I’ve written about water before and in particular, if you’ve been paying very close attention, you might remember that November 12th marks the anniversary of the day, in 1783, that Antoine Lavoisier formally declared water to be a compound rather than an element.

Which means that November is always an excellent time to talk about water. But this time, I’m going to focus on its solid state: ice.

A few days ago I stumbled across some beautiful images of hair ice, which prompted me to make a #272sci Twitter post (keep an eye on that hashtag for similar small bits of interesting science). The story behind hair ice is a fascinating one, and not something I could truly cover in 272 characters – so here’s the slightly longer version…

This form of ice is found on dead wood, and it has a few other names, including ice wool or frost beard. Of course, ice naturally forms at 0 ℃ at standard atmospheric pressure, but the form we’re most familiar with looks, to the naked eye at least, rather more random. In fact, it was snowing here just yesterday, which means I have photos!

As you can (hopefully) see, there’s some regularity to the individual crystals, but they’re sort of growing all over the place. So, how do ice crystals form, and why?

We need to start with the structure of water. Now, you might imagine that a molecule with the formula H₂O would have its atoms arranged in a straight line, like this: H–O–H. But it doesn’t, and the reason it doesn’t is that the oxygen atom in the middle has two pairs of electrons which aren’t involved in bonding – which chemists call ‘lone pairs‘.

Imagine, for a moment, that you have a bunch of balloons made up of two sausage-shaped balloons and two round ones, all attached at the neck. What shape would they make, as a whole? Probably, the two long balloons would form a sort of rough V, and the two round ones would stick out, opposite each other.

If you have some balloons to hand, give it a try. It turns out this is actually a pretty good model for water. We end up with a roughly tetrahedral shape, with oxygen in the middle, hydrogen atoms in two of the corners, and the lone pairs in the other two corners.

This is important because those lone pairs don’t just sit around doing nothing. The element oxygen is very electronegative, which means it likes to attract bonded electrons. Hydrogen, by contrast, is more electropositive, which essentially means it doesn’t.

The result of this is that, although it is very definitely a covalently-bonded molecule (and not made up of ions), the oxygen atom in water has a partially negative charge, while the hydrogens have a partially-positive charge.

Since positive charges attract negative charges, and since molecules don’t exist in isolation. The result is that the hydrogen atoms in one water molecule are attracted to the oxygens in other water molecules. This is called hydrogen bonding.

If your head is spinning as you try to imagine this, take a look at the image below. White is hydrogen, red is oxygen, and the dashed lines represent the attractions between partially-positive hydrogens and partially-negative oxygens.

Do you see the shapes that form? Yes – hexagons!

And how many sides does a snowflake have? Yes – six!

It’s not a coincidence: as the temperature drops, molecules that previously had freedom of movement gradually stop moving so much and pack into these hexagonal shapes. Then, water vapour in the air deposits onto this skeleton and, voilà, we end up with six-sided ice crystals.

Now, normally, this happens fairly randomly. Yes, all the snowflakes are hexagonal (and there are images of the different patterns that can form in this graphic from Compound Interest) but, as my photos of ice crystals suggest, they tend to stick out in all directions.

Hair ice is different. The ‘hairs’ appear at what are called wood rays, that is, lines perpendicular to the growth rings of the wood, and it turns out that if a piece of wood forms hair ice once, it will probably keep producing it – which makes things rather easier for the potential photographer!

Each of the hairs is about 0.02 mm thick and, assuming the temperature doesn’t rise above freezing, they can hang around for hours and even days.

Which leads to the question: why don’t more ice crystals grow on top of the threads and break up the hair-like structures? After all, if it’s cold enough for ice, it ought to be cold enough for, well, more ice – oughtn’t it?

A quick aside: you’ve probably heard of Alfred Wegener, discoverer of continental drift – an idea that ultimately led to modern tectonic plate theory. These days, those ideas are pretty universally accepted, but when Wegener first proposed continental drift in 1912, he faced a lot of opposition. There was more than one reason for this, but one major one was that Wegener was seen as an outsider to the field of geophysics. His background was in meteorology and polar research. In other words, he spent a lot of time in cold weather conditions.

Which brings me back to the main, ah, thread (sorry). Alfred Wegener described hair ice on wet dead wood in 1918, having observed it the year before, and suggested that mycelium, the thread-like part of a fungal colony, could be involved. He thought this because he could actually see mycelium on the branch surface, which was confirmed by his consultant, Arthur Meyer. Meyer, however, was unable to definitely identify the fungal species at the time.

Some years later, in 1975, scientists named Mühleisen and Lämmle actually managed to grow hair ice on rotten wood in a climate chamber and later still, in 2005, the physicist Gerhart Wagner again suggested that a fungus was involved, although he had no knowledge of Wegener’s observations when he first did so. He went on to carry out experiments with Christian Mätzler in which they were able to reliably grow hair ice on a balcony on nights with freezing conditions.

After lots of painstaking (and cold!) observation, they concluded two things: firstly, hair ice forms from water stored in the wood, not atmospheric water – which goes some way to explaining why the structures aren’t more random, as you’d expect if the ice were forming from water vapour in the air.

Secondly, the fungus, as a product of its metabolism, was generating gas pressure, and that was pushing water through the wood rays to the wood surface, where it was fanning out into fine, curling strands.

So, yes, in a way, hair ice is the product of freezing fungal farts. (Yes, yes, very tenuous, but I couldn’t resist ‘freezing fungal farts’, let me have this one.)

There’s a much more scientific explanation in this 2015 paper, the full text of which is freely available online (lots of great photos too!). The culprit turns out to be a fungus called Exidiopsis effusa. Inside the wood, attractions between the water molecules and the wood surface lower the melting point of water slightly, keeping it liquid. Products of wood decomposition left by the fungus also (probably) help to prevent ice forming inside the wood itself.

Once the outside temperature drops, though, the formation of ice crystals on the outer surface of the wood has the effect of drawing out more water, and the result is that the crystals grown in long, thread-like structures – although the fine details of how the fungus does what it does are still a bit of a mystery. Still, it’s nice to find a not-quite-answered science question, isn’t it?

One final thing: just in case you were thinking, oh, come on, is that first picture really real? On the Chronicle Flask Facebook page, a user named Tiarra Friskie commented that they had pictures of this very phenomenon, taken in British Columbia, Canada, and kindly gave me permission to use them. So, here you are: a tiny bit meltier than the picture above, but nevertheless, two guaranteed genuine photographs of hair ice!

If you live somewhere in the vicinity of the latitudes between 45 and 55 °N (which includes most of the UK, by the way), keep an eye out for rotten wood in your local broadleaf forest – if the weather gets cold enough, you might just spot some hair ice yourself.

---

A little admin note: the chronicle flask blog is now (yikes) almost nine years and 150 posts old. Life is increasingly busy and so, after December 2021, I’m going to stop making monthly updates. But not to worry! You can still follow the Twitter hashtag #272sci for regular tiny pieces of science, and I’ll pop back every now and then. Oh, and please do consider supporting the Great Explanations book project here!

Plus, why not take a look at my fiction blog: the fiction phial? And you can also find me doing various flavours of editor-type-stuff at the horror podcast, PseudoPod.org – so head over there, too!

Content is © Kat Day 2021. You may share or link to anything here, but you must reference this site if you do. You can support my writing my buying a super-handy Pocket Chemist from Genius Lab Gear using the code FLASK15 at checkout (you’ll get a discount, too!) or by buying me a coffee – just hit this button: