For this particular experiment, it’s important to keep in mind that the p:s ratio is correlated with lathering, not necessarily the explanation for better lathering.
The two recipes with cocoa butter also use 30-31% avocado oil, while the other recipes do not include avocado oil. I was purposely avoiding stearic dominant recipes for this round of testing and added the avocado as a more cost-effective way to increase the palmitic relative to adding more cocoa butter. I couldn’t add more RBO due to the linoleic + linoleic, which I wanted to keep at 8-9, and I was intentionally staying away from olive oil. At this point, I don’t know how important the shift to avocado is for the behavior of the soaps. After testing my way through two sets of soaps this fall, I noticed that for both sets, the balanced p:s soaps were just a wee bit smaller at the end. I didn’t keep track of weights or the time I used each soap, etc. but it does make me wonder. I’m not ready to rule out a potential role for unsaponifiables, or something else about avocado oil or cocoa butter in soap lathering behavior.
I’m also mindful of the role CO or similar may play in lathering. My series of soaps are relatively low CO (lauric + myristic FAs =13; CO
< 20%). In my somewhat limited experience with higher CO soaps, you can’t keep the bubbles down in a soap with 25-30% or more CO, perhaps regardless of the p:s. A “Holy Trinity” recipe made with 1/3 palm, 1/3 olive and 1/3 coconut has a p:s of 5.5(!) and longevity = 26. (This makes me want to go sorting through my archived soaps.)
Last but not least, I also think the 40-42% oleic is a key to the performance of these soaps. Back when I was formulating with more linoleic, I had to lower the oleic as a tradeoff to keep the longevity at 28 or 29, castor in the recipe, etc.. The soaps were making big bubbles, but the foamy component of the lather was weak. Then an experienced maker sent me a palm-based soap to try. The recipe is relatively low cleansing like the ones in this test, and close on the longevity, includes some cocoa butter, has a higher p:s than my palm + shea test soap and oleic above 40%. The soap makes beautiful lather, with all components, including the dense foamy layer. I also kept thinking about one of my favorite recipes of all times, which has a pathetic longevity number (20% palmitic and stearic), 13% lauric + myristic, 50% oleic, 3% ricinoleic and a p:s of 1.9. It makes a dense, silky, foamy layer and some larger bubbles. It’s not a soap for the super creamy lather lovers, but I love it. The recipe was formulated with the help of an experienced maker who is an avid fan of high oleic EVOO soap. Although I can’t definitely link the foamy component of lather to the oleic percentage, I do think it plays a role in the overall performance of recipes used for this test series.
This lather test was inspired by the love/hate relationship I’ve had with palm-based soap for my entire soap making career. It works works well at low percentages, but IMHO is hard and waxy when used at high percentages (and too fast to trace for my liking). A light bulb came on when I was using a new palm soap recipe in the shower one day and liking it. On closer inspection, the FA profiles almost matched my lard recipe.
The property/FA profiles of the recipes used for this test look remarkably similar to the ones favored by some of the more experience makers on the forum, as I learned in this
thread. I noted that
@DeeAnna is using higher oleic and no castor. I haven’t tried it yet, but it’s on my list of to dos.
For whatever reason(s), the palm + shea soap was not a star performer. Maybe that will change if I can get the bubbly lather up without compromising the creaminess. It’s likely that my next round of testing will focus on tweaking that recipe by adding a little more oleic vs. a little more CO vs. dual lye.
Formulating by the numbers works reasonably well, but it doesn’t explain everything. I still want to know which fats I’m using, even if I don’t know why it’s important. The size and shape of a soap and how it is used affect lathering. We can’t control water hardness or user preferences. To some degree, testing soap may be a soapy version of the
“Judgement of Paris” wine tasting experiment, especially when the soap recipes are more similar than not. When asking people to blind test soaps that represent minor recipe tweaks, it might make sense to include ”end members“ (very low CO, higher CO, 100% OO, etc.) that could be used to help standardize the results.
This post has been edited to improve clarity.
Bad idea! If you can, keep your beautiful soaps safe from DOS!
