Ambiguity in NIST-UNIFAC subgroup 309 allocation

[WingsOfThePhoenix]

Thanks for such a great project, Caleb.

Not a problem with the software as such, but the NIST-UNIFAC (2014) method). Whilst looking at fragmentation method in the NIST paper, I noticed that two components have been allocated subgroup number 309: acetal (CH2(O)2) in main-group 74; and oxime (CH=NOH) in main-group 82. These two components have different van der Waals surface area and volume and the main-groups have different binary interaction parameters; however it will always be interpreted as oxime because it is referred to later in the NIST-UNIFAC parameter allocation in the thermo module.

Suggest that maybe a warning is issued when using NIST-UNIFAC subgroups as a group contribution method? As I said, not an issue in the software, but ambiguity in the method.

UPDATE: NIST-Unifac parameters defined in the SI of https://www.sciencedirect.com/science/article/pii/S0378381214007353?ref=pdf_download&fr=RR-2&rr=8e508110f8779553

[CalebBell]

Hi,
You are dead on correct. This issue wasn't caught by any of the reviews of the NIST UNIFAC method, although many mistakes were. Sure enough their error is in thermo:

NISTUFSG[309] = UNIFAC_subgroup(309, 'CH2(O)2', 74, 'Acetal', 0.9347, 0.708)
NISTUFSG[309] = UNIFAC_subgroup(309, 'CH=NOH', 82, 'Oxime', 1.499, 1.46)

The second one, Oxime, overwrites the first. There is no way to use the acetal group. A different ID would need to be assigned to one of the groups, probably the acetal one. Someone could email the authors of the paper and ask for a correction and a correct ID. Without that, I think it makes sense to create a new ID for acetal that will be sure not to conflict with one in an updated NIST paper - maybe 10309 or 10000.

Would you like to prepare a PR? I believe just changing the ID and making a small note of it in the "Data for NIST UNIFAC (2015)" section of the documentation would be sufficient.
Sincerely,
Caleb