26 No. 1
|Up for Discussion
||A forum for members and member organizations to share ideas and concerns.
Send your comments by e-mail to [email protected]
Questionable Stereoformulas of Diastereomers
Gerd Kaupp and M. Reza Naimi-Jamal
Wedge bonds are still in use for the characterization of relative configurations of asymmetric centers in racemic diastereomers (e.g., in Angew. Chem. Int. Ed. Engl. 2003, 42, 2550). Furthermore, there seems to be an increasing misuse of such formulas as eye-catchers in graphical abstracts, when the paper itself deals with achiral compounds or racemates, but not with absolute stereochemistry or enantioselectivity (e.g., Angew. Chem. Int. Ed. Engl. 2003, 42, 384; 2003, 42, 151; 2003, 42, 549 [does not deal with the usage of wedges in metal complexes]).
Our inquiry with journal editors points to a kind of desperate unwillingness to unify the usage of wedges for an unambiguous description of diastereomers. The violation of pertinent IUPAC recommendations prompts us to point out to the scientific community what we feel is a widespread bad habit and to urge an efficient remedy. Wedged and hatch-wedged bonds are excellent and unambiguous devices for the identification of absolute configurations in organic structural formulas. A chemical formula drawn in that manner is thus immediately attributed with R/S-descriptors by databases such as CAS/REG/SCIFINDER and stored for retrieval. There is no room for a debate that a structural formula must be correct by itself, irrespective of textural context or specifications (frequently it is not mentioned that a racemate was formed).
Indeed, the Beilstein database keeps with interpreting stereochemistry from the context of the formula or on its own judgment, adds and stores the word “racemate” if appropriate for its stereosearch facility not in the field “absolute” but in the fields “relative” and “racemic.” However, the database adheres to the wedged formulas and this may be the reason for frequent data errors by embracing epimers under one and only one “racemate” (Scifinder also frequently does this), as can be experienced with examples from papers cited below.
If an actual racemic diastereomer is published with one single-wedged formula (e.g., in J. Am. Chem. Soc. 2003, 125, 158; Eur. J. Chem. 2003, 1779; Tetrahedron 2003, 59, 3769; Schemes 3-5 in Angew. Chem. 2003, 42, 549), authoritative databases will at the end “pretend” absolute configurations and asymmetric syntheses. Still more disturbing is the reference of an unambiguous wedge formula to a single molecular structure image from an X-ray analysis under textual usage of starred R- and S-descriptors (e.g., [2R*,3R*,4S*,5R*] in Angew. Chem. Int. Ed. Engl. 1991, 30, 420), which are stored in the CAS/REG/SCIFINDER databases with the descriptors (2R,3R,4S,5R) (with suffix or prefix “rel”). In review articles this unambiguously formulated enantiomer may then easily appear unlabeled (e.g., in Chem. Rev. 2000, 100, 1025; on p.1048) and an absolute asymmetric synthesis is “pretended.” In that situation only the application of stereochemical principles and the analysis of the reported X-ray data can unambiguously reveal that in fact a racemate had formed and not only the depicted enantiomer. Numerous problems can be envisaged.
We therefore propose to face this widespread long-term nuisance and use wedged chemical formulas only for indicating absolute configuration. Thick and hatched (alternatively broken) bonds should be used for the representation of relative configurations in racemic diastereomers, as these, contrary to the wedges, do not necessarily indicate an absolute orientation in space. They are therefore more suited for the characterization of racemates with only one formula. Enantiomers of diastereomers with known absolute configuration should always be drawn using wedges (e.g., different than the usage in J. Am. Chem. Soc. 2003, 125, 3534). The automatic recording of configurations will then easily recognize the information “racemate” or “enantiomer.”
We wish to take the liberty to ask authors, referees, and editors to act accordingly in order to avoid further harm to our science. In the figure below, formula A, or even more distinct formula B (traditionally with exactly the same meaning), will then rightfully look like formula C (alternatively like formula D), if it is not the (1R,4S,5S)-enantiomer, but the racemate of this diastereomer
We have noticed that some authors apparently feel the same discomfort and already draw their formulas in the suggested manner (e.g., in J. Am. Chem. Soc. 2003, 125, 761; Eur. J. Chem. 2003, 1733; Angew. Chem. 2003, 42, 694).
Prof. Gerd Kaupp <[email protected]> and Dr. M. Reza Naimi-Jamal are at the University of Oldenburg in Oldenberg, Germany.
Click here to read the Response from
the IUPAC Chemical Nomenclature and Structure Representation
last modified 7 January 2004.
Copyright © 2003-2004 International Union of Pure and
Questions regarding the website, please contact [email protected]