Simulation of Retronasal Aroma Using a Modified Headspace Technique: Investigating the Effects of Saliva, Temperature, Shearing, and Oil on Flavor Release

Abstract

A device that simulated retronasal aroma was constructed from a 1 L blender incorporating purge-and-trap, synthetic saliva addition, temperature regulation to 37 °C, and blending at shear rates reported to occur during eating. Volatiles were collected on a silica trap, solvent desorbed, and quantitated by GC/FID or GC/MS with high precision (CV < 5%) and sensitivity (micrograms per liter). Increasing the temperature from 23 to 37 °C and adding shear increased volatility. The addition of synthetic saliva to a model grape beverage (pH 2.6) increased the pH and the volatility of the bases, 2-acetylpyridine, methyl anthranilate, o-aminoacetophenone, and 2-methoxy-3-methylpyrazine, relative to a model neutral compound, 1,8-cineole. The data were consistent with a sensory test that showed a significant shift in the perception of minty to nutty upon the addition of synthetic saliva to a mixture of 1,8-cineole and 2-acetylpyridine in an acid medium. The volatility of eight flavor compounds was investigated in a soybean oil versus water matrix. The volatilities of α-pinene (log P = 3.75), ethyl 2-methylbutyrate, 1,8-cineole, 2-methoxy-3-methylpyrazine, and methyl anthranilate decreased by factors of 8000, 130, 100, 7, and 3 upon oil addition ; however, butyric acid did not decrease, and polar maltol (log P = 0.02) actually increased.