A celebratory triple-chocolate mousse cake

March 4th, 2020

What is the difference between a foam and a mousse?

A “mousse” is generally defined as a foam in which the continuous liquid phase has gelled or solidified in some way. The colloquial usage of the word, “gel” is not completely consistent with some of the technical definitions, and gel networks are their own completely distinct category of materials that I will discuss some other time. I will just say for now that there are materials called gels in which the microstructural network can break and reform on a relatively fast timescale, and others in which the network reformation or “healing” process will never occur spontaneously or will at least take an unreasonable amount of time for humans. The former category are yield-stress fluids and the latter—while often still rheologically interesting—are not. Unfortunately for me, the non-yield-stress fluid type of gel seem to be much more common in food and cooking, with a classic example being eggs. Above a certain temperature, the microscopic components of an egg will link to one another and form a solid gel network. If we take a cooked egg white and pull on it, it does not seem to exhibit flow but will instead break apart and won’t re-heal. This is the mechanism by which many mousses are made. For a basic mousse, a mixture containing eggs will be foamed, taking advantage of surface active components that stabilize the air bubbles; this foam may be incorporated in some other mixture which will then be heated, cooking the egg-based continuous phase which solidifies permanently. With these concepts in mind, we can start to think about what’s going on as we bake this mousse cake.

Making the cake

For this cake I made a few modifications but largely followed a recipe that I found here, but which appears to have originated here. As I mentioned at the beginning, this cake is constructed layer-by-layer in a spring-form pan which has removable sides.

The bottom layer is a fairly typical flourless chocolate cake, stiff and strong enough to support the layers above it without breaking. I created a foam of egg whites and brown sugar forming the well-known “soft peaks” that I’ve always found an overly vague qualifier. This heuristic test of whether peaks can form is determining the presence and magnitude of a yield stress and the stiffness of the material. If there are insufficient stable air bubbles packing together, the mixture has no yield stress and cannot hold the shape of a peak. The longer you foam the mixture, the higher the volume fraction of air that is incorporated. The air bubbles pack together more tightly and become smaller, this results in a higher yield stress and stiffness. Despite my fascination with them, foams have always been difficult for me to intuitively guess a yield-stress value since they have such a low density, but these foams likely have a yield stress between 1 and 3 Pascals and perhaps even less (see here for some context of yield-stress values). My guess for this recipe as to why soft peaks are preferable to stiff peaks is due to the fact that we are mixing the foam with a chocolate and egg-yolk mixture. Yield-stress fluids are notoriously difficult to homogeneously mix, and thus we want the foam to have a relatively low yield-stress as possible while still ensuring the bubbles are stable and suspended. At the end of the mixing process we have a mixture of chocolate particles, fat (from butter, chocolate, and egg yolk), sugar, and water as the continuous phase, with dispersed air bubbles stabilized by the egg white proteins. With such a complex mixture, it is difficult to know precisely what material structures might be occurring and contributing toward the yield stress, but it is highly unlikely that the air bubbles are concentrated enough to contribute meaningfully. Regardless of the structure, the resulting mixture has a fairly high yield stress and must be smoothed using a spatula to evenly fill the spring-form pan. I estimate it having a yield stress of the order of 100 Pa, a little less than that of peanut butter. After baking, we have the desired stiff chocolate cake that can support the next layers.

The middle layer is the first of the “mousse” layers. Instead of a foam of egg whites, we’re now whipping cream and sugar but we still use the heuristic of soft peak formation. Whipped cream has a fascinating and complicated microstructure, and exists somewhere between a foam and a weak gel network; I’ll give more attention to whipped cream in a later entry. Here we incorporate the whipped cream into a simple mixture of chocolate particles and fat in water. The apparent yield-stress fluid that we obtain for the middle layer is much softer than for the bottom layer, approximately around 50 or so Pascals, around double that of Nutella, but it still needs to be smoothed by spatula over the bottom layer in the pan.

The top layer is another whipped cream mousse but is much weaker and softer, thus we’ve ended up with an interesting gradient of mechanical properties through these layers, going from very stiff to very soft. In my cake, I also substituted in different types of chocolate to make a gradient of flavors. The bottom layer was very dark chocolate and the top layer was milk chocolate rather than white; the middle layer was half of each type. I highly doubt that people were able to perceive this when eating my cake, but they seemed to enjoy it. Regardless, the primary reason I substituted out the white chocolate was because of my wife’s distaste for it.

Getting back to the rheology, though both the middle and top layer contain the same amount of cream and very similar amounts of chocolate, the mechanical properties differ significantly due to the different processing steps and the proportion of air bubbles. The recipe calls for a significant proportion of the cream to remain un-whipped in the top layer, adding it in later and effectively reducing the volume fraction of air. The top layer clearly still has a yield stress, as can be seen from the rough shapes that are held in place at the top surface, however the yield stress is low enough that the material yield under its own weight. Looking at the profile image of the cake, this gradient in the yield stress is apparent. The outer edge between the bottom and middle layers contains gaps due to the fact that the middle layer did not yield under its own weight; the interface between the middle and top layers is completely intact. The top layer yields under its own weight; if it remained a pure foam rather than a mousse it would flow away as soon as the confining walls were removed. Thus, we gel the continuous phase of the chocolate foam by incorporating gelatin, significantly strengthening the top layer after it has been chilled. To finish off the cake I added some chocolate granules to the top surface that rest on top of the gelled mousse without sinking into it.