The chromatogram from one of the first compound-specific S isotope analyses.
Measuring the abundance of stable isotopes in individual organic compounds is a very useful analytical technique that can provide detailed geochemical information about the origins of the molecules, geochemical processes that have altered them, and even the nature of environmental conditions millions or billions of years ago. Traditionally, these measurements are accomplished by connecting a standard gas chromatograph (GC) to an isotope-ratio mass spectrometer (IRMS) by way of a chemical reactor. This reactor either oxidizes or reduces organic species to simple molecular forms (H2, CO2, N2) that can then be analyzed by the IRMS (for technical details, see this paper). This is the technique that was pioneered by John Hayes and others in the 1980's, and is still used widely today. Its commonly called "CSIA", or 'compound-specific isotope analysis'. If you look at some of the other projects going on in the lab, you'll see that we use this method commonly to measure 13C and 2H.
Unfortunately, this chemical conversion does not work well for sulfur, and so there has not been any efficient way to measure sulfur isotopes in individual organic compounds. Given that the sulfur isotopic compositions of things like gypsum, pyrite, and bulk organic matter have already been extremely useful for understanding the history of the Earth's surface environment, we expect that we could learn quite a lot if only we were able to measure sulfur isotopes (32S and 34S) in specific organic compounds.
To pursue this idea, Alon Amrani and myself have been collaborating with Jess Adkins to couple a standard GC to an inductively-coupled plasma (ICP) mass spectrometer. The ICP uses a super-hot plasma to disintegrate organic compounds into their constituent atoms, including S. The idea is that this hot plasma bypasses all of the problems with chemical conversion mentioned above, and allows us to make isotopic measurements directly on atomic S ions (ie, 32S+ and 34S+).
We have been working on this for a little more than a year now, and everything basically works. We can measure d34S values with an accuracy of about 0.3permil in individual organosulfur compounds containing as little as 50 nmoles S. The figure above shows a typical chromatogram produced by this instrument.
So what comes next? Well, the world is our oyster at this point, and we will undoubtedly find a lot of sand along with a few pearls. But we are basically exploring a range of geologic and oceanographic samples, trying to understand which molecules can be readily measured, what sorts of isotopic variations exist, and what natural processes are represented in those isotopic variations. Representative examples include ancient high-sulfur crude oils, kerogens from the Monterey Formation (a Miocene-age organic-rich black shale), and dimethyl sulfide from ocean surface waters. There is obviously lots to be done here, and we welcome interested collaborators.