Towards In-Line Detection of Milk Spoilage Using Nuclear Magnetic Resonance

Abstract

Milk is one of the most important foods in the world, consumed by over 6 billion people annually. Its nutrient rich content makes it a complete food source, however this also leads to that microorganisms can thrive inside the milk. This causes spoilage and the milk can no longer be sold to the consumers. This thesis investigates a new method to detect milk spoilage using NMR. The method would be able to be used as an in-line process and would increase the amount of tested products. Three different microorganisms were studied, Leuconostoc mesenteroides, Exiguobacterium undae, and Pseudomonas chlororaphis. For each microorganism NMR spectroscopy, relaxation times and diffusion were measured by three different instruments and magnetic field strengths, 300 MHz MRI, 60 MHz NMR and 20 MHz TD-NMR. The values were extracted by fitting signal models and the overall trends were observed. Packaging materials effects on the signal-to-noise ratio of the NMR signal and on the relaxation times were also studied at 20 MHz to understand the potential of using low-field NMR in industry. Diffusion experiments showed low potential for industry use due to lacking significant differences for treated and non-treated samples. For T2 clear overall trends could not be observed and microorganisms dependent trends were more clear. T2 showed significantly lower standard deviation than T1 making it interesting for future research. On the other hand, T1 showed the most overall clear trends with an increase over time indicating that the chemical environment is changing due to contamination. The complexity of milk spoilage made it difficult to determine which relaxation time was best indicator for spoilage, but it demonstrated how NMR signals are affected by milk spoilage. More research needs to be done to build on the findings of this thesis, specifically with more replicates and focusing on a specific instrument. Results from packaging material experiments demonstrated how it would not be the hindering factor for an in-line process, it showed that low-field NMR signals could be detectable through materials, even with an aluminium layer. The thesis provided a broader understanding of milk spoilage, its causes and how it affects relaxation times and spectra. All three microorganisms showed similar trends for all instruments suggesting that general spoilage mechanisms can be detected even if the underlying microorganisms differ. The results suggest that low-field NMR could be a viable method for milk spoilage detection.