Kueh Seri Muka

January 31st, 2021

What is seri muka?

To my mother’s slight chagrin, I do not speak any Malay. I therefore had to look up the literal meaning of kueh seri muka as “pretty face cake”, and thus I have broken my brief streak of preparing literally named foods for this website. Seri muka is a layered dessert composed of glutinous rice topped with a custard that stands out due to its green color. This coloring is due to the use of juice made from pandan leaves, which I have seen referred to in Southeast Asian cooking as somewhat analogous to how vanilla extract is used in western cooking.

At a group potluck dinner during my PhD to which I brought seri muka, my former advisor asked me whether the top layer was a yield-stress fluid. What we decided was that the topping belonged to the same group of slightly ambiguous materials that I discussed when making cheesecake. Though visually the topping may have seemed to be a yield-stress fluid, the microstructure was primarily a gel network that could not reform once it was broken. When making the dessert this time, I used a different recipe which at the very least amplifies this ambiguity, and I think pushes the topping over the line towards being an actual yield-stress fluid, as I’ll discuss.

Though my former advisor and I debated over the rheology of the top layer, if I recall correctly, we completely missed the novelty of the bottom layer. I believe the bottom layer of glutinous rice is unambiguously a yield-stress fluid of the cohesive granular material variety. Glutinous rice is also called sticky rice, and thus the name refers to its “gluey” nature rather than referring to gluten, of which it contains none. The stickiness of this material occurs due to the composition of the starch. Unlike other varieties of rice which I found more common in the US and which have a starch composed of a mixture of amylose and amylopectin, sticky rice starch is nearly all amylopectin which forms a gel in hot water. Cooked sticky rice can therefore be considered a yield-stress fluid in the same way as wet sand, with the grains of rice stuck together with gelled starch. Sticky rice is much more common in Southeast Asian dishes, but I also know it is used to make Japanese mochi and a few Chinese dishes like Zongzi. Something surprising I found while reading up on sticky rice is that it is apparently adhesive enough to have been incorporated as building mortar in part of the construction of the Great Wall of China.

Making the kueh

My understanding is that traditionally, kueh is prepared in an "approximate" way. Thus, recipes vary between people and without precise measurements of the ingredients; I’m not skilled or practiced enough to be able to do that, and so I simply merged various aspects of these two recipes, which differ in some key ways from my mother’s recipe that I can vaguely recall. I began by letting the glutinous rice soak for an hour before draining it. I then mixed it with coconut cream, water, and salt before steaming it along with 2 pandan leaves to impart some fragrance. While fluffing the rice after this initial steaming, I was struck by its interesting rheology that I had previously ignored. The material is clearly moldable, and seems to have a yield stress of several hundred Pascals, between that of peanut butter and chewing gum. Using a spatula, I compressed the rice into a flat layer as I continued to prepare the topping.

For the top layer, I blended and filtered pandan leaves to obtain a juice which would give the layer its eventual green color and mixed it with sugar, eggs, coconut cream, cornstarch, and flour. This mixture was cooked similarly to the preparation of an egg cream for tiramisu, constantly stirring it over simmering water. Similar to the tiramisu, this mixture gradually became much more viscous and developed a yield stress. Like with other egg creams, as the mixture cooked it is likely that the egg proteins formed microgel particles which were kept separate or broken down by the constant stirring.

Though I followed the recipe, I believe the top layer was overcooked. While it was exciting to me to have a yield-stress fluid to play around with, the yield stress was high enough to make pouring a flat layer difficult and also led to the trapping of significant air bubbles inside the material. I ended up smoothing the surface with a rubber spatula and popping the largest bubbles as they became apparent during the final steaming process.

In my recollection of my mother's recipe, the top layer was cooked very little and remained a viscous liquid prior to pouring it on top of the rice for steaming, thus making it easier to obtain a flat, bubble free layer. I believe this difference in preparation also accounts for the significantly different mouth texture of the final dessert that I can recall. The dessert I made this time had a distinctly "cakey" texture, dense and yielding in a somewhat ductile manner. This was likely partially due to the incorporation of flour and starch, but I believe it was also affected by the proportion of egg proteins that formed microgel particles versus a sample-spanning gel network. After steaming, the egg proteins that were not incorporated into microgels during the stirred cooking formed a percolated gel network that does not reform after yielding. Similar to my experiment with cheesecake, repeatedly breaking and molding the top layer material made it noticeably softer than the unyielded gel, but there was not a huge difference. Though there is a network that breaks and does not reform on yielding, the topping I made this time had a microstructure with a significant proportion of microgels that were largely unaffected by the yielding process. For this reason, I believe that while the topping this time could also belong to the ambiguous material category, it is much more reasonable and useful to categorize it as a yield-stress fluid.

In my mother's recipe, since the top layer was not cooked as much before pouring and steaming, far fewer of the egg proteins would be in the form of microgels. This gave that dessert more of the texture of a brittle gel which did not have a reformable microstructure in any way. Overall, I think this recipe went well, however I do think I prefer the top layer to be a bit more jelly-like rather than cakey. The cakey texture also made it more difficult to smoothly cut and remove the pieces at the end, which makes them not look as nice. As much as it will pain me, perhaps the next time I attempt this recipe I'll try to refrain from making a yield-stress fluid. Taste-wise, my wife really enjoyed this dish, saying that the sweetness of the top layer and the saltiness of the bottom layer are a good combination to make you want to never stop eating.