Reconsidering the importance of interfacial properties in foam stability

In food industry, protein isolates are often used to help in the formation and stabilisation of food foams. Subsequently there is great interest in (1) understanding the effect of processing parameters on the functional properties of the isolate, and (2) methods and techniques that can help to predict the foam properties. This article describes the foaming properties of proteins that were modified in the Maillard reaction. From these relatively simple experiments results were obtained that indicate that for certain protein solutions the foam properties can vary significantly, while the interfacial properties are constant. Commercial protein isolates originate from only a few sources, mainly egg white and whey, and sometimes plant proteins (e.g. soy). Despite these limited sources a large variety of isolates with a wide range of properties is produced. One source of variation is the isolation procedure, but at least equally important are the conditions used before, during and after drying the protein solution to form the dry powder. From the literature it was found that one of the major changes to the protein during processing of the isolates is the covalent coupling of sugars via the Maillard reaction. To study the effects of these reactions, a model system was produced that consists of proteins that were glycated to different degrees using Maillard reaction. For each sample, interfacial properties (e.g. surface pressure, dilatational modulus) were determined, and foam experiments were performed. The results show that at constant concentration of both the protein (0.5 g/L) and sugar (0.7 g/L), the foam-ability and stability could be significantly improved (e.g. non-modified lysozyme does not foam, the highest modification is easily foamed and the foam has a half-life time of 200 s). Interestingly, the improved foam properties could not be related to any change in interfacial properties. While foam stability improved with increasing modification, the measured interfacial properties were not significantly affected. These observations seem to go against the general view that changes in foam behaviour should be reflected in changes in the interfacial properties. Additional experiments on thin liquid films were performed, where the disjoining isotherm was measured. These isotherms did not show significant differences in the interactions between the adsorbed layers. This indicates that the electrostatic and steric interactions between the adsorbed layers do not depend on the degree of modification. Only the thin film stability against rupture was found to increase with increasing modification. The thin film experiments lead to the hypothesis that aggregates (or oligomeric proteins) formed during modification might become trapped in the film. The presence of these oligomeric proteins could result in an increase of the apparent viscosity in these films, or in gelling or jamming of the liquid phase between the two interfaces. In other words, the observed behaviour is the result of the confined geometry of the thin films. The results confirm other observations that Maillard reactions improve foaming properties. Moreover, strong indications were found that to predict foam stability we need more than the traditional parameters (i.e. (dynamic) surface pressure, interfacial reology, and disjoining pressure).