Evaluation of PTR-MS analysis as rapid and non invasive tool for quality control in agroindustry: the effect of storage and packaging on anhydrous milk fat

M. Pedrotti, Iuliia Khomenko, Luca Cappellin, V. Fogliano, Franco Biasioli

Research output: Contribution to conferencePosterAcademic


1.IntroductionThe growing awareness of consumers for food quality posed new challenges to the quality control (QC) programs of agroindustry. The reliability in sensory and performance experiences of a product has been recognized as an important feature of product quality [1]. Thus, assuring the quality of products on a sensory basis is a fundamental corporate goal. However, sensory analysis, due to its high costs in terms of both time and resources, is not always feasible and adaptable to the high number of samples and fast decision times required by industry [2]. To answer to these needs, new rapid and non-invasive instrumental methods have been developed in the last years. Among the different approaches, analyzing the volatile organic compounds (VOCs) of the different matrixes seems the most promising since VOCs have a key-role in defining food flavor and aroma [3]. For long time gas chromatography (GC) has been the reference method but in the last years, new direct injection mass spectrometry methods have been developed in order to improve time resolution leading to high sensitive and faster analysis. Proton-transfer reaction mass spectrometry (PTR-MS) is one of these techniques. Based on proton transfer from a protonated reagent, usually H3O+, it implements efficiently a soft chemical ionization that makes this technique suitable for rapid characterization of food products [3]. PTR-MS and other techniques have been widely used in different researchers to investigate the different flavors and off-flavors originating from dairy ingredients leading to the identification of more than 230 VOCs [4-6]. The aim of this research was to evaluate PTR-MS as a rapid tool for the quality control of anhydrous milk fat (AMF). AMF usage in pastry, confectionery and ice-cream industry, is rapidly increasing due to its convenient features. The effect of different type of packaging – Bag-in-Box (BIB) and cardboard packages with a layer of plastic film (CT) – on the volatile profile and the quality of AMF has been investigated during a shelf life of 8 months at refrigerated storage (4°C). AMFs were valuated both during shelf life and accelerated shelf life (ASL) where AMF was exposed at 50°C for a total of 11 days. 2. ExperimentalThree production lots (0321, 0322 and 0323) from the same production facility were sampled in three different days. Each lot was then split in the two different type of packaging (BIB and CT). Fresh standards (STDs) were obtained for each time point from the same producer and stored in glass jars. When samples reached the maturity were stored at -20 °C until the day of measurement. On the first day of measurement (day 0) samples were melted in a thermal bath (50°C) and for each sample, five 2.5 mL aliquots were transferred into sampling vials and stored at 50°C till measurements. Samples were prepared for each day of measurement (usually day 0, 2, 4, 7, 9 and 11) following a factorial design for each shelf life point: 45, 120, 180 and 240 days. Empty vials were used as blanks.All the vials were incubated for equilibration at 50 °C for 30 min before being measured in an automated way by using a multipurpose GC automatic sampler (Gerstel GmbH, Mulheim am Ruhr, Germany) connected to the inlet of a commercial PTR-ToF-MS 8000 instrument (Ionicon Analytik GmbH, Innsbruck, Austria) in its standard configuration (V model) with flow rate of 35-40 sscm.. The ionization conditions in the drift tube were as follows: drift voltage 557 V, drift temperature 110 °C, drift pressure 2.30 mbar affording an E/N value of 141 Townsend (1 Td= 10–17 V-1 cm2 s-1) while the mass resolution (m/Δm) was at least 3800. Dead time correction, internal calibration of mass spectral data, peak extraction and concentration in ppbV calculation were performed according to the procedure described elsewhere [7-9]. Principal component analysis (PCA) was performed for data exploration while to determine significant changes in VOC profiles between different packaging types and ASL, one-way ANOVA was carried out by using RStudio software (Version 1.0.143) on the reduced dataset. 3. Results The final dataset on which analysis were performed was composed of 129 mass peaks that were found significantly higher than the blanks through one-way ANOVA (p < 0.01 with Bonferroni correction) in all the time points and after elimination of signals related to interfering ions (O2+, NO+ and water clusters) at m/z 30, 32, 37 and 55. PCA analysis performed on mean centered an scaled dataset of each time point showed a good repeatability of the technical replicates, minimal differences between the different lots and a clear separation of the different type of packaging (CT, BIB and STD). ASL was noticed to reduce these differences probably due to oxidation processes that occurs at 50°C: at the last day of analysis (11 days), especially BIB and CT samples were not well separated. At all the time points, significant differences were found in the aroma profile of the samples and about 40 VOCs were tentatively identified by literature research [4, 5, 10, 11] and GC-MS data. At all-time points the mass ions m/z = 73.064, m/z = 87.041 and m/z = 89.060 correspondent respectively to 2-butanone/butanal, to diacetyl and to butanoic acid three of the key aroma compound in butter [12, 13], showed high concentrations and significant differences between the different packaging types. Other important compounds that showed significant differences between BIB and CT packaging are the mass ion m/z = 101.096, correspondent to hexanal, one-key biomarker of fat oxidation [14] and the mass ion m/z = 115.112 tentatively identified as the 2-heptanone, found by Peterson and Reineccius as one key aroma compound of heated butter [15].4. ConclusionsIn this study, PTR-MS technique was tested as rapid and non-invasive method for evaluation of AMF volatile compounds. The analysis was successful in discriminating each type of packaging at the different storage times. Changing in the VOC profile were evaluated during an ASL. Major significant VOCs characterizing normal and heated AMF were assigned to sum formulas and tentatively identified. These VOCs belong to various chemical classes that are known to affect the profile of the final finished product and that are modified by oxidation processes. PTR-ToF-MS seems a promising tool for controlling the quality of this product and to be used, in combination with sensory and other instrumental analysis (i.e Rancimat, peroxide numbers) for building discriminative models to be implemented in the AMF quality control agroindustry programs.
Original languageEnglish
Publication statusPublished - 2017
Event5th MS Food Day - Bologna, Italy
Duration: 11 Oct 201713 Oct 2017


Conference5th MS Food Day
Internet address


Dive into the research topics of 'Evaluation of PTR-MS analysis as rapid and non invasive tool for quality control in agroindustry: the effect of storage and packaging on anhydrous milk fat'. Together they form a unique fingerprint.

Cite this