Laser spectroscopy spots fingerprints of fake foods

09 Aug 2023

Purdue University LIBS technique can tell authentic artisanal food from imitations.

Global food fraud, or "economically motivated adulteration" in the terminology of the US FDA, is estimated to be worth $40 billion each year, and stops customers getting what they think they have paid for.

Spotting the adulterations can be challenging, since global supply chains are complex, and replacement of an authentic ingredient with a similar product can occur at many different points.

A team at Purdue University has applied laser-induced breakdown spectroscopy (LIBS) to the problem, and developed a portable device that could provide an accurate and simple to use solution.

Reported in the journal Foods, the project's platform is intended to provide information about the atomic composition and chemical structure of a food sample, enough information to pinpoint its ingredients, preparation and potentially the point of origin.

"Think about the difference between a free-range ham from Portugal, aged in a cave for two years, and a ham you buy at WalMart," said Bartek Rajwa from Purdue University.

"They are both pig meat, the same ingredients, but they have a very different taste, smell and texture. To tell them apart, we need a system that can quantitatively analyze those characteristics. It's a big challenge."

LIBS involves using a laser to create a small plume of plasma from the surface of a target material. The light emitted by this plasma can be analyzed to reveal data about the constituents of the material, as an alternative to slower and more expensive fluorescence analysis.

The technique was developed for materials science applications, such as the detection of particular metals in scrap for recycling and quality control. A LIBS device is also currently active onboard NASA's Curiosity Mars rover, where the ChemCam instrument continues to take aim at Martian rocks and analyze their constituents.

LIBS plus Raman: atoms and molecules identified

LIBS has not previously been widely applied to food samples in general and liquid food samples in particular, according to the Purdue team, which tested different types of balsamic vinegar and vanilla extracts alongside solid samples of cheeses, coffee and spices. The raw LIBS spectra were processed using chemometric and machine-learning analytical routines to identify the constituents.

These trials found that LIBS and numerical analysis as a standalone analysis protocol was 99 percent accurate in distinguishing imitation vanilla flavoring from real vanilla extract, and about 90 percent accurate in identifying European cheese branded as Gruyère versus a Gruyère-style cheese produced in Wisconsin.

However, for more complex foods the LIBS spectrum alone may not be sufficient for reliable fraud detection. So the Purdue team has now developed a two-step procedure using both LIBS and Raman spectroscopy, allowing both the elemental and molecular composition of foods to be assessed.

Presented at this year's SPIE Defence + Commercial Sensing conference, a LIBS plus Raman approach should be capable of spotting not just fake ingredients but the presence of molecules associated with contaminants such as pesticides, fungicides or antibiotics in food.

Purdue's preliminary results indicated that the dual-modality method made it possible to assess complex food matrices polluted with widespread organic contaminants, by combining the different optical spectroscopies very rapidly.

"The techniques form a complementary pair; what one cannot detect, the other can," said Bartek Rajwa. "LIBS gives you the amount of each atom, and Raman tells you how they are organized."