The Column - June 2008 - (Page 8) Bonnefille The Column www.thecolumn.eu.com June 2008 sensory panel to discriminate good (three batches: G_S001 to G_S004) and bad ones (four batches: B_D01 to B_D10). The set of good and bad samples allows qualitative models to be built with the FOX E-Nose. Finally, the E-Nose was used to analyse and determine the quality of three blind samples (U_001 to U_003). For routine analysis purposes, statistical quality control can be used (Figure 3). To establish the range of acceptable grade (blue band), all good samples were taken as the reference quality. All future samples analysed that are plotted in this band will be recognized as good and all samples outside will be rejected as out of specifications. Using this quality control card, the three blind samples were assessed as bad. To follow-up onion aroma quality over time and monitor their shelf-life, samples were tested at different times. Samples of “good” onion were analysed when freshly Figure 3: Quality control chart of onion samples. Out of specifications Distance (Odour unit) 0.01 0.00 G_S01 G_S01 G_S01 G_S02 G_S02 G_S02 G_S03 G_S03 G_S03 B_S07 B_S07 B_S07 collected (day 0 D_00), then at various intervals of time (day 1, 3, 4, 7, 8 and 9: D_01 to D_09) by the E-Nose and sensory panel to determine good and bad products. To follow-up onion quality over time, a principal component analysis was built up (Figure 4). At day 4, the quality was still good whereas at day 7 it turned bad. Intermediate analysis was required to precisely determine the deadline for safe consumption. Quality Control of Smell/Taste of Food Grade Polyolefin Resins Plastic pellets aimed at producing food packaging (caps, liners, closures, bottles, water pipes etc.) need to be tested for possible residual solvents or monomers that could be responsible for a “plastic smell” in finished products and could transfer an unpleasant flavour to food products. Currently, these tests are performed by expert sensory panels, which require days for sample preparation and panelists! training. A FOX Electronic Nose instrument was used to test the aroma/flavour conformity of polyethylene (PE) pellets in correlation with sensory panels measurements. The sensory panel ranked the PE samples based on a German standard using the scale shown in Table 1. The nnb score corresponded to good quality; 1–2 and 2 were medium and 2–4 and 4 were bad grades. The Electronic Nose was first trained by analysing PE samples previously evaluated by a sensory panel. This phase allowed a quality control model to be set up showing the various bands of quality on a map (Figure 5). In routine analysis at the production floor, more than 400 batches of three PE types were analysed over a four and a half month period. All samples considered as good by the analyser were also scored as good by the panel (no false positive). The Electronic Nose rates were always more critical, with 5% of the Table 1: Scale used to rank polyethylene samples. Odour evaluation descriptors Sensory score defined by the standard nnb 1–2 2 2–4 4 Good quality 0.01 0.02 0.03 2 4 6 Out of specifications B_S10 B_S10 U_001 B_S08 B_S10 B_S08 U_001 B_S09 U_001 B_S08 U_001 B_S09 B_S09 8 10 12 14 16 18 20 22 24 Figure 4: Principal component analysis of fresh and aged onions. Discrimination index 94 0.6 0.5 0.4 PC2 - 2,281% 0.3 0.2 0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.5 0.4 0.3 0.2 0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 D_08 D_09 D_07 D_04 D_03 D_01 D_00 Ageing evolution Bad onions Good onions No odour/flavour Slight odour Medium odour Strong odour Very strong odour PC1 - 96,893% 8 http://www.thecolumn.eu.com
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.