Is first and second crack really an exothermic reaction? 

Hello and welcome to my possibly not so controversial opinion that first crack is in fact an endothermic reaction. By understanding this part fo roasting to be exothermic many people are basically sacrificing the entire roast profile for a sensor glitch or miss-understood concept.

Now let me explain myself. However here is my disclaimer, I am not a scientist and I have no real test as I am not a scientist. And since asking this question to a few people in the industry I have also heard that I am not alone in this question.

One thing of note, depending on the machine you roast on or the probes you use you might find a spike in temperature or a crash in temperature - airflow has a big impact on this as well, therefor more conductive roasters seems to show a spike in temperatures and more convective roasters seem to show a crash ** this really does show up differently on machine to machine **.

What I am going to do here is lay out my understanding of what is going on, give a little scientific backing on what I have found and give my conclusion. 

This would be a wonderful place to really start discussing and hopefully getting more data about what exactly we are doing when we roast because to me sometimes it feel like closing my eyes and running near the cliff edge hoping I remember the foot positions. 

So my argument is: First crack is an endothermic reaction and that to roast better we should understand that this might be the case and adapt our roasting to this new concept. The key point is to understand when does coffee go from endothermic to exothermic.

The overall theory that is taught is that first and second crack are exothermic. I believe that by second crack this is probably true but that at first crack the coffee is endothermic and still requires heat application.



An exothermic reaction is a chemical reaction that releases heat. It gives net energy to its surroundings. That is, the energy needed to initiate the reaction is less than the energy released.

eg: Those winter hand warmers that you crack and then get warm. 

Exothermic reactions can be strong - gunpowder - that are almost impossible to control with air or a cold environment. And there are weaker ones that can be stopped with an increase of cooler air.


A process which requires or absorbs energy from its surroundings, usually heat. It may be chemical or physical. 

eg: Ice melting.  

My backing to this argument. 

Vaporisation (water changing state into steam) is an endothermic reaction. If first crack is the release of steam trapped inside the bean then this means first crack cannot be exothermic.

If you have ever turned off the gas or just dropped the beans into the cooling tray you will find that basically the roast stops. Yes there is still the internal temperature inside the bean that makes some beans pop but this is heat retention rather than exothermic reactions.

** This is the contentious point. At what point does this mass of beans become self heating (exothermic). And how does airflow affect it.  

You see the bean itself is a mass of pores. These internal pores trap the steam/gasses but also act to insulate which makes the internal temperatures quite stable. If you open the dryer you also see temperature spikes (downward) but we do not freak out so much over these. So why is it that in first crack the spike (upward) has made us all so worried? 

I do not understand this. You see unless roasting on very small batches (100g or less) there seems to be enough thermal stability in the total mass of the beans to maintain temperatures. 

So why do we see a change in temperature? 

Well it is the water vapour that has built up under pressure. Pressure like in a pressure cooker increases the effectiveness of heat, still endothermic, and as such the released steam is HOT! This steam then hits the probe and….. presto. You have a temperature spike. 

Now just for fun it turns our condensation is exothermic. That sweet warm vapour touches something cold and returns to liquid, this liquid then encases that object and changes its temperature until both liquid and object equalise. 

What we need to understand are a couple of simple points:

  1. At what point does the coffee become exothermic?

  2. How does airflow affect the exothermic reaction? Is it that high airflow actually stops the beans becoming exothermic?

  3. If airflow can halt exothermic reactions should we be moving back to conduction based roasters?

This is my theory. We are not focusing on the mass of coffee we are roasting and thinking about that those first crack spikes is irrelevant so we should roast like they are not there. And perhaps we have miss-used the terms and labeled first crack as exothermic because of what the temperature probes are doing rather than thinking about what the coffee beans are actually doing - which in my opinion is actually endothermic.

The point of this article is to really try and get a discussion going. Let us work out what it is that we are doing when we roast and hopefully with better understanding we can once again raise that bar on roasted coffee.

Perhaps I have got this all wrong and misinterpreted all the books and theories out there but if this leads to a clarification about the science in the bean I am happy I wrote this.

Let me know what you think.