A fresh approach to a standard analytical technique can detail a complicated mixture’s components without destroying the sample, allowing researchers to perform a ‘virtual dissection’ of a glass of beer.
Understanding the chemical contribution of each component in a complex system, such as a living cell, often means physically separating those components. That destroys the system and changes its chemistry. A technique called nuclear magnetic resonance spectroscopy (NMR) can identify an individual chemical by its unique radio-frequency signal, which shows up as a peak on a graph. But in complex mixtures, these signals overlap so much that disentangling them is almost impossible.
Instead of opting for the usual workaround — dissecting the sample — Mathias Nilsson at the University of Manchester, UK, and his colleagues divided the NMR measurements into small regions of the radio-frequency spectrum and examined one region at a time using statistical techniques. The approach, which the scientists dub SCALPEL, makes the individual signals stand out more clearly.
With this refinement, the team easily identified the maltose, lactose, glucose and ethanol in a glass of good British stout.