|Harnly, James - Jim|
|Bergana, Marti - United States Pharmacopeia|
|Adams, Kristie - United States Pharmacopeia|
|Moore, Jeffrey - United States Pharmacopeia|
|Xie, .zhuohong - United States Pharmacopeia|
Submitted to: Journal of Agricultural and Food Chemistry
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/11/2018
Publication Date: 7/11/2018
Citation: Harnly, J.M., Bergana, M., Adams, K., Moore, J., Xie, Z. Moore, J. 2018. Variance of commercial powdered milks analyzed by 1H-NMR and impact on detection of adulterants. Journal of Agricultural and Food Chemistry. https://doi.org/10.1021/acs.jafc.8b00432.
DOI: https://doi.org/10.1021/acs.jafc.8b00432 Interpretive Summary: Commercial samples of skim milk powder and non-fat dry milk were analyzed by proton nuclear magnetic resonance (NMR) spectrometry based on a non-targeted approach, i.e., we approached the analyses with no intended target compounds. Initially, we found an unusual noise source related to the data processing of the NMR data. After eliminating the noise source, we could accurately the determine variations in the data arising from the type of milk powder, the producer, the manufacturing site, and the day of analysis. Each factor significantly affected the measured spectra. In addition, the reduced noise allowed better detection of adulteration. Six different concentrations of eight different adulterants were spiked into authentic milk powder samples. Strong nitrogen adulterants like melamine and dicyandiamide were easily detected, but sugar and protein sources were more difficult to detect. These data indicate that a local standard is better for determining the quality of the milk product and detecting adulteration than a global standard.
Technical Abstract: 1H-NMR spectra for 66 commercial powdered milk samples were analyzed by PCA, SIMCA, and crossed multivariate ANOVA. It was found that the sample type (skim milk powder or non-fat dry milk), the production company, the production site, the processing temperature (high, medium, or low temperature), and the day of analysis provided statistically significant sources of variation. Interestingly, inexact alignment (deviations of ±0.002 ppm) of the spectral reference peak was a significant source of variation and fine alignment was necessary before the variation arising from the other experimental factors could be accurately evaluated. Using non-targeted analysis, the lowest detectable adulteration for dicyandiamide, melamine, and sucrose was 0.05%, maltodextrin and urea was 0.5%, ammonium sulfate and whey was 5%, and soy protein isolate was undetectable using methods described herein. The measurement of variance and detection of adulteration was unaffected by the resolution. Similar results were obtained with un-binned data (0.0003 ppm resolution) and binning of 333 data points (0.1 ppm resolution).