Stereochemistry of Tetrachloridoplatinum(II) with istopes 35Cl and 37Cl is not recognized (Richard Hartshorn)

[gblanke02]

Gerd,

Further to my questions around interactions of the stereochemical and isotopomer developments:

Tetrachloridoplatinum(II) is a square planar complex (and not tetrahedral, which is presumably a distinction that will already be covered). If there are two 35Cl and two 37Cl, there are two stereoisomers (cis and trans) which will have different properties (e.g. different IR spectra) and may need to be discussed, at least by theoreticians who might be calculating those properties.

mer-[Co(NH3)3Cl3] could, for example. have two 35Cl and one 37Cl, in which case there are two stereoisomers depending on where the 37Cl is located…

Richard

Remarks: The cis and trans versions of Pt 35Cl2 37Cl2 should have two different equivalence classes, one describing the cis the other the trans isomers of this flat Pt complex. Both versions are not distinguished althoughthe different isotopes are recognized:

InChI=1/Cl4Pt/c1-5(2,3)4/i1+0,2+0,3+2,4+2

Alternatve: Identify the equivalence classes during the determination steps of the stereochemistry, where an additional step is needed to didferentiate different isotopes of the same element.

[schatzsc]

@gblanke02

Important point - differences in isotopes need to be included in the determination of the equivalence classes. In the example above, 35Cl has to go into one equivalence class and 37Cl into another one (and if there are more Cl with un-defined isotope assignment, those need to go into a third equivalence class)

[fbaensch-beilstein]

@gblanke02

Important point - differences in isotopes need to be included in the determination of the equivalence classes. In the example above, 35Cl has to go into one equivalence class and 37Cl into another one (and if there are more Cl with un-defined isotope assignment, those need to go into a third equivalence class)

If I see it correctly, this does not work with either organic or inorganic substances.

[schatzsc]

@fbaensch-beilstein @gblanke02

"Does not work" in the sense of "is currently not implemented"?!?

Consider 3-methyl-2-butene with the molfile for one of the isotopomers attached below. If you exchange the 12C and 13C isotope positions in the methyl groups, you get two InChIs which differ in the /b (= double bond?!?) layer.

InChI=1S/C5H10/c1-4-5(2)3/h4H,1-3H3/i2+0,3+1/b5-4-

InChI=1S/C5H10/c1-4-5(2)3/h4H,1-3H3/i2+0,3+1/b5-4+

However, if you go to the "Convert molfile to InChI" option from the webdemo, the equivalence class of the two C atoms is the same?!?

  -INDIGO-12092518012D

  5  4  0  0  0  0  0  0  0  0999 V2000
   10.2920   -5.1750    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
   11.1580   -4.6750    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
    9.4260   -4.6750    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
   10.2920   -6.1750    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
   12.0240   -5.1750    0.0000 C   0  0  0  0  0  0  0  0  0  0  0  0
  1  2  2  0  0  0  0
  1  3  1  0  0  0  0
  1  4  1  0  0  0  0
  2  5  1  0  0  0  0
M  ISO  2   3  13   4  12
M  END

[fbaensch-beilstein]

"Does not work" in sense of there are no different equivalence classes for isotopes.

[schatzsc]

@fbaensch-beilstein @gblanke02

The algorithm to determine the equivalence classes needs to be expanded to also include isotope composition then.

In that context, it is also important that "no isotope distribution specified" (= natural isotope distribution?!!) has to be an additional equivalence class separate from those with specific isotopes.

Furthermore, is it correct that the molfile cannot distinguish between "natural isotope distribution" and "isotope distribution unknown"?!? (Although I assume that the latter would be very rare ...)

[fbaensch-beilstein]

I would like to note that we could run the risk of altering existing organic InChIs or creating a divergent behavior for isotope treatment between organic and molecular inorganic structures.

[schatzsc]

@fbaensch-beilstein @gblanke02

Well, you could always build an additional "equivalence class 2.0" on top of the exisiting system and only apply it to structures containing metals but I think that will not really help.

For example, think not of the 35Cl/37Cl example from Richard Hartshorn where the different isotopes are directly bound to the metal but a situation like Di(n-butyl)(propyl)(ethyl)tin where one of the distal (away from the tin) carbons of the terminal CH3 group is 12C and the other 13C than this would be a chiral Sn center. And the same would even be the case when you replace Sn by Si, Ge, and even carbon.

And if these central element are to main group for you, just imagine using a Ti(IV) instead ...

[gblanke02]

Hi Uli, You are correct. The molfile does not distinguish between “natural isotope distribution” and “unknown isotope distribution”. Best Gerd Von: Ulrich Schatzschneider ***@***.***> Gesendet: Donnerstag, 11. Dezember 2025 01:04 An: IUPAC-InChI/InChI ***@***.***> Cc: ***@***.***; Mention ***@***.***> Betreff: Re: [IUPAC-InChI/InChI] Stereochemistry of Tetrachloridoplatinum(II) with istopes 35Cl and 37Cl is not recognized (Richard Hartshorn) (Issue #175) [https://avatars.githubusercontent.com/u/88166560?s=20&v=4]schatzsc left a comment (IUPAC-InChI/InChI#175)<#175 (comment)> @fbaensch-beilstein<https://github.com/fbaensch-beilstein> @gblanke02<https://github.com/gblanke02> The algorithm to determine the equivalence classes needs to be expanded to also include isotope composition then. In that context, it is also important that "no isotope distribution specified" (= natural isotope distribution?!!) has to be an additional equivalence class separate from those with specific isotopes. Furthermore, is it correct that the molfile cannot distinguish between "natural isotope distribution" and "isotope distribution unknown"?!? (Although I assume that the latter would be very rare ...) — Reply to this email directly, view it on GitHub<#175 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AT3EDCITBA7NTNI2U45TNDL4BCYGFAVCNFSM6AAAAACNNCO2VWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTMMZZGQZTMNZYHE>. You are receiving this because you were mentioned.Message ID: ***@***.******@***.***>>

[schatzsc]

@gblanke02 Well, what's not in the molfile is not accessible to the InChI algorithm, as it needs to be "self-contained" and cannot refer to information that's outside the molfile and the code itself.

Thus, for an atom we can only have the two cases 1) "one particular isotope specified" or 2) "no isotope specified".

(Specifying that one atom is either 13C or 14C, for example, would also not be possible in the molfile, right?)

Still, we need one equivalence class for each specified isotope of an element and one for "not specified".