Milk fat is widely used in food products especially fat spreads and chocolate products. The usual method to determined milk fat is by fatty acid profile analysis by gas chromatography, where butyric acid is regarded as the principle marker of authenticity. However, this method is quite laborious to carry out, and more recently analysts have turned to proton NMR as a more rapid method. Romanian researchers have evaluated the published NMR method, and have discovered that the triplet used in the NMR spectra for butyric acid and milk fat authentication is also produced by n-3 fatty acids such as linolenic acid (C18:3), and therefore may not be suitable for distinguishing milk fat from adulterated or non-dairy fat blends. They verified this by determining the fatty acid composition of 3 synthetic non-dairy fat blends, and obtained fatty acid compositions similar to milk fat, allowing for its misclassification as genuine milk fat.
The 1H-NMR method was revised by using 3 new NMR fat descriptors based on various integral ratios of signals associated with the CH2 moiety versus signals associated with butyric and n-3 fatty acids. This combined with chemometric analysis allowed the correct classification of both butter fat samples and simulated butter fat compositions, as well as, other vegetable (linseed, coconut, palm stearin, palm kernel oil, Lallemantia iberica seed oil) and animal fats (beef and sheep tallow).
Read the full open access paper
Comments
Hi Mark
I presume that this does not get over the most problematic aspect from a compositional viewpoint – i.e. the factor to be used when calculating the percentage of milkfat in a product. When I was still active the enforcement authorities tended towards using 3.6% butyric acid in 100% milk fat, the trade wanted to use a rather lower figure.
For me, the topic brought back “happy” memories as it was one of the first collaborative trials carried out under Ministry auspices. We had 40 laboratories which participated. For amusement, the results of that trial were interesting as it also proved analysts to be “honest” as known duplicates, blind duplicates and split level samples were all used in the trial.
But this was 40 years ago when finance for analytical work was much more readily available?
Roger
(See report:
The Determination of Milk Fat in Admixture with Cocoa Butter
David O Biltcliffe and Roger Wood
J Assoc. Publ Analysts, 1982, 20, 69-88)