Crema and the laws of physics

August 9th, 2008

It took me a while to notice that one of the prettiest things about espresso was a little bit confusing.

Watch this video and perhaps you will see what I mean:

In theory it doesn’t make sense. The bubbles contain CO2, which is much less dense than coffee so they should be rising quite quickly and not falling. What would make a bubble sink instead of float? You’ll see the same thing happening if you watch a pint of Guinness settle out – and it was their research that gave me the answer to this question.

What is happening is that plenty of bubbles of crema are rising, but they are rising predominantly in the middle of the shot glass and the pressure they create as they rapidly rise effectively sucks down the smaller bubbles close to the walls of the glass. There is nice explanation of Guinness here (with pictures!)

Another mystery solved…..

Comment Policy

When posting comments please use your name or your initials, please do not use your company name. Please don't link to your company name in the comment text either. This will be considered spam, and will be deleted. Opinion of any kind is welcome, but any personal attacks will also be deleted.

  • Pingback: Crema and the laws of physics

  • http://www.coffeegeek.com Mark

    Nice find, James. Brings another dimension to the “Guinness Effect” comment about settling crema. They look similar because they ARE similar.

  • http://www.relishwadebridge.co.uk hugo

    Far be it for me to disagree with the Guinness marketiers….I think they’re wrong and something more simple and logical is happening.
    I’ll try and explain, but I’ve had a beer so bear with me…
    1-Crema is made of bubbles of various sizes, all very, very small and on exiting the pf spout, in a ‘suspension’ of sorts in the coffee liquid.
    2-Coffee is a viscous liquid that drags bubbles with it and in the video liquid can be seen flowing down the glass edge, not because it’s being sucked down by bubbles rising in the centre of the glass (although this is happening) it’s merely moving relative to the bigger, less dense bubbles..
    3-All bubbles are lighter than the coffee therefore all want to rise above the coffee and do.
    4-A tiny bubbly is relatively heavy compared to a bigger bubble. The bubbles seen moving down are merely sinking relative to the bigger bubbles which can be seen as clumps of immobile crema. The important thing is it’s all relative. Tiny bubbles sink below bigger bubbles but all the bubbles stay above the liquid.

    As proof.. try stretching milk badly (unevenly), let it settle for a minute, then dig a spoon in. As you dig down the bubbles get smaller and finer and the layers are even from the centre to the edges. Just next to the fully liquid milk you’ll find the smoothest, silkiest foam.
    It got there by rising the slowest through the milk because it’s the heaviest relative to the bigger, faster rising bubbles.

    Clear as mud, maybe, but right.

  • Pingback: another world

  • Matt

    Hugo I completely agree with you, and was my theory before even reading your response. There is only so much espresso and the bubbles that are larger rise quicker too.

  • http://espressolab.ca Jimmy Oneschuk

    I’m not so sure the Guinness theory is correct either, or rather too simple to answer a much more complicated process that is espresso.

    While technically they are correct, there is drag on the side of the glass, the surface area of the glass relative to the overall volume of liquid in the cup shouldn’t be significant enough to create a change in the potential energy of the movement of bubbles which are contained in a lubricant. Espresso is a much heavier liquid than Guinness with more resistance for the movement of air molecules within the crema.

    The same is true of the smaller bubbles theory. Yes, smaller bubbles should fall if their overall mass is higher than the other components of the crema phase of the espresso. I believe the answer lies in the polyphasic nature of espresso and that liquid is being continually added to the espresso until brewing is completed. Adding liquid means heavy components of the liquid need to sink amid the rising components. Those heavy components might include mechanically unmixed oils and parts of the woody matrix of the bean which come to form the crema phase of the espresso. It is likely these components are influencing the motion. But to question the original premise, that air should always rise begets the definition of espresso as an emulsion of CO2, oils and brewing water. The specific gravity of the emulsified oils compared to the other components interacting in the crema theoretically should also influence the overall visual effect…

    not an answer I am satisfied with either, but when it comes to crema in espresso I think a simple answer cannot account for all the processes :)

  • http://www.relishwadebridge.co.uk hugo

    Erm… yeah. Matrix futtock orthodoxy theory me thinks…

  • Pingback: Crema de la Crema: a blog on espresso » Blog Archive » Great article on Crema

  • Richard Kirton

    @Hugo, And there’s me thinking fluid dynamics is a complicated subject!

  • Arthur Wynne

    so i was watching food network this evening and alton was talking about bernoullis principle,which i thought only involved how planes fly, low pressure-high pressure=lift….well he also has an equation that explains fluid dynamics on wiki so if you scroll down the page theres an explanation about viscosity,theres an in depth explanation of what hugo was talking about.theres a lot to read on the viscosity section.quite interesting.there are a load of links on liquids,gases and temperature which is all relative to espresso ,happy reading

  • Tomasz Nguyen

    Hi!
    on a slightly different note, when are you going to rid of you records? I’m after one, badly :)

  • Chris Weybright

    should we not also consider that many of the bubbles are in the process of bursting, and becoming part of the liquid portion of the shot/sinking to find their level?

  • Sarah Dooley

    That was simple and really cool. Thanks.

  • http://caffenation.blogspot.com rob berghmans

    i have to admit ; i miss your posts. i understand, but meanwhile i’m hoping to hear more about you in the future. always interesting.

  • Humberto

    I was thinking along the same lines as Chris there. Surely some of what we’re seeing is the “coffee” draining off the surface of the bubbles as well as their “liquefying”.

  • Ishida

    I think it’s not the bubbles that are sinking, It’s the liquid itself that’s sinking and the bubbles are rising, merging and acomodating…

  • Sebastian Storholm

    Neat, another (almost) useless fact I can pester lesser minds with! ^^

  • alex

    Hmm, I’m not convinced.  The bubbles act the same way only in NITROGEN draft beer–they’re unusually tiny and flow down, unlike carbonation bubbles which are larger and rise.  Most coffee containers are nitrogen flushed to preserve oils/flavor/prevent oxidation.  Maybe this nitrogen is getting into the espresso, causing the same effect.