsorption_vdW family and ii) make a special library reaction for the van-der-waals adsorption like [O][O] + X <=> O=O.X. Then check how O=O.X would behave in any other vdw families, like Surface_Addition_Single_vdW (it's already there, entry 11) and Surface_Dissociation_Double_vdW and Surface_Dual_Adsorption_vdW (currently blocked)

I guess a generic question: should O2 ever be van-der-waals adsorbed and reacting in all these families? or are we better off just blocking it entirely? (I know vdw O2* exists - there are binding energies for it in cathub - so I think the answer is that we should deal with it - but does it react in any of these ways?)

[alongd]

Related: S2(g) and SO(g) are both also triplets at the ground state, should probably be treated the same as O2(g), perhaps supporting the approach of using a template rather than a library.