Abstract
This research work was aimed to develop methods based on the fast, simple and sensitive analysis of volatile organic compounds (VOCs) for the monitoring and control of the quality of milk samples. At this purpose, a direct injection mass spectrometer based on proton transfer reaction (PTR-MS) was used. PTR-MS is a soft ionization technique that allows the direct measurement of VOCs. Typically, VOCs are firstly protonated with hydronium ion (H3O+ ) ions in a drift tube. Finally, the protonated ions (generally denoted as MH+ ) are detected by means of a mass spectrometer. Such analytical approach provides a characteristic VOCs fingerprint of each milk sample. The suitability of this technique was tested with a number of case studies. The first case study was based on the oxidation of milk samples induced by Cu++ ions (0-32 mg.L-1 ). During the storage of milk (4°C), the mass fragments m/z 55 and 83 evolved in intensity and proportionally to the content of Cu++ ions. The evolution of these fragments was explained with the production of hexanal due to the oxidation of milk fat. Among these two fragments, the m/z 83 was selected because it was less affected by interferences from other VOCs fragments. Accordingly, it was considered the main volatile biomarkers for the monitoring of milk oxidation. The overall analytical performance of PTR-MS showed that the signal of m/z 83 was linear (R2 = 0.999), highly sensitive (8.8 µg.m-3 per µM Hexanal in the sample), repeatable (rsd% < 6) with a very low detection limit (LOD = 0.5 µM). Such results supported the use of PTR-MS for the rapid screening of milk samples in terms of their susceptibility towards oxidation. Finally, the fast response, the simple analysis and the high sensitivity of the detector allows the use of PTR-MS also for the quality control at line or in line of the process. A second work was focused on to development of a rapid and non-invasive method for the quality control of raw milk was developed by using a multivariate control chart built with proton transfer reaction mass spectrometry (PTR-MS) data. For this purpose, raw milk samples were (collected by the South Tyrol Milk Federation in 2015) sampled from a pool of farmers affiliated to three distinct milk producers, located respectively in Meran, Brunick and Bolzano (all South Tyrol, Italy). For each milk sample, a representative VOCs fingerprint was collected, containing the intensity of about 180 mass fragments from m/z 21 to 200. Four mass fragments respectively, m/z 45, 59, 73 and 89, were selected and used to build a multivariate control chart based on the Hotelling T2 statistic. An upper control limits based on the Hotelling T2 statistic was also defined. Such control chart was then successfully applied to predict out-of-control samples because of (1) different thermal treatments, (2) different geographical origin or (3) different altitudes of the farms where the cows were grazed. The control chart discriminate all the “foreign samples” beyond the upper control limits as outliers. In conclusion, this work exemplifies the use of multivariate control charts with the data collected by PTRMS and provides evidence of its suitability for the quality control of raw milk samples. Finally, the third experiment aimed to understand the photooxidation reactions of sulfhydryl compounds in milk. Although photooxidation of milk is a well-known problem that is responsible for off-odours development, there is still some lack of knowledge on the initial mechanism governing such reaction. Especially, it is not clear with intermediate compound is formed at the beginning of the photooxidation reaction. To fulfil such gap, this work aimed to monitor the off–odour development in methionine model solution (75 mg.L-1 ) and pasteurized semi-skimmed (1.5% fat) milk by PTR-MS before and after their exposure to fluorescent light. Sulfur compounds were mainly contributing to early stage light activated flavour development. In particular, this work focus on the monitoring of off-flavor development during the oxidative degradation of methionine. PTR-MS revealed that during the early irradiation of methionine (< 100 min), methanethiol is formed following two paths, simultaneously. In one path, methanethiol can be formed via oxidative degradation of methional. However, in a second path, methanethiol can also be formed directly from methionine. As expected, such direct reaction was also faster. During early stage of photooxidation reaction, the simultaneous occurrence of both pathways was not reported before. It should be highlighted that such evidence was only possible due to the high sensitivity and fast response achievable by PTR-MS technique. Moreover, the findings of this research will be helpful to develop new strategies for the prevention of light induced off-flavours, such as in the case of extending the stability of milk during storage. Overall, the findings of this research project suggest that PTR-MS is a promising tool for the quality control of milk. The rapid analysis of VOCs has allowed to better understanding the kinetic mechanism of some relevant dairy reactions, such as the rancidity of milks induced by copper ions. Moreover, it was helpful to explain the occurrence of out-of-control milk samples because of the different geographical origin or altitude of the farmer. Finally, the superior capacity of PTRMS to analyse the VOCs evolved from milk in real-time was exploited to follow the initial stage of photooxidation reactions. When the rate of changes is fast and when the time resolving analytical power is a matter of concern, PTRMS technique offered its best performance.