Regular readers will be aware I do a little coffee roasting at home, the majority of which occurs with green coffee beans bought online. A couple of years ago I published a five-part series on the techniques involved with washed processing of coffee, taking a very small batch through the processing stages from cherry off the tree and into the cup.
Back in November last year I republished these in a single post, in the knowledge I would be running through a similar experiment using natural processing methods. Around the same time I also put together a short piece about the app Day One, and how I planned to use it to monitor my progress.
One of my main aims in writing here is to push myself along in learning a little more about the specific details of my interests — in this case, coffee processing. To that end, it will help to look at some background information before I get started.
Coffee processing — what are the options?
Although the following is not a detailed explanation of each, it should serve as a general indication of the differences between the various processing options available after coffee is harvested from the tree. The three more popular methods at the current time are natural (dry), washed (wet) and honey processing. I will also briefly describe the semi-washed or wet-hulled method, and a semi-dry method: pulped natural processing.
The information that follows is a put together from the following sources, all of which I’d recommend for more detailed information and accompanying images:
Cafe Imports – Coffee Processes
D.R. Wakefield – The difference between semi-washed and fully washed methods
Perfect Daily Grind – Indonesian Wet Hulled Coffee: Your One-Stop Guide
Origin Coffee Roasters – Black. Red. White. A Guide to Costa Rica’s Honey Process
Cafe Britt – Honey Processed Coffee: What makes it so special?
Seattle Coffeeworks – What on Earth is Honey Process?
James Hoffmann – The World Atlas of Coffee (Book)
The article on the Seattle Coffeeworks blog is particularly useful to view the “anatomy” of a coffee bean, and provides a little more detail in describing each of the layers, which might be helpful in gaining a better understanding of some of the terms described below.
For reference, the following schematic diagram names each layer in the entire coffee cherry:
- Centre cut
- Bean or seed (endosperm)
- Silver skin (epidermis)
- Parchment (hull, endocarp)
- Pectin layer (mucilage)
- Pulp or flesh (mesocarp)
- Outer skin (pericarp, exocarp)
Wet processing methods
Described in its simplest form, wet processed or washed coffee involves extracting and fermenting the bean from the flesh of the pulp, followed by drying, or more specifically:
- The external fruit pulp (exocarp and part of the mesocarp) is removed by mechanical de-pulping machines;
- The remaining mesocarp or “mucilage” is insoluble in water and clings to the parchment layer strongly, needing removal by fermentation and washing, or friction in mucilage removal machines;
- The beans may then be taken through a second washing to remove any remaining debris;
- Drying is then undertaken on patios or raised drying beds.
(Note: this method is what I have described in my five-part series of posts linked to above; the above stages are all completed by hand rather than machine when done on a small-scale)
Wet-hulled or semi-washed
Typically a method used in countries such as Indonesia, where humidity is high, and rainfall abundant. Wet-hulling (giling-basah) is a traditional processing method used in these countries, and was developed due to the particular climate and conditions:
- After harvesting, the pulp is removed, leaving the mucilage and parchment covering the bean;
- Beans are fermented in concrete tanks or plastic rice bags overnight (this process assist with breaking down the pectin in the mucilage for easier removal);
- The beans are then washed to remove the mucilage (or sold to a larger processing mill to continue the processing);
- After mucilage removal, remaining is the wet parchment, hence this coffee may also be called “wet parchment coffee”;
- Sun drying occurs for 2-3 days, reducing the moisture content to 20-24% (wet processed coffees are dried to a moisture content of 11-12%, rendering the parchment brittle for easier removal);
- The parchment is then removed using machinery specifically designed for wet parchment coffee (greater friction is required);
- Further drying occurs in the sun by day, with the coffee stored in bags overnight to continue the fermentation until the moisture content is reduced to the desired level.
An interesting point to note here is a point about wet-hulled green coffee being recognisable by its “dark-green and patchy colour” which may also be described as blue. I’ve certainly noticed this in the green coffee I have purchased which was processed in Sumatra, and it is very bluish in colour — quite noticeable compared with green coffee from other regions around the world.
I’d highly recommend reading the original article on Perfect Daily Grind, as it goes into considerable detail around the origins of why the processing is used, possible defects, and advice for getting the most out of these types of coffees (including roasting).
Semi-dry processing methods
Although not typically a category you often see listed individually, semi-dry processing does use significantly less water than the washed or semi-washed methods described above.
Pulped natural processing is apparently a common method used in, but not unique to Brazil — and you will certainly see this description on coffee originating from other countries. The actual stages are similar to the washed process described above, however the mucilage is removed with a pressure washing apparatus (taking care of steps 2 and 3 above), which removes the need for fermentation.
Without requiring large tanks and significant water for fermentation, the pulped natural method doesn’t quite fit in the wet processing category, however it is not fully dry either — therefore ending up in its own semi-dry category.
I must also point out that up until this point, I had not really looked closely at the subtle differences between pulped natural and honey methods of processing, however they are distinctly different, given the mucilage is removed in the former, however remains on in the latter (described further below).
Although consistency is apparently increased with pulped natural processing, as the risk of too much or too little fermentation is eliminated, thinking goes that this consistency possibly comes at the expense of flavour. Pulped natural coffees commonly tend to have lower levels of acidity in their flavour profile.
Dry processing methods
Natural processing is probably the most common of the dry methods, and involves leaving the coffee cherry intact, sun drying, then removing all outer layers from the bean:
- The coffee is harvested, and in some cases drying has already been allowed to commence on the tree;
- The pulp or flesh is left on, and the cherry is dried intact, with drying completed on raised beds or patios until the desired 11-12% moisture level is achieved;
- The entire hull (dried pulp and parchment layer) is then removed mechanically;
- The remaining seed or green coffee bean is then ready for export.
As you can see, this method is the simplest in the form of steps or intervention required, however I am certainly not suggesting it is necessarily easy to do well.
Honey processed coffees are a further subgroup of those classed as naturally processed, containing additional sub-groups based on specific criteria. This method originated and became popular in Costa Rica, progressively spreading to other countries in recent years.
The term honey is believed to have derived from the stickiness of the mucilage on the outside of the bean, which has a honey-like texture. Believe me – its sticky alright. Here is a short video demonstrating this to some degree, as producer Graciano Cruz of Los Lajones Estate talks about the honey processing method.
Farmers may separate their crop after harvesting, with the aim of developing different flavour profiles based on varying the amount of mucilage left on the bean and altering drying times. These discrete groups provide the various classes of honey processed coffees:
- White honey — all mucilage removed from the coffee seed; sun-dried uncovered
- Yellow (or golden) honey — approximately 50-75% mucilage removed; gains a yellow colour; dried in minimal shade over 7-10 days;
- Red honey — approximately 25-50% mucilage removed; longer drying time and perhaps combined with cloud/shade cover; dried over 14-21 days;
- Black honey — no mucilage removed; 100% of the sticky outer coating remains over the seed; covered to elongate drying time up to 30 days.
A point to note on the above list: most articles on honey processing will point to variations that occur from mill to mill or between individual producers. For example, I have seen descriptions indicating black honey processing involves drying the beans covered, with a relatively quick drying time — yet others suggest the drying time is longer. The percentages above are also approximations based on different articles I have read on the subject.
Why use dry or semi-dry processing methods?
For me personally, the answer is a simple: because I am interested and would like to see how things turn out. On a larger, professional scale, the reasons are obviously very different, and in a competitive marketplace, are a little more involved.
Perhaps to oversimplify things a little here, the answer seems to come down to two main aspects: flavour and environment (in which I include both climate, and the producer’s location and built environment — that is, access to technology and logistics around production etc).
When we talk of flavour (obviously quite important in the coffee game), Timothy Hill, of Counter Culture coffee summarises the flavour thinking in relation to natural processing:
With the natural processing, you’re letting the fruit and the nutrients from the skin and the sugars go into the seed through the drying process
In general, there seems to be the philosophy whereby natural coffees are a truer reflection of the flavour of the coffee cherry itself. This is not to say washed coffee holds none of these inherent characteristics, however from a consumer’s perspective, it is not often the two can be directly compared. In saying that, more producers seem to be showcasing different processing methods of similar or same harvest lots side by side.
As far as the other factors I have mentioned in relation to climate, technology, environment and overall capacity of the producer, Hill also goes on to note:
…the setup cost is extremely low—you don’t have to have a pulper, concrete, the machinery, electricity to run it — if you’re running a pretty large-sized farm. The labor and how to do it right is really tough. So it definitely is easy to take on; it’s really hard to do well.
And from Sweet Maria’s on climate suitability:
Natural coffees are perhaps the original method to process coffee. If the first coffee grown as commercial crop, for trading, was in Yemen, the climate is well-suited to dry processing. The western parts of Ethiopia such as Harar are traditional dry-processing areas. Older coffee-growing areas of Brazil as well as newer ones (Cerrado) have distinct seasons ideally suited to dry process method. If a farm can dry coffee from ripe cherry to hard, dried pod in 20 days, it is probably well-suited to DP (dry processing) methods.
Compared to washed processing, the natural method is clearly better suited to areas where water might be a scarcer resource, given its considerable utilisation in the various stages of the wet processing method. It then flows on, that drier (read less humid) climates would also be well suited to natural processing, which they appear to be.
Clearly in my humble backyard experiment, set up and ongoing costs are not a consideration at any scale, however with the coffee tree in my backyard now over a metre tall, there will be a time in future when processing considerations come into play. Although my time and effort required to process any coffee I produce will be a consideration, I suspect the local climate will perhaps dictate which method might be best overall.
Although I have called this series of posts a natural processing experiment – I use that term fairly casually. The plan is to outline in basic terms the methods I’ve used; data collected; recount how things turned out; and perhaps consider what I may do in future. What you won’t see are the usual scientific journal headings under which these components reside, nor a formal write-up with p-values, correlations, effect sizes or the like.
In simple terms, what you’ll read about is how I had a crack at it, followed by a tale of what happened.
Considering my natural processing options from those listed above, some thought was given to whether the methods used would be readily reproducible; easy enough to accomplish in terms of equipment, technique and time; and amenable to some type of tracking. Finally, it would be nice to utilise more than one type of processing and perhaps compare the two outcomes.
Of course there were some limitations in what I planned to do — climate for one. Many articles mention humidity as an “enemy” of natural processing, with Indonesian producers having adapted their methods to account for this, as I’ve described in the wet-hulled processing above. Although I have no real control over this, the drying period occurring in late spring provided some hope of finishing this phase before the full brunt of Brisbane summer humidity arrived.
Also, I have no real means of measuring the moisture content of the beans, with devices capable of this more expensive than what I’d be prepared to outlay at the current point in time. It is also beyond the scope of what I wanted to do, however would have provided decent points of reference for both data monitoring and as an indicator of when the process might be done. To that end, based on what I had read to date: “that seems long enough” would have to do.
In the end, I decided to try natural processing (entire cherry intact and sun-dried), along with a sample of the honey processing method, most closely aligned with what I have described as “black honey” above (pulp removed, with the entire mucilage remaining on the outside of the bean). In retrospect, the black honey method may not have been the wisest choice in a humid drying environment, however trying to remove some of the mucilage first and to what degree, did not easily satisfy my “ease of use” criteria — whereas removing the pulp and simply leaving them did.
So things were now underway. A small lot of coffee cherries halved into even smaller batches, one natural and the other honey processed. Hopefully I have done a reasonable job above in explaining the basic differences between methods to provide you with some understanding of each — no doubt many of you will already be fairly familiar, however I though it would be a good starting off point.
Future post(s) will report back on progress and ultimately the outcome of my own little experimentation into natural processing methods — hopefully you’ll check back in and see how things went.