Raman spectroscopy spots contaminated alcohol in the bottle

Detection of unwanted chemicals in counterfeit or contaminated alcoholic drinks is a serious concern for both public health and the drinks trade, with the presence of methanol in spirits posing a particular problem.

Optical spectroscopy techniques are attractive approaches to the task, not least since they could potentially be used to analyse spirits in the bottle, without needing to extract a sample for conventional chromatography.

But spirits come in a variety of glass containers, and anything other than clear uncolored glass can hinder the analysis. In addition, fluorescence from both sample and container can be orders of magnitude stronger than the underlying Raman signals of interest.

A project at Adelaide University and the University of St Andrews has now demonstrated a non-invasive Raman technique able to quantify methanol in a range of bottled spirits, and described its findings in J Phys Photonics. Raman is particularly useful for methanol detection, as it can selectively indicate methanol in the presence of ethanol thanks to their spectrally distinct Raman peaks.

The project builds on previous research at both institutions. In 2020 St Andrews demonstrated how a Raman approach using a shaped laser beam could generate a spectroscopic response from the contents of a bottle of whisky with minimized interference from the glass. In that instance the laser illumination was an annular beam formed on the bottle surface which refocuses inside the bottle.

For its new technique the project also employed a beam focused to a point inside the bottle, but combined wavefront shaping and wavelength modulation for the first time, as a way to suppress signals from the container while enhancing key features of the Raman response from the contents.

Customs checkpoints, distilleries and food manufacturers

A wavelength of 785 nanometers was selected, to minimise fluorescence emission from the spirits while ensuring adequate transmission through green glass bottles, with the wavefront shaped into a Bessel beam by a suitable lens. The Raman signal of the liquid sample is excited at the focal point inside the bottle and collected through the middle 'dark' region of the annular beam, said the project, avoiding any signal from the container.

"Wavelength selection alone cannot fully overcome sample fluorescence while maintaining optimal transmission through the bottles," noted the team in its paper. "To further suppress fluorescence and enhance sensitivity, we combined the wavefront-shaped excitation beam with wavelength-modulated Raman spectroscopy (WMRS). In WMRS, the excitation wavelength is tuned, shifting the Raman peaks while the fluorescence background remains constant."

For its experimental trials, the project applied the technique to seven commercially available spirits bottled in different glass containers. The method achieved quantification of methanol through colored spirit bottles, with a limit of detection of 0.2 percent by volume methanol in 40 percent ethanol, well below the reported maximum tolerable methanol concentration of 2 percent by volume.

These findings point towards uses for this Raman technology in practical real-world scenarios, including customs checkpoints, distilleries, food manufacturers or quality assurance facilities.

"Being able to identify the contents of a sealed bottle without opening it has enormous potential," commented Ralf Mouthaan from Adelaide's Centre of Light for Life. "At Adelaide University we're now adapting this technology to tackle problems that directly affect Australian industries, including wine authentication, food quality and product safety."