By Mike Ebert
In my early days of learning to roast coffee, someone I came to respect deeply stated his favorite coffee origin was Guatemala—not only because of the beautiful country and people, but the various coffee-growing regions and their diversity in the cup. He then went on to say it’s also because they roast the right way. Now, I may have missed his sarcastic smile as he said it, but I was puzzled, so I asked him why. He said because they roast perfectly; they move through the drying phase smoothly, hit yellowing all at once, proceed this way into cinnamon, have a first crack that’s tight and loud, control the momentum easily, and roast evenly almost every time.
He went on to explain he was being completely sarcastic, and in his opinion, no coffee roasts “correctly”—they all have their unique differences in how one might want to roast them. This started a conversation about which origins seemed most difficult to roast, as well as other elements that come into play, such as density and size differences within a particular lot (think Yemen), but we ended up agreeing it was processing methods that gave us the most “anxiety” during roasting.
In those days, parchment-dried (washed) coffees were the preference for most specialty roasters. Yes, we had fruit-dried (natural), pulp-dried from Brazil (pulped natural or honey-processed), and seed-dried (wet-hulled) from Indonesia, but for the most part, we were focused on parchment-dried coffees. There were many reasons for this—some valid, some not. My early mentors usually frowned upon any hint of fruit in the cup; when I cup today and taste any fruit notes, I have those same mentors sitting on my shoulder whispering in my ear to stay away. “It may cup well today, but in short order, it will become sour and fermented,” was their common refrain. At the time, this assessment was not entirely false, due in part to the lack of market access, as well as low demand for these coffees. However, I think the main reason was a lack of understanding, education, and scientific experimentation throughout the supply chain regarding these coffees.
This has changed dramatically. Not only have fruit-dried and pulp-dried coffees risen in prominence among specialty roasters, we are also now seeing all sorts of new approaches to coffee processing, such as carbonic maceration, anaerobic fermentation, double-washed, and fermentation using yeast additives. It is an exciting time in coffee, and the sensorial qualities of these coffees can be amazing. But for me personally, these coffees really mess with my mind when I roast them; they do not roast as evenly or predictably as the coffees I have roasted for decades. Even when I am prepared beforehand mentally, I scratch my head and wonder, should I be doing something else? Increasing air? Lowering energy? In conversations with other roasters, I began to ask about their approaches, looking for some clarity, and found not as much clarity as I would have liked, but some common thought processes and considerations. This article is meant to share what I learned, and what I am still learning.
Post-harvest processing types
TRADITIONAL PROCESSING METHODS
Let’s start with what I call the “traditional” methods with which most of us have experience. For starters, you may be confused by the terms I used earlier— fruit dried, parchment dried, etc. These are somewhat newer names that resulted from work conducted by World Coffee Research. Speaking with a diverse group of coffee professionals, agronomists, sensory professionals and other experts, I found that many who do not work in the coffee industry were confused by these newer terms. In addition, there is a lot of slang, country to country and even within country. It was in this meeting that someone stated, “We should name the processing methods by how the seed was dried.” Their point was that sensorial differences in coffees from the same lot yet processed differently are dependent on what state the seed was in when it was taken down to 12 percent moisture.
CURRENT TERMS FOR TRADITIONAL PROCESSING METHODS
FRUIT DRIED: The cherry is picked and laid out to dry to 12 percent moisture.
PARCHMENT DRIED: The fruit is pulped and the bean comes out with some pectin still on the parchment. It is then soaked in water for a period to remove the pectin, then it’s dried down to 12 percent moisture.
PULPED DRIED: The fruit is pulped and the bean comes out with some pectin still on the parchment, then it’s dried down to 12 percent moisture.
SEED DRIED: The fruit is pulped, dried to 20 to 30 percent moisture, and then hulled (parchment removed) and the seed alone is dried to 12 percent moisture
In general, the approach most of us take with parchment-dried (washed) and fruit-dried (natural) progress similarly:
For parchment-dried coffee, typically a higher charge temperature is used, with more aggressive energy throughout the roast. These coffees are relatively easy to “tame” in terms of getting the development time (time after first crack) that we desire.
For fruit-dried coffee, that same approach works, but with lower charge temperatures, less aggressive energy, and some care going into first crack, otherwise scorching can occur. In general, though, these coffees tend to go through color changes in a similar way.
When it comes to pulp-dried (pulped natural or honey-processed) coffee, things start to get more complicated. Color changes and first crack intensity are different. Most of the time, as I head into first crack, one voice in my head is saying “Do something!” and the other is saying “Do nothing!” For the first half of my career, I roasted very few of these coffees— except for pulp-dried Brazils, which tend to have more consistency of roast when compared to other origins.
“Pulp naturals/honey coffees can have distinct flavors that can easily be roasted out if they’re developed too deeply—typically, more so than full naturals,” says Paul Thornton of Thornton Family Coffee Roasters. “However, recently I’ve noticed some processors are leaving honey coffees in their cherry form after picking a bit too long, giving their term ‘honey’ an extension into pulpy flavors—they sometimes have way more fruit-forward flavors than they should have when they’re pulped too long after picking. For this purpose, I am referring to standard pulp-natural/honey-processed coffees that are pulped same day as picked.”
Thornton asserts that there is no sense in buying honey-processed coffees if they are going to be roasted darker than around 55 on the Agtron Gourmet scale (whole bean) because the subtle differences (when compared to fully washed coffees) can easily be roasted out. “Light-roasted honey coffees often will offer a sweeter flavor if they are processed and roasted with the intent to create and preserve this subtle flavor difference,” he says.
Thornton’s approach is to “start with high heat and as high of airflow as possible to reach first crack, embellishing acidity, dramatic heat slowdown immediately as first crack comes into play, while maintaining a high level of convection (airflow),” he explains. “Steady slope from here on out with an aim to get that sticky chaff that’s often difficult to get off the surface of light-roasted coffee bean. I generally will pull this roast as first crack tails off.”
Seed-dried coffees—or “wet-hulled,” as is found in Indonesia—do roast differently. Typically, they have higher moisture, but not as dense of a bean as one might expect. A general approach is to use high energy, like for a parchment dried, but slow things down as the color changes occur to have a gentler effect and more control heading into first crack.
To be clear, these approaches are not written in stone, nor do they create the absolute perfect roast; they are more of a general approach that I think many roasters would agree upon.
“In my experience, I always start my roasts based on the moisture content, screen size/variety, batch size and roasting room conditions—guided by the overall end roast level profile intent/strategy,” says Anne Cooper of Equilibrium Master Roasters. “I then turn my attention to dealing with the density and process through Maillard into first crack and to the end of the roast—never at the beginning of a roast.”
Yeast Fermentation
NEWER AND EXPERIMENTAL PROCESSING METHODS
All coffee processing methods involve both fermentation and drying. With the traditional methods, most of the sensorial differences result from the state the seed is in when it is dried, as described in the previous section. With the newer methods that have started to become more widely used in recent years, other factors, including the fermentation method, have a significant impact on the sensory qualities of the beans. In this section, we’ll take a look at some of these newer methods.
Carbonic. Maceration. This is a method of fermentation that involves placing whole cherries in a stainless-steel barrel and allowing them to sit in a carbon-dioxide-rich environment. Basically, the cherries ferment from the inside out. This was inspired by the wine industry and is said to bring about more aromatics, though it also results in a lower concentration of acetic acid, which some consider a drawback to this approach.
Anaerobic Fermentation. Oxygen is removed from the vessels in which the coffee cherries are fermented when the coffee is added at the beginning of the process, and valves on the tanks keep. them free from oxygen while also allowing CO2 to be released as it builds up during fermentation. Truth be told, this is not 100 percent new; it has been done on a small scale in some locations for years. The goal is to introduce different flavors in the coffee.
Double Washed. This is like parchment dried (washed), except after the initial fermentation in water, the water is removed and clean water is added to the tank for an additional fermentation period. This is the primary method used in Kenya. To be honest, it doesn’t really affect the roast profile, but I felt it should be mentioned. If anything, in my humble opinion, it makes the coffee roast very much like my old friend’s description of Guatemalan coffees—they roast “right.”
Cascara Fermentation. Once again, similar to parchment dried, except instead of soaking in water, the coffee is soaked in a cascara tea. In other words, the skin of the coffee fruit is dried and a tea is created using the dried skin, then the tea is used for fermentation instead of plain water.
Yeast Fermentation. This is also similar to parchment dried but with yeast added to the water during fermentation. Scott Labs, which has focused on yeasts for wine, is now developing yeasts specifically for coffee.
Lactic Processing. This processing method takes place under anaerobic conditions, with lactic acid bacteria cultures added to the fermentation liquid. The bacteria feed on the sugar contained in the pulp, which produces lactic acid. When complete, the beans are soaked in clean water to stop the growth of bacteria.
Wine Process. This method involves leaving the coffee cherries on the shrub for two weeks after maturity, allowing the fruit to over-ripen. Once the coffee fruit is harvested, it is sun-dried for several days. The result is a uniquely strong, fruity and fermented cup profile.
These are not the only newer methodologies being used or experimented with, and each has its variations. There hasn’t been much research into how these methods affect the chemical makeup of the coffee, but Figure 1 (above) shows the basic sugar change with parchment-, fruit- and pulp-dried processing methods. Notice the higher levels of fructose and glucose in the fruit- and pulp-dried coffees; both play a significant role in the beginning of the color changes as you enter into the Maillard reaction.
“If I had a natural Colombian, even the experimental processes as well, that had 11 percent moisture and 71 kg/hL density, I would still charge with a decent charge temp and heat application to deal with the initial moisture, and vice versa if it was low moisture,” says Cooper. “I would then work my way through my profile milestones and approach first crack with the appropriate rate of rise to then deal with the density and natural process and develop accordingly to the intended profile roast level. Generally, though, it would be a shorter development time so as not to over-roast or mute the process.”
NO “ RIGHT” WAY TO ROAST
One must remember there is no single way to roast any coffee. At some level, the quality of a roast is still the most subjective element in specialty coffee.
“Especially with these experimental processes, driven by the overall profile intention/strategy, I always encourage roasters to never forget the origin they are roasting first and start accordingly, then deal with the density and process toward the middle/end of the roast,” Cooper says. “And I find this greatly helps to alleviate any doubts or stress in regard to how to approach these processes.”
She also encourages roasters to get to know their machine and probe, so they can use the rate of rise (in degrees per minute) at key parts/milestones in the roast to help make good decisions about when to change or apply heat (along with other machine adjustments) as the specific bean reacts to heat application in line with an intended profile strategy.
“I’ve also found that key events like first crack tend to happen later or are much quieter on these experimental processes, hence it is even more important to be in touch with your probe and rate of rise to help manage heat application and the overall intended profile accordingly,” Cooper says.
As you can see, there’s some agreement and some differences, but overall, the same general theory; coffees processed using these newer methods will roast differently than coffees processed using traditional methods. Initial thermal energy: possibly gentler. Scorching can occur more easily, but it depends upon moisture. With all of the techniques mentioned in this article, the key is to focus on moisture and the overall cup you are looking for.
When approaching a coffee processed using one of the newer processing methods, I’d suggest focusing on these areas:
TURNING POINT: I always suggest keeping this the same, regardless of coffee type or even batch size. I like to have the same “starting point” of energy transfer on every roast, knowing I can adjust later to compensate.
RATE OF RISE: This should be focused on the overall cup profile you are looking for, with a focus on how you are heading into first crack. If you want a longer development time, you want to go into first crack slower or be more aggressive in lowering the energy input, and vice versa. Things go quickly with these newer fermentation methods.
CONTROL OF ROAST: Assure control of roast heading info first crack. Dropping the rate of rise might be tougher with newer methods than with traditional methods, and it might even be more difficult to hear first crack. You might need to focus more closely on temperature points.
END COLOR: The color of the coffee at the end of the roast can be misleading with these newer methods. It may look darker than it really is, and you can roast out any of the nuances of the processing method easily.
My best advice is to sample roast and get familiar with coffees that have been processed using methods that might be new to you. Do not make assumptions based on roast color, and taste everything!
MIKE EBERT is founder of Firedancer Coffee Consultants, dedicated to helping clients create a personalized strategy to ensure long-term success. He works primarily with roasters, producers and retailers, drawing from his extensive experience in all facets of the specialty coffee supply chain to guide clients in reaching their potential. Ebert is an authorized SCA trainer (AST) for the Specialty Coffee Association (SCA); a certified preventive control qualified individual (PCQI); and lead trainer for food safety. He co-chairs curriculum for the SCA Educational Advisory Council, which develops new educational experiences for the specialty coffee community.
