A PTR-MS-Based Protocol for Simulating Bread Aroma During Mastication

The sampling parameters to simulate retronasal aroma during the mastication of white bread has been optimized using a retronasal aroma simulator (RAS) to compare the retronasal bread aroma with the conventional headspace aroma. The volatile composition in breath was compared with that in the effluent from a RAS using proton transfer reaction mass spectrometry, and the optimized RAS parameters were as follows: 2.5 g of bread sample, 250 mL of buffer, 1 L/min of N-2 gas stream, 350 rpm of rotating speed, and 38 A degrees C of water jacket temperature. The increased sensitivity and high reproducibility of RAS enabled detailed measurements of flavor release in the mouth during the mastication of bread. The simulated retronasal aroma was compared with the conventional headspace aromas by gas chromatography/olfactometry (GC/O), and the results demonstrated that the caramel note odorant of 2,5-dimethyl-4-hydroxy-3(2H)-furanone was found to show the highest contribution to the headspace aroma; however, it showed little contribution to the simulated retronasal aroma. These differences appeared to be caused by the volume of buffer added in RAS experiments. The odorant concentrations in the RAS effluent were found to decrease with the increase of the buffer volume, and the decreasing rates appeared to be associated with the chemical types of odorants. Flavor release of typical odorants in a RAS was measured at various ratios of buffer volume, and the results indicated that flavor release in mouth appeared to be influenced by the physicochemical properties of odorants. The results would help flavor chemists to make better prediction of bread aroma in mouth during the mastication.