CHEMISTRY: Convoluted Chromatography

See allHide authors and affiliations

Science  06 Jul 2007:
Vol. 317, Issue 5834, pp. 19b
DOI: 10.1126/science.317.5834.19b

A drawback of chromatographic separations is the waiting time necessary for analytes to travel from the injection site to the detector. High-throughput screens are often limited by this waiting period, during which isolated signal peaks punctuate a largely silent detection baseline. Recently, spectroscopic analysis has benefited from sophisticated mathematical algorithms that facilitate deconvolution of many overlapping signals from a single data set, thereby allowing multiple samples to be analyzed all at once. Trapp has implemented a similar multiplexing approach to gas chromatography. Specifically, he assigned a distinct binary injection sequence to each sample (with each “1” prompting injection and each “0” no action). Multiple samples were then injected continuously onto a separation column in accord with their assigned bar-code sequence, resulting in a much higher proportion of detected signals during a given time period than in traditional chromatography. The overlapping data could be deconvolved into individual chromatograms by means of a Hadamard transform and subsequent matrix manipulations. The author analyzed samples composed of several organic alcohols and hydrocarbons as a proof-of-principle and noted an enhancement in efficiency of nearly a factor of 40. — JSY

Angew. Chem. Int. Ed. 46, 10.1002/anie.200605128 (2007).

Navigate This Article