Washing vs. Fermentation

[Matthew P. Williams]

I've often wondered about the interchangeability of the terms "washing" and "fermenting" in reference to wet-processing. Tom Owens' travelogue of Kenya piqued my interest again with the following:

The key difference here is the extraordinarily long fermentation time. And yet the resulting coffee (usually) has one of the brightest, cleanest cup profiles in the world!

In Coffee: Growing, Processing, Sustainable Production (don't have the book with me to cite the actual author and page number), the author explains that the agents involved in mucilage removal are pectase and pectinase, which are both present in the mucilage itself. This indicates that washing is an enzymatic, not microbial, process.

Of course, water is not always sterile. To what extent does a microbial process contribute to washing?

[Edwin Martinez]

......This indicates that washing is an enzymatic, not microbial, process.

you mean fermenting?

Of course, water is not always sterile. To what extent does a microbial process contribute to washing?

It doesn't.

The reason Tom finds further fermented kenyas are cleaner (as long as it is not OVER FERMENTED) is because this process is removing fruit, allowing the bean to express itself in it's purest form.

If you over ferment you have a SERIOUS defect and you can literally smell it at that stage and every stage that follows. If you under ferment, it is not as big a deal.. potentially not as clean and articulated. Most, obviously prefer to err on the under ferment side. If you've fermented as much as possible with out over fermenting the washing should be a breeze. Washing is just a physical and not a chemical process.

It is my opinion that knowing exactly how long to ferment is critical in separating good from great.

Once all fruit is broken down although the bean is still at full moisture and one with it's parchment, it has lost it's protective membrane (the fruit itself) so the fermented fruit begins to negatively effect the bean through the parchment.

When to stop fermenting has some similarities to wanting to drop your roast into the cooling bin RIGHT before you go into second crack. Once you've crossed the line, it is very clear you've done so. Ironically over fermented coffees can't be used on light roasts as they taste completely rotten. But roasted dark enough you can iron a lot of this out, and if it is not too bad it can pass as fruity.

[fleck]

i'd like to also point out that often times in kenya and other east african countries, the coffee is not always left in the tank with the same water that the mucilage was removed with. in many cases, the elongated "fermentation" time could be better referred to as a "post-ferment soak." after the initial fermentation and mucilage removal, coffees are sometimes rinsed then soaked with new water for a day+, which lowers your risk of the coffee becoming over-fermented, but can often times heighten acidity in a clean, crisp way.

--stephen

[Matthew P. Williams]

......This indicates that washing is an enzymatic, not microbial, process.

you mean fermenting?

I think this is where my confusion lies - it seems like the term fermentation has been loosely applied to the washing process.

Fermentation is an active process (energy is consumed), while enzymatic activity is passive. Fermentation is an anaerobic microbial conversion of carbohydrates to byproducts like acids or alcohols. Pectinase enzymes, on the other hand, catalyzes hydrolysis of polysaccharides in the mucilage.

There may be historical precedence for the misappropriation of "fermentation": In the same section of the Wintgens book Coffee: Growing, Processing, Sustainable Production it is mentioned that in the past, Saccharomyces (brewers yeast belongs to this genus) was sometimes added to washing tanks to expedite the washing process. Adding yeast eventually fell out of favor (I wonder why!!), but this may be the source of the residual misnomer.

Anyway - I'm not at all trying to say that fermentation doesn't occur, I am trying to understand the washing process better, and I have been confused by the interchangeable usage of "washing" and "fermentation". Knowing that enzymes are at play makes me scratch my head when I'm told that fruit is fermented off of the parchment.

Edwin, your post sheds a lot of light on the washing process - I am sure you are intimately involved with it. I'd like to paraphrase the process to see if I am on the right track.

After pulping, the beans are held in tanks, and clean water is added. The beans are soaked until enzymes free the mucilage from the seed. When all the mucilage is off the parchment, washing has reached a critical point: the seed in the parchment is no longer protected by the mucilage layer. If the beans are not washed immediately, they are susceptible to taints from fermentation that may occur in the washing slurry from yeast contaminants fed by mucilage.

Do I understand this correctly?

[James Hoffmann]

I looked into this a while ago for my blog, I never got round to writing part two. I still probably have post of the papers I referenced.

[Poul Mark]

Don't we also need to differentiate between mechanically washed and the wet process which uses fermentation to loosen the mucilage off of the bean, vs a wet mill which strips all or a portion of the mucilage off of the bean during the wet milling? I think so.

[Alistair Durie]

Write in from Peter Baker, CABI Europe - UK :

The discussion on the Coffeed Origins & Tasting Washing vs. Fermentation item is a bit off the mark.

Yes enzymes are involved, but they are produced mostly by bacteria (see attached). Some of these are inside the berry plant itself (endophytic) so even when the berries are sterilized, fermentation can still occur.

As pectins are fermented to simple sugars, yeasts begin to take over and if this goes too far then you get the ‘fermented’ taste problems. So it’s bacteria good; yeasts bad in this case, the opposite of fermentation in beer.

Best wishes

Peter Baker
CABI Europe - UK
http://www.cabi.org

[Matthew P. Williams]

This has been very illuminating! Thanks to everyone who has contributed. My own confusion was a lack of clarity and specifics on the sequence of events in the wet process.

[Peter G]

Sorry to weigh in late here, folks- one of my favorite discussions! A number of ideas have come up here:

On Nomenclature: Indeed, "fermentation" and "washing" are sometimes used interchangeably, which is a mistake. Fermentation is a stage in the traditional production of "washed" coffees. "Washing" is a separate stage of that same process. With the advent of mechanical demucilaging, "washed" coffees can now be produced without a fermentation stage (the demucilaging stage substitutes for the fermentation stage: they both separate the mucilage from the parchment). To further confuse, some use "fully washed" to differentiate between fermented and demucilaged coffees! I find myself saying using "washed" when only fermentation and washing have been used and "demucilaged" for coffees which have been demucilaged and washed.

On "Fermentation": the word fermentation is most commonly associated with beer and winemaking, when it is used to describe the conversion of sugars to alcohol by yeasts (saccharomyces cerevisiae). However, fermentation has come to encompass a variety of processes in food, when various microorganisms are allowed to act upon the food to create a desired result: lactobacillus to make yogurt and cheese, acetobacter to make vinegar, etc. So, fermentation doesn't include only yeasts, but a variety of microorganisms. Therefore, the best definition of fermentation might be something like "when microorganisms are allowed to act upon a food product to produce a desired result". So, does coffee fermentation fit this description?

In much of the literature, there is argument over whether the mucilage removal that happens naturally when depulped coffee is left to sit around is a microbiological or enzymatic process. There is no question that both enzymatic and microbiological processes are happening in that fermentation tank. In 1999, Avalone et al showed that pectins were still mostly complete after the mucilage has separated, indicating that osmotic pressure from bacterial consumption of sugars was responsible for the mucilage separation. In the paper Dr. Baker kindly posted, the same author proposes a slightly different mechanism. In any case, as Dr. Baker indicates, the consensus seems to be that microorganisms are very important in the process. In other words, I think most would agree that the characterization of the fermentation phase in coffee as a purely enzymatic process would be wrong.

Also, we should remember that the purely functional definition of the fermentation stage (separation of mucilage) is incomplete. Most will agree that byproducts (both enzymatic and biological) are created during this stage that can add flavors and complexity to coffee. If allowed to go too far, these byproducts can also damage the quality of the coffee. In any case, the story is more complex than simple mucilage removal.

Since microorganisms are very important here, and since both mucilage removal and flavor production are facilitated by these microorganisms, it seems that fermentation is an apt word to use to describe this process.

There's more to say, of course, but I'll save it for later.

Peter G

[Matthew P. Williams]

It's never too late for relevant contribution. This discussion has piqued a new interest for me.

I've traced the source of misinformation in the Wintgens book:

It should be noted that the term "fermentation" is not appropriate in the case of coffee because, unlike cocoa, no biochemical reaction takes place inside the coffee bean. It would be more correct to refer to mucilage removal by means of a biochemical reaction or hydrolysis of the mucilage which covers the parchment beans. This reaction is caused by enzymes (pectinases and pectase) which are naturally present in coffee cherries. pp. 639-640

Natural fermentation can be accelerated either by adding certain types of yeast (Saccharomyces) or with hot water. These methods were implemented before efficient mechanical demucilagers became available. At present they are rarely used commercially, although enzymes and hot water may still be used in isolated cases to increase fermentation capacity during peak periods. p. 641

Carlos H. J., Brando. "Harvesting and Green Coffee Processing." Coffee: Growing, Processing, Sustainable Production. Ed. Wintgens . Jean Nicolas. Wenheim, Germany: Wiley-VCH, 2004.

It's clear that this information is misleading. Indeed, fermentation within the coffee bean itself is not a goal of wet processing, but the author incorrectly limits the scope of the word "fermentation", ignoring microbial activity outside of the coffee bean.

In the paper submitted by Peter Baker, the authors suggest that enzymes created by pectolytic microorganisms are unable to depolymerize highly esterified pectins. Peter corroborates this in citing that pectins are mostly complete in the fermentation solution. Enzymes, both endogenous and bacterial, do play a role in the fermentation stage of wet processing, but other factors are more productive.

Peter, I'm looking forward to what you are saving for later!

[SL28ave]

I've traced the source of misinformation in the Wintgens book:

FWIW, it's the earlier chapters in that book (on diseases and nutrition, for example) that I find most valuable.

[R Miguel Meza]

I'm excited to see this being discussed. I think we tend to lump what we call washed coffees to much together as if it were only one process. in reality there are dozens of variations all of which make for a slightly differing cup. when using fermentation for fruit removal how long and how much water is added or not added certainly makes for differences in the cup. not using water tends to make for a brighter fruitier/wineyer cup though often with a thinner mouthfeel. and even many who use demucilaging machines may often not remove all the mucilage with these machines but only partially and then ferment away the rest for varying amounts of time. fermentation may also be done multiple times with rinsings in between. all these coffees could be 'washed' or rinsed after the fruit removal stage and clumped as washed coffees. and i suspect many coffees that are only semi-fermented/demucilaged (some mucilage still left on the parchment during drying) are being lumped with washed coffee as well.