Coffee Data Science

Glow in the Dark Espresso in Search of Fines Migration

Finding another way to look at the theory of fines migration

Published in

Towards Data Science

8 min readMay 21, 2021

This is the fourth piece I’ve written about the topic of fines migration. I have postulated with evidence that fines don’t migrate through a chalk experiment, an experiment on coffee swelling in size with moisture, and an experiment with spent coffee.

I wasn’t planning to redo the chalk experiment until my son came home with glow in the dark chalk. I thought this could make for a fun experiment combined with spent coffee. The glow in the dark component could help make the finer particles more visible if they had moved throughout the puck. The chalk dust itself is much less in size than 100um, so it is a great substitute for fines.

The aim was to put glow in the dark chalk dusk on top of spent, sifted coffee grounds (>400um in diameter). If fines migration occurs, it would definitely occur for large coffee particles with more room between particles for fines to migrate.

To review, the fines migration theory in espresso is that very fine particles (fines) migrate during a shot which can then potentially clog the filter or go into the cup. This migration is claimed to be one of the underlying causes of channeling.

To show how the glow in the dark chalk looks in the dark.

Chalk is not easily soluble. There are some studies showing that increasing temperature lowers solubility while increasing pressure decreases solubility, but there isn’t much in the espresso range.

Glow in the dark coloring is generally not water soluble either.

In both cases, drying out the grounds and the subsequent liquid would still allow us to see the glow in the dark chalk.

Data Collection

I started with 18g of spent coffee grounds that had been sifted, so most of the coffee grounds were greater than 400um in diameter. Then I added 0.62g of chalk on top, and I also added a metal mesh screen to reduce chalk from flowing back into my machine.

I made chalk dust, and I let it sit in the sun for a few hours.

I powered up the Kim Express, and I pulled the shot. It was a muddy disaster probably caused by extracting anything that was left to extract. I suspect this would quite possibly be the worst case for fines migration because of high flow. Originally, I wanted to run this test in the dark so I could see if the fines would glow if they did in fact migrate.

After some clean up, I went to work on analyzing the puck.

Puck Analysis

For sanity, I wanted to check if the top was still glow in the dark. So the dark image is in a room with no lighting or windows using a long exposure camera shot.

When I tried to get the whole puck out, only the very top came out. Perfect! the top chunk of the puck was exposed nicely, and I went to take a picture in the dark. Visually, there were no white speckles nor did anything show up in the dark.

I took to a close inspection of that top part, and I still found no white particles aside from the ones at the very top.

I then cut open that part, and I still found no white speckles indicating no chalk penetrated the puck.

As a sanity check, I popped out the other part of the puck, and I found no evidence of white speckles. In the dark, I found no evidence of glowing bits.

Sifted Analysis

I took the top part, and I sifted it using a 400um screen. There were some clumps of chalk that stayed on top, but a lot of fine chalk particles went below.

I look a microscope to a few particles just for fun.

Additionally, I used a toaster oven to dry out all the liquid. I looked at this in the dark as well, and I didn’t see any evidence of glow in the dark chalk. I also didnt’ see any white particles in the tray.

I scrapped off the residue, and there was 0.5g of residue which equates to 2.77% extraction yield. It is common in spent grounds to still have a little residual to extract.

Extra Dark Testing

I put the bottom part of the puck on the tray, and I searched for any white particles, but I didn’t find any.

I tried to use my kitchen lights to recharge the glow in the dark chalk, but I found a UV light, so I was able to recharge and go in the dark.

For the top part, the glow in the dark chalk still glowed, but for the bottom, spread out across the tray, there was nothing. I tried multiple distances of holding the light and moving it across, but nothing showed up.

Then I did the same for the residue, and nothing showed up.

I ran an experiment to try to find evidence of fines migration using glow in the dark chalk and coarse sifted spent coffee, but I found no evidence. I worked through multiple ways to detect evidence using visual inspection in regular light and UV light, but again, I found no evidence that the chalk penetrated the puck at all.

Conclusion: There is insufficient evidence that fines migrate.

It is quite possible I am wrong and conducted a poor experiment, but the experiment is simple enough that others could repeat it or even improve the protocol. I welcome such testing.

If you like, follow me on Twitter and YouTube where I post videos of espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn. You can also follow me on Medium.