Adulteration of food is a public health concern with many documented incidents. For example, milk powder (MP) adulteration surfaced in 2008 where melamine contamination in Chinese products caused nearly 300,000 illnesses and six infants’ deaths. As such, comprehensive food protection requires the development of rapid screening methods to detect adulteration. To address this, a non-targeted technique using near-infrared (NIR) spectroscopy combined with chemometric analysis was developed for MP authentication screening. NIR spectroscopy was chosen due to its several advantages over traditional laboratory methods such as liquid chromatography coupled with mass spectrometry. These include NIR having rapid measurements that require limited or no sample preparation, which is ideal for portable instrumentation use. A single class, soft independent modelling by class analogy (SIMCA) model was developed using a diverse set of non-adulterated, commercial MPs as the training set. The model was then validated using a set of gravimetrically prepared MPs spiked with potential adulterants. The adulterants studied included nitrogen containing small molecules (e.g., melamine and biuret at 0-2.0% w/w in MP), inorganic salts (e.g., calcium carbonate at 0-2.0%), and plant proteins (e.g., pea protein and soy protein isolate at 1-20%). The SIMCA classification model yielded high specificity and sensitivity; it correctly classified all of the non-adulterated test samples as such (no false-positives), nitrogen containing small molecule spiked test samples at concentrations ≥0.6%, and plant protein spiked test samples at ≥5%. These NIR detection and SIMCA models show promise for rapid, on-site evaluation of MP for prevention of adulteration and protection of health.

Rapid detection of milk powder adulteration based on NIR spectroscopy and chemometric analysis

254th National Meeting (2017)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.