June 10, 2021

Coffee roasting & experimental processing methods


Washed, natural, pulped natural, semi-washed, honey… coffee producers around the world process their beans in a number of different ways. In recent years, with the rise of specialty coffee, variations on these techniques have started to appear, often capitalising on the fermentation that takes place during processing. These methods are often grouped together under the label of “experimental processing”.

Some of these techniques have been inspired by movements in other, unrelated industries. Others have emerged as farmers try to reach new markets. But no matter where they come from, once the beans are processed, dried, milled, and shipped, one key question remains: how do you roast these experimental coffees?

Most skilled specialty coffee roasters around the world will know how to roast coffee that has been processed using more “conventional” methods. However, when a new technique emerges, it can sometimes be challenging to profile it and truly bring the best out of the bean. 

To learn more, I spoke to roasters who had a good base of experience working with experimental processing methods. They told me more about how the processing affects the structure of the bean, and what roasters need to consider. Read on to find out what they said.

You might also like our article on how Colombian coffee producers are innovating with processing.

Experimental processing techniques: Bringing out the key flavours

Two of the most common experimental processing techniques in coffee today are anaerobic fermentation and carbonic maceration.

Anaerobic fermentation is when the coffee cherries are locked into an airtight fermentation tank after being harvested, to stop any oxygen from getting in. 

Carbonic maceration, a technique that originated in the wine industry, is similar. The cherries are locked in the airtight tank after harvest, just as with anaerobic fermentation, but are then flushed with CO2. 

In short, processing coffee in this way puts more of a focus on fermentation. Consequently, the beans absorb more sugars and acids from the surrounding cherry. This leads to the development of complex, unusual, and sometimes extreme flavour profiles.

In response to this, roasters have to create profiles for these beans which recognise and appreciate the complexity locked within the green coffee. This means being open to roasting them in different ways.

Yiannis Taloumis is the co-owner and CEO of TAF Coffee in Athens, Greece. He says: “The modern processing techniques have really brought out the complexity in coffee; experimentally processed beans are very rich in flavour. 

“But this is also the catch; if this complexity, this richness is not taken into consideration during roasting, it can lead to aggressive coffee flavours, and profiles that are not acceptable,” he adds. “This is why the roaster needs to experiment and find a balanced profile that will not be dominated by the intense fermentation of these coffees.”

It’s also important to make sure that the profile you opt for doesn’t completely overwhelm your customers. While people who enjoy trying new things might be open to these unusual flavours, they may be too much for many coffee consumers.

Sam Corra is a former Australian Barista Champion, a World Brewers Cup finalist, and the Head of Coffee at ONA Coffee. ONA has been a pioneer in roasting experimentally processed coffees over the years.

“Modern techniques allow for more expressive flavours and aromas for all coffee origins,” he says. “But with the complexity gained we also have found an introduction of boozy, savoury, or umami-driven flavour compounds that need to be managed in the roasting process. 

“My theory is that the heavier the fermentation, the more the beans ‘lose’ sugars for caramelisation as green coffee. This would mean they start to burn earlier, and show more savoury traits and drying tannin-like attributes.”

Things to consider when roasting experimentally processed coffees

When roasting experimental coffees, there are certain issues you need to address early in the roast to make sure you don’t end up with full batches of ruined coffee.

The first of these is heat transfer. Yiannis says there are some “dangers” they have identified when roasting experimentally processed beans.

“As the sugar concentration is higher, they absorb more heat, making it trickier to manage and shape the profile,” he tells me. “Really, you need to focus on different steps of the roasting process [to when you] roast traditionally processed beans.

“For example, the starting temperature point (charge temperature) is very important, because experimental beans can be scorched much more easily.”

Sam tells me that over the years, ONA has developed the ability to pick out certain data points when roasting experimentally processed coffees. This, he says, helps them understand what to expect during the roast.

“Firstly, there is a lag in heat transfer at the start of the roast,” he says. “We often notice a slower turning point and a slight lag in the momentum of the roast. If you’re not prepared for this, you can often see roast times that are far longer than your target duration.

“In the later stages of the roast, the momentum can also become difficult to control if you’re aiming for a constant decreasing curve. If it’s not monitored, and you do not reduce the gas early enough, you can easily lose control of the roast.”

Sam goes on to tell me that this means the rate of rise (RoR) falls at a “far slower rate” than intended.

“This will cause faster acceleration of the roast degree over the desired development percentage,” he says. “In the worst cases, it can even rebound into a positive climb that will effectively scorch all the pleasant sugars, leaving a harsh and hollow cup. 

“To avoid this, we make preventive adjustments to the heat at a far earlier stage in the roast, near first crack, which is much earlier than we would for a traditionally processed coffee,” Sam adds. “For experimental beans, first crack comes at a much later temperature than standard processed coffees, often about 2 to 3°C later. 

“This is good to note for roast planning, as it allows you to predict RoR and the temperatures you need to achieve the desired profile.”

How do the beans look after they are roasted?

As roasters know, quality control is important, and reviewing the colour of the beans after roasting is generally a good early indicator that a bean has been developed in the right way. So, what should you keep an eye out for?

“The biggest thing we noticed is how the roast colour and oils in the coffee seemed far darker in appearance than traditionally processed coffees taken to a similar end degree,” Sam tells me.

“You will also notice that over time, experimental lots seem a lot more oily in appearance the longer they are left off-roast compared to a coffee roasted similarly with a more traditional processing method.”

While oily coats on coffee beans are typically associated with darker roasts, the higher sugar and acid content caused by the experimental processing can be misleading. It may make you think that you’ve gone ‘too dark’.

As such, roasters should adjust their expectations accordingly when looking at these experimentally-processed coffees after the roast.

Other things to consider

There are several ways you can approach experimentally processed coffees when profiling them.

Anne Cooper is a consultant at Equilibrium Master Roasters in Melbourne, Australia. She says that she profiles in a completely different way, which helps her manage and observe how experimental beans behave during the roast.

“I believe many use process & density as a way to start a roast, which I don’t do,” she says. “My profile also has specific RoR targets at certain milestones along the way. So, if I get different feedback from a certain bean, then I can adjust and stay on track, and keep the energy in check to give the coffee every opportunity to shine. 

“I use a baseline profile first, learn what the new bean needs to stay on track to that profile, and then create the specific heat management protocol for that particular bean.”

By using this method, Anne says she can learn more about how each individual lot is processed, and adapt accordingly. This informs her roast management and heat strategy. 

Much like Sam, Anne also notes that first crack is different in terms of “timing and intensity” with experimental lots. She says that this “really challenges your senses and how to finish development at the end of the roast”.

Yiannis says that TAF’s approach to roasting experimentally processed coffees has also been helped by working with a different style of heat application.

“What works for me is radiation, rather than conduction or convection,” he says. “I think it allows much better control for the bean when absorbing heat; I find it very helpful with these beans, especially for those that have fermented for a long time, such as carbonic maceration or static cherry fermentation.”

To conclude, Sam offers one final piece of advice: “Be open-minded to change, and be ready to adapt to new approaches in the pursuit of the perfect expression,” he says. “Don’t be governed by previous notions of roast development, profile times, bean colour, and oil appearance.

“The best advice I can suggest is to let the taste do the talking. If it tastes good, roll with it, even if the approach to achieve that is less conventional.”   

Profiling these coffees is by no means easy, and many roasters will find that there isn’t the same base of knowledge in the coffee roasting community that there is for conventionally processed beans. Understanding how these beans are different may be a good first step. However, as Sam says, the key is to be open to anything. 

Be open to surprises, and prepare yourself for some fine-tuning to make sure you get that amazing, unusual profile you’re after in your roasted beans.

Enjoyed this? Then read our article on how roasters can drive change at origin.

Photo credits: TAF, Josef Mott

Perfect Daily Grind

Want to read more articles like this? Sign up for our newsletter!