Update 30/3/16: More pics were added for visualizing espresso fluids under microscope. Link: Click here for the post


Since I have access to a lab microscope(Olympus fluorescence microscope with camera attachment) and happens to have my Pharos in the office, here are some pics and video for your amusement (inspired by the recent grinder/grind particle distribution discussion). I find it rather interesting to look at so am sharing them here.

As mentioned, the grind was done on the Pharos with dialed in espresso setting. Also, the image quality may be less than perfect. I could edit them but thought it'd be more interesting to present what I saw on the microscope in person.


Without dispersion - manually spread on the slide with a small spatula (and tapped the slides on the bench, hopefully to dislodge the clumps - don't think it works)
400x magnification


note: the horizontal line is just the edge of the glass microscope slide


Same position with different focus field depth:


You can almost see how easy it is to miss them if you're not focusing at the right depth.

Dispersed with ethanol:

400x magnification


Notice how big of a difference dispersion makes to the shape and the overall fines population! Instead of some fluffy particles, all the smaller particles are now mostly separated and you can see how sharp the edges are.

Here's where the fun begins - 1000x magnification with oil immersion (dispersed with ethanol as well)




And the grand finale!

Video @ 1000x magnification
The micro-fines are basically moving(I call it dancing with the flow ) due to the evaporation/capilary action of the ethanol between the slides & cover slip. You can almost relate this to the water flow in espresso flow, except that the flow rate in a puck is at least a few magnitudes greater!

Just to give some ideas how big the particles are in the video. The measurements are the long end length x width in micron. This is a frame taken from the video.



For references, a hole on a VST basket is roughly equal or less than 30 microns. And we're seeing particles here in the range of 1-20+microns.update: as pointed out by Vit, the 30 micron refers to the manufacturing tolerance, not the hole diameter itself. The average holes of VST is around 300 micron as measured by the HB team. How filter baskets affect espresso taste and barista technique


Also, from these, you can see that without dispersion, the micro-fines are almost *invisible* due to sticking to bigger particles. So the next time you're looking at the grind particles distribution graphs, remember to look out for where the micro-fines are (it's almost non-existant on the graphs I've come across). The volume % of micro-fines are probably insignificant, but I would contend that the effect is far from negligible if they do flow/migrate into the cup. Of course, the SOP that I was using might be biased to showing microfines (that's the intention anyway - to show that micro fines below 10 µm do exist).

Sam, this is amazing! Thanks for posting!
Video's and photo's of different grinders would be great to see.

I'd be happy to take samples from my EK and Terranova-M3 and send them to you. But I fear shipping would cause more fines?

Could you post a link to what you're using? And curious, wow much would 70x show? As we can make cheap iphone microscopes. Would that be useful for comparison?

Anyways, I need to head back to work, but looking forward to diving into these later.

Here is the link to the smartphone microscope:

Sidenote:
I've been wondering about using a $10 DIY iPhone microscope along with a cell phone camera and some sort of calibration scale to do optical partical distribution analysis. Think it would work?

Thanks Nick! It's mostly for my own selfish curiosity. I would love to try different grinder too, but don't have any other than conicals.

I'm using this Olympus BX51 (it's a full fledge lab microscope underutilized for coffee geeking ). http://www.olympus-ims.com/en/microscope/bx51p/

70x mght work to show the bigger particles, and really,the biggest problem you will have is clumping/sticking issue. Don't think imaging is good enough for accurate grind distribution analysis(especially when you have large spread of particle sizes). You would need to take thousands of images to have accurate assessment IMO. Additionally, the lower magnification you use, the lower resolution you have. If you compare my 400x magnification and 1000x magnification images, the 1 micron-ish particles are hardly visible. So a microscope would be good for observation(like this), but not for quantification.

nickw wrote: I'd be happy to take samples from my EK and Terranova-M3 and send them to you. But I fear shipping would cause more fines?

You could always ship him the grinders!

I like that!

samuellaw178 wrote:70x mght work to show the bigger particles, and really,the biggest problem you will have is clumping/sticking issue. Don't think imaging is good enough for accurate grind distribution analysis(especially when you have large spread of particle sizes). You would need to take thousands of images to have accurate assessment IMO. Additionally, the lower magnification you use, the lower resolution you have. If you compare my 400x magnification and 1000x magnification images, the 1 micron-ish particles are hardly visible. So a microscope would be good for observation(like this), but not for quantification.

That makes sense, both on the magnification needed for resolving the small particles, and thus the number of images needed to ensure reasonable accuracy. Which explains why we don't see optical particle size analysis at this scale.

I'd still be happy to send you some samples. We could try our luck with damage/degradation which could occur in shipping. I'm not sure how large or problem it would be; I know people do send in grind samples for laser particle size analysis.

First - Thanks! This is amazing!

samuellaw178 wrote:The volume % of micro-fines are probably insignificant, but I would contend that the effect is far from negligible if they do flow/migrate into the cup.

I think you are quite right there.

I got this thought train:
Could tapping (horizontal & vertical, like some people distribute) or WDT also get these microfines loose from the larger particles?
And if so, are there more to be found on the bottom of the filter basket?
And could it be that more of them end up in the cup if the previous two questions are answered with a 'yes'?

samuellaw178 wrote: For references, a hole on a VST basket is roughly equal or less than 30 microns. And we're seeing particles here in the range of 1-20+microns.

Are you sure about that? I have a cheap espresso machine, but I think that holes in baskets are similar - around 0.25 mm in diameter. The link you posted is about 0.03 mm tolerance in diameter

In filter of experimental pump moka pot device I'm using holes are even bigger, around 0.5 mm, but amount of fines seems to be pretty much the same as in "proper espresso" (low)

Btw. nice photos

nickw wrote: I'd still be happy to send you some samples.

That is a possibility, but the effort and cost might not be worth it. If interested enough, I could go around and get some sample grinds (cafe etc).

tohenk2 wrote:I got this thought train:
Could tapping (horizontal & vertical, like some people distribute) or WDT also get these microfines loose from the larger particles?
And if so, are there more to be found on the bottom of the filter basket?
And could it be that more of them end up in the cup if the previous two questions are answered with a 'yes'?

To question 1, I don't think so..at least from my observation. I tried to tap the slides hoping to dislodge the fines, but it doesn't seem to have an effect. Can't answer your 2nd and 3rd question either. But my guess would be the fines migration is more significant during the flow - as evident by pressure/flow profiling. Maybe that can be tested just by preparing two identical baskets (one tapped horizontally and vertically, the other no tap) and see if there's a consistent effect on the initial flow rate.

vit wrote:Are you sure about that?

I'm just quoting what the retailers are stating. One thing for sure is that in espresso/moka, the puck itself is also functioning as a porous filter. So that have reduced most of the grinds that would otherwise get in the cup or clog the filter basket holes.