Level: Apprentice

Reading Time: 7 minutes

I notice you advocate dropping your beans into a hot roaster.  I assume this is because you want Maillard reactions and Strecker degradation products.  Is there a certain temperature that works best for these products?

This isn’t the reason I suggest dropping your cocoa beans into a hot roaster.  I drop beans into a hot roaster so that my roasts are repeatable, so that I am hitting the ground running.  That said, and it really varies per roaster, I find a pre-heat temperature of 325-350 F works well in most cases.  Over 400 F tends to scorch the beans but that generally only holds true if there is a lot of thermal mass in the roaster.   The general trend is that the more mass there is in your roaster (like my Royal) the lower the temperature should be.  I don’t like it above 360 F, but the Gourmia is fine up to 450 F because it is so light.  This is the kind of thing I’ll be going over in the upcoming Roasting seminar.

With that out of the way let’s talk about those reactions.   There are tons of resources about Maillard reactions and I really don’t think I need to dive into the deep chemistry of them.   Generally speaking they are a class of reactions of amino acids and reducing sugars when heated.  More on that in a moment.

Here is a nice write up.

The over view is that they are entire class of chemical reactions that are responsible for many foods turning brown (hence they are also called browning reactions) and gaining an extra depth of flavor.  We are talking about browned meats, vegetables and breads.  And coffee.   What all these things have are the following:

1.       Amino acids (the building blocks of protein)

2.       Reducing sugars (this is a technical chemistry term but the take away is that glucose, fructose, lactose and maltose are reducing sugars.  Really any monosaccharide.  Sucrose (table sugar) is not since it is a disaccharide made up of a glucose and fructose molecule.

3.       Heat above 300 F for extended amounts of time.

Here is the thing.  Given those requirements I am not sure Maillard reactions happen to any great and noticeable extent in chocolate.   

Yes, there are proteins and amino acids. 

Unfortunately there is little data that there are reducing sugars.  Remember, cocoa is a fermented product.  Those free sugars are mostly if not fully consumed during that process.   I have found some studies that showed that there can be some residual sugars, but not very much.  Significantly less than is present in meats, grains and vegetables.

But for the sake of discussion, let’s assume there are enough there.  Now heat becomes the limiting factor and limiting it is.

We roast to a surface temperature of rarely more than 270 F.  Maillard reactions will form there but they take hours.  The 2-3 minutes the beans see a lot of heat just isn’t enough.  After that even though the surrounding  air may be hot, the surface of the beans are not and that surface and internal temperature is what counts.

I need to say thought that I’m unsure.  I’ve also seen papers about chocolate that have examined and listed compounds formed in Maillard reactions, but at rather low levels.  So some similar Maillard products do  form….or the compounds are forming by some other mechanism.   I’m unsure.

Being unsure though led me to wonder if I could test it because at the end of the day it is the results I care about.  Do Maillard reactions affect the taste of our chocolate and can we optimize them?

We know there are proteins in chocolate so that was my given.  Where it gets fuzzy is whether there are enough reducing sugars present and assuming there are, does a traditional roasting profile give a shift in flavor consistent with Maillard reactions.

To test that I took a non-fruity, pretty neutral flavored bean and introduced 5% maltose.  I did this by making a malt syrup and soaking the raw beans in it until they soaked it up and then dried them.


I roasted them as normal and made chocolate.  As a control I also roasted an untreated batch and made a batch of chocolate with an equivalent amount of maltose.

The roasting test beans certainly gave off very different aroma….but I can’t honestly say they were those yummy browning smells you get from meats and breads and long simmered stews.

The roasted beans did have a distinctly browner appearance but the resulting nibs really didn’t.  Maybe a little.  In both cases the maltose soaked cocoa is on the right.

The chocolate.  That is what counts.  The test chocolate had a distinct flavor that the control did not.  It was one that I’ve tasted before.  It is fullness at the back of the tongue and mouth.  And a light tang.  It took me a little but I finally recognized it as the flavor I get in certain milk chocolates. 

Upon a bit of reflection I find this makes sense.  Milk powder is full of proteins, amino acids and lactose.  During the long refining process it would appear Maillard reactions are occurring.  Some years ago I experimented with getting the temperature up to 170 F while refining and the flavor was more pronounced, correlating with heat being a component if applied long enough.

What this proves pretty well to me is that it takes the addition of a significant amount of reducing sugar to illicit a flavor change and that I’m not sure I actually liked the change.  It was not bad, but I can’t say it gave me that extra depth of flavor that well browned meats do. 

It is possible though that the maltose reacted in other ways that were not Maillard reactions and that the roasting temperature was not hot enough and that is why it didn’t give the same yummy outcome.  There really isn’t much of a way to tell except by analysis and to that end I don’t really see the point as the question was not ‘how do I get Maillard reactions?’ but instead did I roast a certain way to favor those reactions.

Again, the answer is no.

Strecker Degradationswere also brought up but I’m not sure it has much place in cocoa either.  Jim Stevens gives a good accounting in coffee.

The Strecker Degradation is an interaction of amino acids (AKA proteins) with a carbonyl compound in an environment with water, resulting in the creation of CO2 and an Aldehyde or Ketone. The later two components are important for volatile aromatics and flavors, and the Strecker Degradation contributes to browning. It involves compounds formed in the Maillard reaction and is therefore necessarily linked to it in coffee roasting.
The Strecker Degradation
This is one I thought I should include because I see it referenced quite a lot, in any text concerning roasting chemistry and is instrumental in the creation of the brown pigment as well as a myriad of volatile aromatics. It falls under the umbrella of, and requires compounds created by, the Maillard Reactions. Unfortunately this one isn’t so simple to explain.
It involves amino acids again, but instead of reacting with a sugar it is reacting with a molecule with two carbonyl groups, which look like this:
The compounds begin to trade parts of themselves, one of the carbonyl groups ending up on the amino acid (the acid having lost on carbon from its chain) creating an aldehyde as well as an amino ketone. This is an intermediary stage and reactions continue to create many different volatiles.

The key here is that in order to have Strecker degradation products you need Maillard products to start with and lacking those you don’t have anything to degrade.

 I admit it would be nice to have these reactions to play with and the huge body of work that coffee roasters have to work with in these reactions, but I remain unconvinced they play much of a role in chocolate making.