Coffee under the microscope

Inside Coffee Affair

There are many great cafés in London serving excellent coffee but inevitably a few stand out. One such café is Coffee Affair in Queenstown Road railway station which ‘inhabits’ a space that really encourages you to slow down and enjoy your coffee while just noticing the environment. An ex-ticket office that whispers its history through subtle signs on the parquet floor and in the fixings. The sort of place where you have to stop, look around and listen in order to fully appreciate it. And with a variety of great coffees on hand to sample, this is a café that is a pleasure to return to whenever I get the opportunity.

So it was that a few weeks ago, I happened to wander into Queenstown Road station and into Coffee Affair. That day, two coffees were on offer for V60s. One, an Ethiopian with hints of mango, peach and honey, the other, a Kenyan with tasting notes of blackcurrant and cassis. But there was an issue with them when they were prepared for V60s. The Ethiopian, “Gelana Abaya”, caused a considerable bloom but then tended to clog the filter cone if due care was not taken during the pour. The other, the Kenyan “Kamwangi AA”, did not degas so much in the initial bloom but instead was easier to prepare in the V60; there was not such a tendency to clog.

What could be going on?

So we had a look under the microscope at these two coffees. Each coffee was ground as if it was to be prepared in a V60 and then examined under the microscope. Was there any difference between the appearance of the Gelana compared to the Kamwangi? A first look didn’t reveal much. Magnifying both coffees at 5x, it could be said that the Kamwangi had more ‘irregular protrusions’ on the ground coffee compared to the smoother Gelana, but it was hard to see much more:

coffee under the microscope

The samples of ground coffee imaged under an optical microscope at 5x magnification. Kamwangi is on the left, Gelana on the right. “500 um” means 500 micrometers which is 0.5 mm.

So, the microscope was swapped to image the coffee in fluorescence mode. It was then that the cell structure of the coffee became clear. Here are the two coffees magnified 10x:

Fluorescence microscopy 10x, Ethiopian, Kenyan, Kamwangi, Gelana

Fluorescence microscope image of the two coffees at 10x magnification. Note the open structure in the Kamwangi and the more closed structure in the Gelana.

and at 20x

Kamwangi and Gelana coffee under the microscope

A fluorescence microscope image magnified 20x – not ‘um’ means micrometers (1/1000 of a mm), so the scale bar represents 1/10 mm.

So there is perhaps a clue in the cell structure. It seems as if the Kamwangi structure is more open, that somehow the cells in the Kamwangi break open as they are ground but the Gelana somehow keeps its cells more intact. Could this be why the Gelana blooms so much more?

Which naturally leads to a second experiment. What happens when you look at these two coffees in water under the microscope? Here the fluorescence images didn’t help as all you could see were the bubbles of gas in each coffee but the optical microscope images were of more interest.

optical microscope image in water

The two coffees compared under the microscope while in (cold) water. Magnfied 5x

‘Bits’ broke off the Kamwangi as soon as water was added but in comparison, there were far fewer bits of coffee breaking off the Gelana grains.

So what do you think has happened? If you remember our question was: when these two coffees were prepared with a V60, the Gelana bloomed a lot but then clogged in the filter (without extreme care while pouring the filter). Meanwhile the Kamwangi did not bloom so much but also did not clog the filter, what could be happening?

From the microscope images, it appears that

  1. Before adding any water, the cell structure in the Kamwangi is more open, the Gelana appears ‘closed’.
  2. When water is added, there are many more ‘bits’ that come off the Kamwangi whereas the Gelana does not show so much disintegration on the addition of water.

If pushed for a hypothesis, I wonder whether these two observations are linked. What is happening is that the cell structure in the Kamwangi is, for whatever reason, fairly fragile. So as soon as it is ground, the cells break up and a lot of the carbon dioxide is released. Consequently when water is added to it, the bits of broken cell quickly disperse through the water and it doesn’t seem to ‘bubble’ that much. In comparison, the Gelana cell structure is tougher and the cells only open up when water is added. I wonder if this means that the ground Gelana coffee will swell rather than break up and so ‘jam together’ as each grain tries to expand rather like trying to inflate many balloons in a bucket. They will push against each other and prevent water from easily percolating through the ground coffee.

Sadly, many more experiments would be required before we could see if there’s any truth in this hypothesis however that does provide a great excuse, were one needed, for many return trips to Coffee Affair. Meanwhile, what do you think? Do any of the images stand out to you and why? What do you think could be the cause of our V60 coffee mystery? I’d love to hear your thoughts so please let me know either here in the comments section (moderated and experiencing a lot of spam at the moment so please be patient), on Facebook or on Twitter.

2 Responses to Coffee under the microscope
  1. Fascinating article.

    Unfortunately, I think you are missing/overlooking another variable: how fresh the roast is. Coffee loses vapour over time, therefore a coffee roasted a few days ago will produce far more gas during blooming than a coffee that was roasted a few weeks ago.

    Of course, whether this degassing leads to the filter paper clogging up is another question…

    However, to correct for this variable, you’d need to revisit both coffees over a period of several weeks, noting the changes (if any) in behaviour. Looks like you could be in Coffee Affair for some time 🙂

    Thanks,
    Brian.

  2. beanthinking says:

    Hi Brian,
    Thanks for the comment.
    Indeed it is true that were the roast date very different, we’d expect different blooms and yet these two coffees had a very similar roast date (and were the same grind size). So I think we need to look for something else to explain the strange behaviour that Coffee Affair are noticing….

Leave a Reply

Your email address will not be published. Required fields are marked *

*

*