Henry Rzepa's Blog

I wrote in an earlier post how Pauling’s Nobel prize-winning suggestion in February 1951 of a (left-handed) α-helical structure for proteins[cite]10.1073/pnas.37.4.205[/cite] was based on the wrong absolute configuration of the amino acids (hence his helix should really have been the right-handed enantiomer). This was most famously established a few months later by Bijvoet’s[cite]10.1038/168271a0[/cite] definitive crystallographic determination of the absolute configuration of rubidium tartrate, published on August 18th, 1951 (there is no received date, but a preliminary communication of this result was made in April 1950). Well, a colleague (thanks Chris!) just wandered into my office and he drew my attention to an article by John Kirkwood[cite]10.1063/1.1700491[/cite] published in April 1952, but received July 20, 1951,^‡ carrying the assertion “The Fischer convention is confirmed as a structurally correct representation of absolute configuration“, and based on the two compounds 2,3-epoxybutane and 1,2-dichloropropane. Neither Bijvoet nor Kirkwood seem aware of the other’s work, which was based on crystallography for the first, and quantum computation for the second. Over the years, the first result has become the more famous, perhaps because Bijvoet’s result was mentioned early on by Watson and Crick in their own very famous 1953 publication of the helical structure of DNA. They do not mention Kirkwood’s result. Had they not been familiar with Bijvoet’s[cite]10.1038/168271a0[/cite] result, their helix too might have turned out a left-handed one!

I record all this because I was today asked to provide an abstract for an NSCCS Themed Workshop shortly to be held at Imperial College on the uses of the Gaussian computational chemistry program in synthetic chemistry. One of the themes will be chiroptical spectroscopy. Gaussian of course deploys much of the theory developed by Kirkwood in the 1950s to make exactly the same sort of predictions that Kirkwood himself used to verify the Fischer convention in 1951. Whilst the majority of modern determinations of absolute configuration are still based on Bijvoet’s method,[cite]10.1038/168271a0[/cite] catching rapidly up are those based on chiroptical calculations. Perhaps in 2012 they are trusted more than they were in the 1950s? At any rate, such calculations are nowadays very much part of a modern undergraduate laboratory experience (slightly less so still in research laboratories I fear).

Here is another coincidence. Both Pauling and Kirkwood worked in the same department (Institute of Technology, Pasadena, California). One can only speculate on whether Kirkwood might have wandered into Pauling’s office in late 1951 to alert him that the protein helix should be right rather than left-handed (oh to have been a fly on Pauling’s blackboard). So alerted, would Pauling have foreseen that eventually such determinations would be routinely made using the very quantum mechanics that he had popularised?

^‡He first proposed a method of calculating absolute configurations as early as 1937[cite]10.1063/1.1750060[/cite], applying this to d-butan-2-ol as he noted it (also known as (+)-butan-2-ol), assigning an (2R)-configuration.[cite]10.1021/ja803119p[/cite] It took a further 15 years for him to apply his method to Fischer’s configuration!

Tags: California, chiroptical spectroscopies, computational chemistry, Imperial College, Institute of Technology, John Kirkwood, Pasadena, spectroscopy

This entry was posted on Tuesday, March 13th, 2012 at 12:56 pm and is filed under Chiroptics, Historical. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.