Computer Simulation of the Pre-Heating, Gelation, and Rheology of Acid Skim Milk Systems - Food Colloids: Self-Assembly and Material Science - page 286

The amount of free whey protein in our simulation, however, is quite high: real
samples typically have 30% free aggregates and 70% coating. 15 To obtain such
figures, longer complexation runs would be needed. More experimental meas-
urements could directly investigate whether indeed the interaction between the
whey proteins and the casein is consistent with our parameter choice.
Higher pH leads to a shorter gelation time, a higher gelation pH, and stiffer
gels. 14 The shorter gelation time can be explained by the denaturation. Even if
the probability of complexation is low compared with the binding probability
of the caseins, the rapid diffusion will lead to faster gelation. The denaturated
b-lactoglobulin in solution is mainly responsible for this. 14 We have found the
same effect for all the simulated systems, as they all had about the same amount
of whey protein in solution. The complexation appears less relevant, as dena-
tured b-lactoglobulin added to casein has the same effect as pre-heating. 15,16
That effect was also found in our reference sample. Disulfide bonds, formed at
ambient temperatures and under acidic conditions, play an important role
during gel formation. 17 For the rheology the complexation did not appear to
have much of an effect in this simulation study, the effects on the gelation
kinetics being stronger. Systems corresponding to low pH were found to gel
more rapidly. The gelation kinetics for these bidisperse systems is quite different
from cluster aggregation in monodisperse models 7 due to the slow diffusion
rate of the casein micelles, which act as spacers in the final gel structure.
Acid gelation of pre-heated milk has been suggested 18 to be a two-step
process: first the whey proteins form a gel, and later maybe the caseins
coagulate. From our results this explanation seems unlikely, as the coated
casein micelles in the simulation are trapped in the whey protein network, in
agreement with observations from confocal scanning laser microscopy. 15 Only
substantial rearrangement of that network would result in casein micelle
gelation. Because of the timescales involved such rearrangements cannot be
investigated using the present model, though rearrangements at the level of
the whey proteins are included. We have used a similar model 19,20 with flexible
bonds in earlier studies of rennet-induced casein gels, containing only the large
particles. Simplified models would have to be developed in order to study
ageing in bidisperse models with a large size difference.
Acknowledgement
The work was financially sponsored by Unilever Research and Development,
Vlaardingen, the Netherlands.
References
1. P. Walstra and R. Jenness, Dairy Chemistry and Physics, Wiley, New York,
1984.
2. A.J.R. Law, D.S. Horne, J.M. Banks and J. Leaver, Milchwissenschaft,
1984, 49, 125.