Identification of milk quality and adulteration by surface-enhanced infrared absorption spectroscopy coupled to artificial neural networks using citrate-capped silver nanoislands.

Milk is one of the most important multicomponent superfoods owing to its rich macronutrient composition. It requires quality control at all the production stages from the farm to the finished products. A localized surface plasmon resonance optical sensor based on a citrate-capped silver nanoparticle (Cit-AgNP)-coated glass substrate was developed. The fabrication of such sensors involved a single-step synthesis of Cit-AgNPs followed by surface modification of glass slides to be coated with the nanoparticles. The scanning electron microscope micrographs demonstrated that the nanoparticles formed monolayer islands on glass slides. The developed surface-enhanced infrared absorption spectroscopy (SEIRA) sensor was coupled to artificial neural networking (ANN) for the qualitative differentiation between cow, camel, goat, buffalo, and infants' formula powdered milk types. Moreover, it can be used for the quantitative determination of the main milk components such as fat, casein, urea, and lactose in each milk type. The qualitative results showed that the obtained FTIR spectra of cow and buffalo milk have high similarity, whereas camel milk resembled infant formula powdered milk. The most difference in FTIR characteristics was evidenced in the case of goat milk. The developed sensor adds several advantages over the traditional techniques of milk analysis using MilkoScan™ such as less generated waste, elimination of pre-treatment steps, minimal sample volume, low operation time, and on-site analysis.