Effect of saliva dilution on the release of diacetyl and 2-heptanone from cream style dressings.

Aroma release from a cream style dressing, consisting of a thickening agent dispersed in the water phase of an oil in water (o/w) emulsion, has been studied by a purge-and-trap (PT) and a dynamic headspace mastication (DHM) model using two representative volatile compounds, viz. diacetyl and 2-heptanone. These isolations have been carried out from three systems: the dressing, the thickening agent dispersion and the o/w type of emulsion after adding different volumes of artificial saliva. Dilution of the samples with artificial saliva influences the amounts released for diacetyl and 2-heptanone differently: diacetyl decreases upon dilution of the thickening agent dispersion, emulsion and dressing. However, the amount of released 2-heptanone decreased only in the case of the thickening agent dispersion. These differences are caused by the distribution of diacetyl and 2-heptanone between the water and the oil phases. The distribution is not so important, when the DHM model is used for the release from dressings. The viscosity of the mixture of dressing and artificial saliva then plays an essential role. In general, the viscosity is considered to suppress the release of flavour. However, it has been found that the amount of volatile compounds released from the more viscous dressing was greater than from the emulsion. Most probably, the DHM model creates a large surface area by adhesion of the dressing on the wall of the sampling flask and the plunger head. This result suggests that the DHM equipment, which mimics the mouth movement, might be used to predict the real release of flavour in the mouth more precisely than other mouth models.