1. Home
  2. >
  3. process
  4. >
  5. How Low Can You...

Jun 13, 2015

How Low Can You Go?

One of my favorite parts of my job is teaching people to make small batches of chocolate in our Chocolate 201 class. It’s not just because it’s fun to make chocolate, but I really love geeking out about chocolate and answering people’s questions.

Every class, I get asked about what else you can put into the mini-melangers—vanilla, cocoa butter, types of sweeteners—and some about inclusions, but people most frequently ask: what is the lowest percentage chocolate possible? Since Dandelion Chocolate only uses cocoa beans and cane sugar, I decided to learn Dandelion-style by making a bunch of test batches to find out.

In class, the lowest percentage that we make is 65%—which describes how much of the bar is composed of ingredients made from cacao. A cacao bean is naturally about 50% fat (cocoa butter) and 50% solids (essentially cocoa powder). At Dandelion, we grind whole beans, but chocolate can also be made by separating the beans into its components and recombining those in different proportions. For example, our 70% chocolate bars are made from 70% cocoa nibs and 30% sugar by weight. A 70% chocolate bar could also be made from 50% cocoa nibs, 15% cocoa butter, 5% cocoa powder, and 30% sugar. The percentage on a bar typically refers to the percentage of ingredients that come from cacao. How much of each component exists, between cocoa butter and non-fat cocoa solids, doesn’t matter.

The limiting factor in making a low percentage dark chocolate bar is the amount of fat available for the solids to be suspended in. Chocolate is a “sol” (a solid suspended in liquid) which is similar to an emulsion (a liquid suspended in a liquid). Without enough fat to suspend the solid particles, the chocolate won’t work.

The lowest percentage of fat possible in chocolate according to the book The Science of Chocolate by Stephen T. Beckett is 25%.  Since I wanted to stick to two ingredients: cocoa beans and sugar, I did some math to figure out a few recipes to test this fact.

Recipe 1: 55% Madagascar cocoa nibs + 45% cane sugar (approx. 25% fat in batch)

Recipe 2: 50% Madagascar cocoa nibs + 50% cane sugar (approx. 23% fat in batch)


The trials are underway.

The first trial is to make a batch of chocolate which contains the “minimum” 25% cacao. The second is to make a batch of chocolate which contains less fat than the claimed minimum. I expected the second recipe not to work, and I suspected failure would either mean the machine would stop turning and the chocolate would seize up, or clumps of chocolate would be tossed around the room (it’s happened before in our R&D space).

Starting both of these batches was painful. Adding that much sugar meant that the chocolate got really thick and as the wheels in the melanger turned, they flung chocolate out of the machine, so I had to put lids on and let them mellow then add more sugar. The entire process of getting these started took significantly longer than normal, and at the end of the night, they looked like frosting and fudge and I was mostly concerned that they would be too thick for the machines to keep turning and that they might ultimately stop over night.

The next morning, I was relieved to find them still running, but they didn’t look quite ready, so I left them to refine for four more hours more than usual and they still didn’t look like normal chocolate.

IMG_3688               IMG_3697

Normally, chocolate flows, but with this amount of sugar, and not enough fat to hold it all, it seemed like wet sand. The following photos show what it looked like after I spread out the chocolate and after I shook it to get it to settle so I could add more. The resulting “chocolate” reminded me of playing with non-Newtonian fluids (e.g. “Oobleck,” or cornstarch and water) as a child.


The consistency was so thick it resembled “Oobleck.”

When I tasted them, the first thing that struck me was mouthfeel, and the gritty texture of sugar granules instead of the smooth, blissful mouthfeel of chocolate converting from solid to liquid. It felt more like eating Mexican drinking chocolate than a chocolate bar.

There was too much sugar to be suspended in the small amount of fat for both recipes, which made the entire process difficult, so I wouldn’t recommend either of these recipes to someone making chocolate at home. Unless, of course, you get a kick out of pushing the boundaries of what’s possible.

The Results:

In the end, it looks like I’ll have to do another round of recipes with a higher percentage of fat. The only way I could see a 55% cocoa batch of chocolate turning into something other than a gritty bar is by adding an additional ingredient like cocoa butter or soy lecithin, and that is a theory for another day. Even though we only use two ingredients in our chocolate and we’re quite happy with our recipes at Dandelion, it’s always interesting to explore how chocolate reacts to different things. We won’t be changing anything, but I’ll still be playing around with this idea in the meantime. If you’re curious about the math I used to work out the recipes, it’s all below.

How to calculate the percentage of fat in chocolate:

Last year, we sent a bar of 70% Madagascar to be tested and found out it had 32% fat. Other chocolate makers who use Madagascar beans will get slightly variable results because when beans are roasted in the husk, small amounts of cocoa butter migrate to the husk of the bean. Since sugar has no fat, all the fat is from the nibs:

.32 fat  x  1 bar   = .32/.7 = .457% fat/nibs

1 bar  =  .7 nibs

Our test batches are 1000 grams.  To make a batch with 25% fat, I needed 250 grams of cocoa butter (fat).  So 250/.457% is about 550g nibs and 450g sugar.

To try a lower percentage of fat, I had to decrease the nibs in the recipe.  50% sugar, 50% cacao which means 500g cocoa nibs * .457% fat/cocoa nib = 228.5g fat, out of a 1000g batch is about 23% fat. Since this is lower than the 25%, theoretically it shouldn’t work. In the end, it didn’t break the machines, but it turned out incredibly grainy and not at all like the smooth chocolate we enjoy eating. I imagine it would be impossible to temper too.


  1. Joel

    This was interesting to me and I wanted to try for myself. I’m not starting with nibs so I can control the exact amounts of cocoa butter to cocoa solids. I’m making 70% dark chocolate, and of that 70%, 40% is cocoa butter and 60% is cocoa powder. I’m also using coconut sugar because it’s low GI and really good. I usually go with a 55/45 butter/solids ratio, which flows nicely of course – but this is pretty thick in the melanger. We’ll see how it goes.

  2. Derek Pedersen

    Interesting. I wonder if the grittiness was due to the high percentage of solids to fats as you suggest? Or due to grinding time? It sounds like the grind was overnight + 4 hours. Assuming maybe 16 hours, this isn’t a lot of time. I usually consider 20hrs minimum with average for a small melanger to be around 24-30hrs for a good mouthfeel.

    I was hoping to learn if a very high solid % could be gritty even if you grind to <20 microns due to lack of fat. If you know the answer, I would love to hear it.


Leave a Reply

%d bloggers like this: