Surface Topography of Heat-Set Whey Protein Gels: Effects of Added Salt and Xanthan Gum - Food Colloids: Self-Assembly and Material Science - page 477

NaCl. Also, according to the classical friction theory, the friction force should
remain constant at various sliding speeds (if no acceleration is involved) and it
should be zero for zero surface load. However, the experimental data in
Figure 2 contradict these predictions: the friction force increases with increasing
sliding speed and there is a non-zero friction force at the extrapolated zero surface
load for both protein gels. This contradiction can be attributed to the existence of
surface water and the presence of adsorbing polymer chains. 2 For wet protein gels,
it has been shown 11 that contributions to surface friction can be attributed to three
different sources: surface roughness, surface hydrodynamic flow, and surface
adsorption of polymer chains. The contributions from hydrodynamic flow and
polymer chain adsorption can be clearly seen from the friction force and sliding
speed relationships in Figure 2(a). The contribution of polymer chain adsorption
increases with the sliding speed and reaches a maximum at some medium speed.
At very high sliding speeds, this contribution will become negligible because there
is not enough time for polymer chains to adsorb to the substrate. 2 In this case, the
friction force will show a linear dependence on the sliding speed. Therefore, in
theory, it is possible to estimate the thickness of surface water based on the speed-
dependence of surface friction when polymer chain adsorption becomes minimal.
Unfortunately, the maximum speed of our device was only 40 mm s 1 ; but we
have noticed that even at this speed the gel sample tends to deform because of the
high level of friction and its relatively weak mechanical strength. Examining the
friction force data in Figure 2(a), we do not see a linear relationship between force
and sliding speed. We conclude that the contribution from hydrodynamic flow is
still not large enough to dominate. But based on the fact that the salt-containing
gel has lower speed-dependence (Figure 2(a)), one can infer that this gel has a
thicker layer of surface water. This agrees with the load-dependence test results: a
rough surface would have more void spaces to hold water.
Figure 3 shows CLSM surface images for the two protein gels with (a) no salt
and (b) 200 mM NaCl. We can see that the former has a much smoother and
flatter surface. There is hardly any measurable surface irregularity for this gel.
In contrast, the salt-containing gel has a rough surface with numerous clearly
identifiable peaks and valleys. It is plausible to assume that the peaks are
composed of protein aggregates and the valleys are the void spaces containing
surface water. The CLSM images have been further examined for surface
roughness estimation. Figure 4 shows surface profiles of the two protein gels.
The surface roughness has been quantified in terms of the root-mean-square
roughness, R q . The gel with no salt addition has R q E

0.2 mm, but the gel

containing salt has R q E 2.4 mm, i.e., more than 10 times larger.

It is well known 11 that the presence of salt leads to a coarser microstructure
and weaker mechanical strength for heat-set WPI gels. Our observation of a
much rougher surface for the gel containing 0.2 M NaCl seems consistent with
existing observations of bulk microstructure. However, it is worth noting that the
peaks and valleys at the surface of the gel in Figure 3(b) appear to be oriented
towards the surface. While this directional preference for the protein aggregates
at the surface has not been observed in the bulk microstructure of whey protein
gels, 9 it is unclear yet what causes this surface anisotropy. One possible