Under pressure

What do you notice about this iced latte? The cup is rich with physics, but for this post, the important bit is the floating ice on top.

A coffee should be a time for relaxation, for reflection. As we come to the end of summer here in the northern hemisphere, we may want to enjoy one last iced coffee before we return to the warming coffees of winter. If on the other hand you are reading this from the southern hemisphere, the equatorial region, or some time after it was originally posted, you may be just starting to enjoy your iced coffees again. Either way, ice is remarkable and it is good to make some time to enjoy it. One of the things that makes it remarkable is what seems to be its very ordinariness: it floats.

Ice floats because the solid form of water is less dense than the liquid form. This is actually fairly unique to water. Most liquids get more dense as you cool them. As they transform into solids, they get denser still. This would mean that if you were to cool a liquid until it starts to solidify, the solid would sink, not float, on the liquid. If water were like most other liquids, all the ice in our iced-coffee would be at the bottom, not jiggling at the top. In addition to what would be an almost aesthetic problem for the coffee, this has consequences for life itself. When a lake or a pond freezes over, the fish and other aquatic life, can survive under the ice in the denser water. This odd property of water has helped life to evolve.

The reason for this strange behaviour lies in the way that water molecules bond together. Each water molecule can bond to a neighbour through a hydrogen bond. This optimises the structure to a layered form of well spaced hexagons (link here for an interactive model of water ice). Each corner of the hexagon is an oxygen atom. The size of the hexagon means that, if they weren’t arranged into a regular lattice, the water molecules could get closer together than they do in the solid phase. Which is another way of saying that the liquid can get more dense than the solid. Ice will float on water.

The layered structure of the ice crystals also means that each hexagonal face will tend to glide over the one below it or above it. It is this property of ice that means that we can determine the direction of glacial flow in centuries past. When fresh snow falls on top of a glacier, the density of the snow layer is about 50-70 Kg/m3. For comparison, the density of water is 1000 Kg/m3. Although each snow crystal is hexagonal, they have random orientation as they fall. As new snow falls, it pushes down on the old snow and compacts it until, about 80m down into the glacier, the density of the (now) ice is 830 Kg/m3. As the depth increases still further, the density increases to 917 Kg/m3 which is as dense as a glacier can be but is still much less than the density of water; a glacier would float. When the snow crystals are pressed down, the hexagonal layers of ice will glide past each other in the direction of push and the crystals will re-orienate. They will also grow as they merge with other crystals and as a result of the heat from the bedrock beneath them. This means that deep in the glacier, more of the crystals will be orientated in the direction of the push. Taking a vertical core of ice and looking at the orientation of the crystals in 0.5mm thick cross sections therefore reveals how they have been pushed as a function of depth. This in turn reveals which way the glacier has flowed in the past.

Sun-dog, Sun dog
A ‘rainbow’ of colour as seen in a ‘sun dog’ observed in central London. Note the order of the colours.

The structure of ice has one other surprise for those of us who are enjoying more coffee outside. Depending on the weather conditions, high up in the atmosphere, hexagonal ice crystals form. Because they are hexagons, they are, in effect, a section of a 60 degree prism. This means that light entering through one face, will be refracted twice to emerge from the crystal at 22 degrees relative to where it came from. If there are enough of these crystals high in the atmosphere, a bright circle will form around the Sun. For reasons that are probably obvious, it is known as the 22 degree halo. It seems fairly difficult to observe this halo. What is far more common to see are two bright regions of light at the 9 o’clock and 3 o’clock positions on the halo. In addition to being brighter than the rest of the light circle, these two regions often appear like a ‘rainbow’, but with the red on the inside of the halo and the blue on the outer edge. Known as “sun dogs” or parhelia, they too are a consequence of the ice crystals. As the ice crystals fall, they are more likely to fall flat so that each hexagonal face is horizontal. More of these ice crystals means that there is going to be more light refracted at the position horizontal to the Sun and so the light there is intense. They appear as separated colours for the same reason that the colours disperse with a prism: each wavelength of light has a very slightly different refractive index and so gets ‘bent’ by a slightly different amount. The ice crystals are bending the red a bit less than the blue.

This is a good time of year to keep an eye out for Sun dogs and haloes. And if we can do so while enjoying a well made iced coffee with the ice cubes floating at the surface, all the better.

Please do share any photographs you have of coffee with 22 degree haloes or sun dogs, either here or on Facebook or Twitter.

General Science history

Super cold brew

Cold brew coffee with ice
Cold brew coffee served with ice. Image from

How cold do you drink your cold brew? Poured over ice? As an experimental physicist who works with liquid nitrogen (& helium), I was initially quite intrigued to learn of nitro cold brew coffee. Could it be coffee that somehow uses liquid nitrogen to fast-cool it, what would that do to the taste? You would expect liquid nitrogen (at -196ºC) to rapidly cool the coffee below its freezing point, after all, it is how Heston Blumenthal makes ice cream. To make a drink-able cold-brew with liquid nitrogen would require great skill, especially given the potential health risks. It would be another situation where you may well ask yourself, “what’s the point?”

However, it turned out that the reality was far more mundane, gaseous nitrogen is passed through cold brew coffee to create a drink with a silky mouthfeel. A smooth drink that comes straight from the tap just like stout. Such a drink is going to behave as an ordinary liquid and chilled only to the point that it is kept in the vat. The novelty would presumably come from the mouthfeel introduced by the many tiny bubbles distributed through the drink. Just as with water, if you cooled the nitro-brew below its freezing point it would solidify and form coffee cubes. No real difference to get excited about. But what if there was a very different sort of liquid, a “super liquid”, that didn’t behave like water, coffee or even liquid nitrogen but one that could leak through solid cups?

Superfluid helium is such a liquid. Like water, oil or even liquid nitrogen, when you cool helium (the same gas that is in party balloons)∗, it becomes an ordinary (but very cold) liquid at -269ºC. But unlike those liquids, when you cool it further, below -271ºC, it does something very odd indeed. It becomes a superfluid in which the liquid moves with zero friction or equivalently, zero viscosity (honey is very viscous, water is very much less so).  And it is because of these properties that it can do some astonishing things such as stream through cracks in containers that were thought impermeable (see the video at 0:52m), or even climb the walls of the container it is in (1:13m)!


To explain the behaviour of superfluid helium it is necessary to use quantum mechanics. Indeed, Fritz London (1900-1954) is said to have described both superfluidity and superconductivity (which happens in solids) as “quantum mechanisms on a macroscopic scale”. At the heart of the theory of superfluidity is the idea that the helium atoms fall into the lowest energy ground state possible, a Bose-Einstein condensate. To form a Bose-Einstein condensate, the particles (atoms of helium) have to  be bosons rather than fermions. All particles in nature can be categorised as either bosons or fermions. The difference between the two types comes from another quantum property of particles, the spin. Spin is related to the angular momentum of the particles and, this being quantum mechanics, can take only discrete values, either whole number or half integer numbers.

cold brew, doublemacbex
Another photo of cold brew coffee, this time from Bex Walton (flickr) – note the condensation around the rim, much could be said about that. Image CC licensed.

Bosons are particles with integer values for spin, fermions are particles with half integer values. Most of the elementary particles you will have heard of are fermions: electrons, protons, neutrons, they’re all fermions. Some particles however, such as the photon (the particle of light) are bosons. Helium 4 atoms are effectively composite bosons, because of the combination of 2 protons, 2 neutrons and 2 electrons that make up the atom. When you add their individual (half-integer) spins, you will get an integer spin, hence a boson not a fermion. The distinction is important because while bosons can share a lowest energy state (the Bose-Einstein condensate), fermions cannot. Quantum mechanically, no two identical fermions can share an energy level (the Pauli exclusion principle), so you can never get to a state where all the fermions are in the lowest energy state. There are practical, every day consequences of this for us, such as the way metals such as copper conduct electricity and heat, the fact that the electrons in the metal are fermions turns out to be crucial for us to understand how metals ‘work’. In contrast, the fact that the helium atoms are in the lowest energy state in super-fluid helium means that the ‘liquid’ behaves very strangely indeed.

We seem to have come a long way from the idea of a cold coffee. But perhaps next time, if someone offers you a “super cold brew” take a moment to think of the physicists who get to play with some real super cold superfluids†. Hope you enjoy the video.


*Technically it is Helium 4 that becomes superfluid at 2.2 K (-271ºC). The rarer isotope, Helium 3, does not become superfluid until much lower temperatures and even then, the superfluidity has some very special properties.

†Although I do get to work with liquid helium (and although it is mostly helium 4), I work at the relatively ‘hot’ temperatures at about -269C. At this temperature the interest is not so much in the liquid helium itself but its use as a coolant for other materials.



cafe with good nut knowledge Coffee review Observations Science history

Crystal Perfection at Workshop, Holborn

Workshop coffee Holborn
Diamonds are forever, Workshop coffee Holborn

The brand identifier of Workshop coffee is a diamond, a representation of which hangs on the wall as soon as you enter the Holborn branch. I had arrived at Workshop in order to try their coffee after I’d had a great espresso made with beans roasted by Workshop at Knockbox in Lamb’s Conduit Street. The coffee brewed in their own café certainly did not disappoint. I enjoyed a very good La Soledad filter coffee and a cake (which was confidently nut free, this brings me to another plus point for Workshop, they know the ingredients of their cakes!). The interior of the cafe, just beside Holborn Viaduct, is quite spacious and, if you sit at the back, you get a great view of the workings of the espresso machine as different people come in to get their ‘take out’ coffee. It is very possible to spend quite some time here in order to relax and enjoy your coffee while taking in your surroundings. To a physicist who studies materials (like me), the diamond logo of Workshop represents a fantastic material. A material in which the structure of the crystal determines so much about its properties. Were the carbon atoms in diamond bonded slightly differently, they would form the soft, pencil lead material ‘graphite’, rather than the hard, transparent material of diamond.

unit cell, repeating structure
The floor at Workshop reminds me of my crystallography text books.

Whether it was intentional or not, the crystal theme of the logo was replicated in the floor tiling of the Holborn branch. Crystallography is a branch of science that probes the building blocks of solids. It reveals how the atoms that make up different solids are arranged to form the solid. The atoms could be arranged in a simple cubic arrangement (as with salt) or hexagonally (as is the case for graphite). To establish the crystal structure you need to find the smallest repeating unit in the whole. Many introductory solid state physics or crystallography text books use 2D examples of repeating structures to help the student to understand how to build up these “unit cells” into full blown crystals. Many of the examples of such lattices look stunningly similar to the floor at Workshop.  Fundamentally, the idea of the crystal is that it is a simple repeating structure, just like the floor of Workshop. Indeed, the word “crystal” as used by Pythagoras implied perfection, harmony and beauty, a sense that is really conveyed by the crystal structure of the diamond logo of Workshop.

Crystal cake, LaFeSi cake, grape atoms
When a colleague left our lab, we made her a  cake that was a representation of part of the crystal structure of the material that she had worked on. Chocolate grapes and profiteroles represent different atoms in the structure.

The ancient Greek term for “crystal” actually implied the type of hard ice that is wonderfully clear and transparent. And it is ice that connects the area surrounding Workshop with a famous chemist who won a Nobel prize for his work in crystallography in 1962.  Max Perutz (1914-2002) described crystallography as a technique that “explains why diamond is hard and wax is soft, why graphite writes on paper and silk is strong”. Once you have enjoyed your coffee at Workshop, if you head down the stairs on the viaduct and descend to Farringdon Road you quickly get to Smithfield Market. It was here that, during the Second World War, Perutz helped to develop the material Pykrete. A “secret weapon” of World War II, Pykrete was developed five floors below Smithfield Market in a room cunningly disguised with animal carcasses. The planners in the war effort had wanted to design a boat made of ice but the problem was that when it was shot at, ice shattered. Could scientists develop a type of ice that would not shatter if it got hit by enemy fire? Pykrete was the answer. Pykrete uses the fact that materials such as plastics can be strengthened by adding fibres to them. In the case of Pykrete the “fibres” were sawdust and the material to be strengthened was ice. Not only does it not shatter when shot at (instead, the bullet creates a crater in the ‘boat’), it takes a lot longer to melt than ordinary ice. The sawdust encased in the ice acts to insulate the ice and increase its longevity.

Perutz’s Nobel prize was for his work to determine the crystal structure of haemoglobin, it took ‘just’ 25 years to do so. The field of crystallography continues to enrich our understanding of the behaviour of solids, though now we’re expected to get results more quickly than the 25 year time frame Perutz enjoyed. If you know of a good café where lots of physics goes on, or of a good café near a site of special (or unexpected) scientific interest, (or even just a good café) please do share your story either in the comments section below or by contacting me on email, Twitter or Facebook.

Workshop Holborn is at 60 Holborn Viaduct, EC1A 2FD

Quotes and other useful facts taken from:

In our time, 29th November 2012: Crystallography“, (BBC Radio4)

Max Perutz “I wish I’d made you angry earlier” (2002),

Ichiro Sunagawa “Crystals, Growth Morphology and Perfection”, Cambridge University Press (2005)