Thursday, January 5, 2012

A Chiral Conundrum

While browsing on Nature News this afternoon, I came across a puzzling chemical graphic in the article "Frontier Experiments: Tough Science." Under the heading "Seeing through the molecular mirror," came this picture of two hands, showing the chirality around a Re center...
Source: Thomas Porostocky / Nature News


OK, I understand the point of the graphic, but I'd like to tackle a few pedagogical questions. First, has anyone ever seen a real-life compound like this? I searched Reaxys, but couldn't find any examples of a Re center bound to three different chalcogens. Closest I could find were some rhenium selenide clusters. 


Second, what's the valency around that Re center? My good buddy Wikipedia (source of all truthful information on the internet, right?) informs me that Re can have oxidation numbers from -1 to +7. Well, OK, so I'll assume double bonds between all the heteroatoms and Re (which also doesn't come up via Reaxys or our buddy Google), but I'd still question whether, as drawn, we have Re(+4), five, six, or seven!


So let's assume a stable compound...is it actually "chiral?" 


Textbooks on inorganic chemistry indicate that Re(VII) complexes with oxo ligands prefer the square pyramidal geometry. Well, you say, it has two other bivalent chalcogens, so maybe it's tetrahedral after all? My ligand field theory spider sense tells me that Re(VII) is d0, which should have a tetrahedral geometry, but I don't know enough about this peculiar complex to know if it takes on solvent molecules (becoming octahedral?) or undergoes inversion. Maybe in the gas phase (lasers!), in a vacuum, it's just a plain ol' tetrahedron.


I'm not attacking the artist, he's a graphical designer, not a chemist. And I get the message - it's a physical chemistry model experiment to show how chiral molecules show slightly different physical parameters, perhaps ones that can be measured. The designers use a single periodic column to show how molecular number assigns chirality (C-I-P convention) and to enforce a striking enantiomeric disparity. It's just hard for me, as a non-physical chemist, to go this far out on the edge.


Please, to illustrate this concept, let's use known examples (amino acids, anyone?) first, and then transition out to the edge. Then we won't suffer Feynman's cargo cult caveat: "...you should not fool the layman when you're talking as a scientist."


Update (1/5/12, 3:44PM) - I realize that the Darquie experiments referred to in the text involve (potential) formation of such a compound. I simply find "strange bedfellows" for text that leads with biological homochirality, since Re(VII) compounds are unlikely to be something one encounters in cells!


(1/5/12, 4:19) - Cleaned up text in grafs 2-3 to focus the post.

1 comment:

  1. I didn't see this post until just earlier today, so I hope there are no objections to a belated comment.

    The major issue I always saw it (which is mentioned in the preprint, and has been previously noted in other papers on this topic) is that the energy difference scales as a function of the heaviest nuclear charge present (Z to the fifth power). I don't think the article emphasized the magnitude of the difference very clearly. In principle, it's going to be a lot easier to unambiguously measure this difference when you've got rhenium (Z = 75) at a chiral center versus one of the second-row elements (take your pick) that compose amino acids and carbohydrates.

    So, definitely - doing these measurements on alanine and glucose isomers would be directly more relevant to the entire "origin of biomolecular homochirality" goal. But the spectroscopic state-of-the-art seems to currently restrict us to these heavy-metal compounds as the best candidates to unambiguously detect such difference. Of course, the worst-case-scenario goal would be that we learn how to construct better spectrometers, further test our knowledge of physics and discover new things, and then build even better spectrometers with that knowledge so we can try to detect energy differences between L-alanine and D-alanine. Heh.

    ReplyDelete