Blue Steam

As a photographer I see my camera as a tool with which to capture and create. As an engineer, I see my camera as a wonderful machine and opto-electric toy. But sometimes a situation arises that makes a camera into a scientific tool. To make visible some hidden property of reality itself.

After this poetic introduction, it might seem a little mundane to tell you that this post is about the steam the comes out of my coffee machine. And in fact you don’t even need a camera – this is something that you can see with your naked eye. We own a Delonghi Icona home espresso machine, which looks like this:
DeLonghi_IconaLike most machines of this type it has a milk frother side-arm (on the right), that can be used for making cappuccinos. It works by forcing hot steam out at high velocity. The steam’s velocity is controlled by a valve which one opens by turning the round black knob on top.

Playing around with it one day, I was surprised to see a strange thing: as one opens the valve more, the apparent colour of the steam exiting the machine changes colour from neutral gray towards blueish-gray. Video follows:

To see it more clearly, I put a white LED headlamp behind the nozzle, shining towards to steam. Firstly, with the espresso machine’s steam knob turned partly open, the steam appears gray-white:

Partially opened, the espresso's milk frother produced white steam

Partially opened, the espresso’s milk frother produced white steam

This is not strange, as steam is normally neutral in colour due to something called Mie scattering. Mie scattering happens when light interacts with little spherical droplets that are larger (by, say, a factor of 10) than the wavelength of light. This is the same effect you see in clouds, and as it is not colour-dependent one sees clouds to be grey or white.


Mie Scattering (Image by Thomson Higher Education)

Mie Scattering (Image by Thomson Higher Education)

But now, look what happened when I opened the steam knob further:

Opening the knob further, the steam noticeably changes in colour, turning blue. What happened?

Opening the knob further, the steam noticeably changes in colour, turning blue. What happened?

Something strange was happening here. Remember this is still only water (steam) coming out of that nozzle. How can it now be a different colour? This blueness was not only a subjective illusion – the RGB values clearly confirm it – look at the third value (blue) relative to the two other (red and green).

Left: RGB = (205,204,209) Right: RGB = (116,134,210)

Left: RGB = (163, 157, 169)
Right: RGB = (116, 134, 210)

Before you think this is an issue of white balance, take note that I manually set my Nikon D600 to have identical white balance in both photographs, calibrated in Adobe Lightroom to match the white LED light, visible in the background.

Now let’s go on to the science of it. The first explanation that I could think of was that this had to have something to do with the size of the steam particles. Having studied physics, I remembered the sky being blue because of something called Rayleigh scattering, which occurs when blue light is selectively scattered more than other colours off molecules in the air.


Rayleigh Scattering.( Illustration by Rich Watkins and Joe Vallandingham, Rowan University)

If the steam particles were *really* small, they may behave in the same way that the Earth’s atmosphere does. Rayleigh scattering occurs when the particles with which light interacts (the steam, in this case) are much smaller than the wavelength of light; typically less than 1/10 of the wavelength. However herein lies the problem:

  • Visible light has a wavelength spanning the range of 0.4 μm (blue) to 0.7 μm (red).
  • Mist and clouds contain relatively large water droplets (1 μm – 1000 μm, mean 15-20 μm). This does not cause Rayleigh scattering, but colour-neutral Mie scattering instead.
  • Steam that is close to evaporation can have much smaller droplet sizes (0.2 μm – 1 μm).
    However even this is still too large for Rayleigh scattering to occur, as 1/10 of the visible spectrum’s wavelength is only about 0.05 μm.
  • The effect of Rayleigh scattering is typically quite weak. One sees the blue sky because of the huge amount of air that comprises the atmosphere, but won’t see the effect in a small volume of air. But our blue steam is rather spectacularly intense.

Only after googling this a bit did I come across the something called the Tyndall Effect. In principle the Tyndall effect is quite similar to Rayleigh scattering, but occurs for larger particles, i.e. particles that are approximately the same size as the light’s wavelength. Much like Rayleigh scattering, the Tyndall effect preferentially scatters blue light. However, the effect is actually stronger than with Rayleigh scattering, and mathematically somewhat different.

Of course this effect can only work if there is some blue-wavelength light contained in the white light of the source. A friend asked me if LED light satisfies this requirement, as it isn’t a natural continuous source like the sun. As it turns out, white LED light contains a large blue peak in its spectrum, which probably explains why the blue came out so strongly when I used my headlamp:

(Diagram by Olympus Microscopy)

(Diagram by Olympus Microscopy)

Because droplets of superheated steam from the espresso machine might be of a similar size than the wavelength of light, these droplets can exhibit the Tyndall effect.

Tydall scattering in a colloidal silver solution

Tyndall scattering in a colloidal silver solution

Amazingly, Wikipedia says that this is also the reason why blue eyes are blue!


The Tyndall effect is also what makes blue eyes blue (according to Wikipedia)

And with that, I surely learnt something amazing. Who knew that one could learn something about optical science from a coffee machine?