Rule set of molecular inorganics in wrong order

[fbaensch-beilstein]

It looks as though the rules for inorganic molecular chemistry are being applied in the wrong order.

This structure

produces the following InChI:

InChI=1B/C30H20O14Ti/c31-11-1-3-15(17(35)5-11)27-25(39)29(23-19(37)7-13(33)9-21(23)41-27)43-45-44-30-24-20(38)8-14(34)10-22(24)42-28(26(30)40)16-4-2-12(32)6-18(16)36/h1-10,31-40H/q+2/p-2

The two coordinate bonds would first have to be converted into single bonds, and then the rules would apply and break the titanium-oxygen bonds, as titanium has a valence of 4.

stereochemistry.ds/mol_stereochemistry.ds_145.mol
Generated by WebMolKit

  0  0  0     0  0            999 V3000
M  V30 BEGIN CTAB
M  V30 COUNTS 45 52 0 0 1
M  V30 BEGIN ATOM
M  V30 1 Ti 7.3506 0.0000 0.0000 0
M  V30 2 O 6.1371 1.3858 0.0000 0
M  V30 3 O 8.5641 -1.3858 0.0000 0
M  V30 4 O 8.5641 1.3858 0.0000 0
M  V30 5 O 6.1371 -1.3858 0.0000 0
M  V30 6 O 8.1318 5.5124 0.0000 0
M  V30 7 O 6.5694 -5.5124 0.0000 0
M  V30 8 O 3.4466 2.8024 0.0000 0
M  V30 9 O 11.2546 -2.8024 0.0000 0
M  V30 10 O 3.3839 8.2243 0.0000 0
M  V30 11 O 11.3173 -8.2243 0.0000 0
M  V30 12 O 10.3725 1.5381 0.0000 0
M  V30 13 O 4.3287 -1.5381 0.0000 0
M  V30 14 O 14.9023 4.0839 0.0000 0
M  V30 15 O -0.2011 -4.0839 0.0000 0
M  V30 16 C 6.6006 2.8123 0.0000 0
M  V30 17 C 8.1006 -2.8123 0.0000 0
M  V30 18 C 8.1006 2.8123 0.0000 0
M  V30 19 C 6.6006 -2.8123 0.0000 0
M  V30 20 C 5.7985 4.1534 0.0000 0
M  V30 21 C 8.9025 -4.1534 0.0000 0
M  V30 22 C 8.9025 4.1534 0.0000 0
M  V30 23 C 5.7985 -4.1534 0.0000 0
M  V30 24 C 6.5694 5.5124 0.0000 0
M  V30 25 C 8.1318 -5.5124 0.0000 0
M  V30 26 C 10.4025 4.1360 0.0000 0
M  V30 27 C 4.2987 -4.1360 0.0000 0
M  V30 28 C 4.1715 4.1157 0.0000 0
M  V30 29 C 10.5297 -4.1157 0.0000 0
M  V30 30 C 5.7664 6.9281 0.0000 0
M  V30 31 C 8.9346 -6.9281 0.0000 0
M  V30 32 C 3.3361 5.5124 0.0000 0
M  V30 33 C 11.3651 -5.5124 0.0000 0
M  V30 34 C 4.1389 6.9281 0.0000 0
M  V30 35 C 10.5621 -6.9281 0.0000 0
M  V30 36 C 11.1374 2.8284 0.0000 0
M  V30 37 C 3.5636 -2.8284 0.0000 0
M  V30 38 C 11.1676 5.4263 0.0000 0
M  V30 39 C 3.5336 -5.4263 0.0000 0
M  V30 40 C 12.6374 2.8110 0.0000 0
M  V30 41 C 2.0638 -2.8110 0.0000 0
M  V30 42 C 12.6674 5.4089 0.0000 0
M  V30 43 C 2.0338 -5.4089 0.0000 0
M  V30 44 C 13.4023 4.1013 0.0000 0
M  V30 45 C 1.2988 -4.1013 0.0000 0
M  V30 END ATOM
M  V30 BEGIN BOND
M  V30 1 2 2 16
M  V30 2 2 3 17
M  V30 3 1 4 18
M  V30 4 1 5 19
M  V30 5 1 6 22
M  V30 6 1 6 24
M  V30 7 1 7 23
M  V30 8 1 7 25
M  V30 9 1 8 28
M  V30 10 1 9 29
M  V30 11 1 10 34
M  V30 12 1 11 35
M  V30 13 1 12 36
M  V30 14 1 13 37
M  V30 15 1 14 44
M  V30 16 1 15 45
M  V30 17 1 16 18
M  V30 18 1 16 20
M  V30 19 1 17 19
M  V30 20 1 17 21
M  V30 21 2 18 22
M  V30 22 2 19 23
M  V30 23 1 20 24
M  V30 24 2 20 28
M  V30 25 1 21 25
M  V30 26 2 21 29
M  V30 27 1 22 26
M  V30 28 1 23 27
M  V30 29 2 24 30
M  V30 30 2 25 31
M  V30 31 2 26 36
M  V30 32 1 26 38
M  V30 33 2 27 37
M  V30 34 1 27 39
M  V30 35 1 28 32
M  V30 36 1 29 33
M  V30 37 1 30 34
M  V30 38 1 31 35
M  V30 39 2 32 34
M  V30 40 2 33 35
M  V30 41 1 36 40
M  V30 42 1 37 41
M  V30 43 2 38 42
M  V30 44 2 39 43
M  V30 45 2 40 44
M  V30 46 2 41 45
M  V30 47 1 42 44
M  V30 48 1 43 45
M  V30 49 1 1 5
M  V30 50 1 1 4 CFG=3
M  V30 51 9 1 2 CFG=1 DISP=COORD
M  V30 52 9 1 3 DISP=COORD
M  V30 END BOND
M  V30 END CTAB
M  END
>  <dataset>
stereochemistry.ds

>  <source_file>
mol_stereochemistry.ds_145.mol

>  <previous_comparison>
differ

>  <std_full>
InChI=1S/2C15H10O7.Ti/c2*16-6-1-2-8(9(18)3-6)15-14(21)13(20)12-10(19)4-7(17)5-11(12)22-15;/h2*1-5,16-19,21H;/q;;+2/p-2

>  <molinorg_full>
InChI=1B/C30H20O14Ti/c31-11-1-3-15(17(35)5-11)27-25(39)29(23-19(37)7-13(33)9-21(23)41-27)43-45-44-30-24-20(38)8-14(34)10-22(24)42-28(26(30)40)16-4-2-12(32)6-18(16)36/h1-10,31-40H/q+2/p-2

>  <comparison>
differ

[nnuk]

@fbaensch-beilstein Thanks for your research in finding out this issue. I have modified the concerned condition to resolve the issue. It can now be tested on the InChI Web Demo.

[gblanke02]

The issue is seen as well with NaOH, NaOCMe, or LiOH in case these compounds are drawn with coordinative bonds. As found in our (Nauman, me) discussion yesterday the adaption of tehvalences and the neutralization of charged atoms linked by the corrdinative bonds does not work properly. Attached you find examples for NaOH and NaOCMe that show the same issue and how they can be corrected by adapting the valences properly according to our coordinative bond exchange valence rules amd deleting the opposite charges of neighbored atoms.

Coordinative bond issue.zip

[nbehrnd]

@fbaensch-beilstein I would like to comment the structure as depicted in your initial post (uploaded May 26) because it looks like to contain an error. If used (vide infra) I would orient the arrows about dative bonds from the Lewis base pointing to the Lewis acid. To me, owing the non-bonding electron pairs, both oxygen atoms of the \beta hydroxy carbonyl moieties are electron rich (i.e., Lewis bases), the transition metal cation the moiety with the empty / accepting orbital (i.e., the Lewis acid). Thus the arrows would point in the direction opposite to the one depicted in the illustration by May 26.

Another point could be if / how «dative bonds» differ from «normal bonds». While a discussion in the forum of RDKit (link) noted some programs ([in 2023/2024] still) required this bond type, input provided by Ulrich Schatzschneider / @schatzsc for instance initiated Avogadro2 a revision; since January 16, 2024, its ligands (public fragments repository) don't use the arrows any more: