Icy response to isotope ratios
Publication of a study on C isotope ratios of methyl halides and chlorofluorocarbons has kicked up a storm among experts in the field.
Atmospheric concentrations of halogenated volatile organic compounds (halo-VOCs) are small, but have a major impact on atmospheric chemistry, particularly by depleting the ozone layer. Isotope ratio mass spectrometry (IRMS) is used to determine the sources and sinks of gases after an initial GC step, but the quality of this GC separation has become the subject of intense debate.
The trace halo-VOCs are so dilute that a preconcentration step is required to enable detection. Researchers at Queen's University, Belfast, UK, have designed and validated a five-component preconcentration system1 linked to IRMS and GC/MS analysers for measurement of isotope ratios and identification and quantification of each halo-VOC. They evaluated the preconcentration system piece-by-piece for six standard compounds to confirm that none of the five components introduced isotope fractionation, then applied the methodology to local air samples.
This work has drawn sharp criticism from other researchers working in similar areas. One of the major concerns is with the methodology. Allen Goldstein, a biogeochemist from the University of California, Berkeley, US, says that likely interferences from closely eluting peaks render some of the data highly uncertain, especially for methyl bromide, and could lead to confusion for readers unfamiliar with this field. This point is echoed by Jochen Rudolph, a researcher in atmospheric chemistry at York University, Canada, who published the first application2 of preconcentration of atmospheric VOCs for IRMS.
Kelly Redeker, one of the authors of the paper, replies that peak overlap was not an issue with standard compounds, where there was sufficient separation between each compound in the chromatogram. For air samples, he agrees that methyl bromide gradually became a problem due to baseline drift and has subsequently been dropped from the analyses, but the remaining analytes had good separation.
Redeker also says that the method reproducibilities can be defended against potential separation and baseline problems, but insists that 'we were not asked to do so by the reviewers of the article'.
John Monaghan, editor of Rapid Communications in Mass Spectrometry, which published Redeker's paper, said: 'We at RCM can only proceed on the basis of what our referees tell us and in this case they both supported publication with minor changes.'
Previous studies on systems of this type validated the preconcentration system as a whole. 'The explicit testing of the components of the condensation system has not been performed previously [within IRMS methodology],' said Redeker. 'We felt that.it was conceivable that one component could enrich while the next depleted the compound [isotope ratio] signature.'
Nevertheless, Rudolph says many statements in the paper imply that previous studies did not adequately validate the measurement procedure.
Rudolph also suggests that the performance of the method as a whole and the basis for extrapolating from ppb levels of standards to ppt levels of atmospheric samples were poorly explained.
After debating these points with Redeker, Rudolph concludes that the paper, as published, still has several problems. In their defence, Redeker agrees that more discussion would have clarified some of their conclusions, but claims that their work supports previous studies and is supported by them, since their ambient air data are similar to reported data for CFC-113 and methyl chloride. In addition, they have increased the number of halo-VOCs measured in ambient air. Steve Down
1 M E Archbold et al, Rapid Commun. Mass Spectrom., 2005, 19, 337
2 J Rudolph et al, Geophys. Res. Lett., 1997, 24, 659