How to make clear Ice for cocktails

Clear Ice

Alternate Title: “If it’s not perfect, what’s the fucking point?”

My name is Ben. I like it when things are done properly*. This post, therefore, has been quite a while in the making.

It turns out making clear ice is hard. Very hard in fact. There’s a very good reason that companies like this exist. In fact, if you’re based in, or near, London – you should probably just stop reading this, man up, buy yourself a chest freezer to house some of their wares, and give them a call.

If like me though, you’re out of their delivery range, you’re going to need to make your own**.

Luckily, every single one of you has the requisite equipment for this, so there really aren’t any excuses. You can make big, clear ice with zero investment. Sounds like I’m trying to get you to buy a timeshare. I’m not. Promise.

“Why is clear ice so important” I hear you cry? Three reasons:

Firstly, Aesthetics. A massive chunk of clear ice floating in a drink looks amazing. Don’t underestimate the ‘first bite’.
Secondly , Dilution. A massive load of little microscopic bubbles in your ice (that’s what the cloudiness is) has the effect of hugely increasing the surface area of your block of ice, meaning it’s going to dilute your drink much faster.
Thirdly – Stability. Even assuming you take your ice chunk out of the freezer and let it temper properly, it’s going to be much more likely to crack into smaller pieces (and we’re back to point 2 – dilution) if different areas of the chunk have different densities.

Before we get onto the recipe – a bit about the theory behind this (or my hideously simplistic interpretation of it at any rate). There’s no shortage of people that will tell you that all you need to do to make clear ice is boil it. Or boil it twice. They are wrong. There are also people that say that all you have to do is use Bottled water, or filtered water. They, also, are wrong.

I’m going to use water that’s both filtered and boiled in my method below, but neither (or both) of these methods alone produces clear ice. The problem is actually pretty straight-forward once you take the time to sit down and think about it. It’s just simple state-change dynamics.

Y’see ice has 3 particular properties that make our lives very tricky when we’re trying to make a massive block:

1) Ice is less dense than water (0.92 is the exact figure), which means that water expands as it freezes.
2) Ice isn’t gas-permeable (at least not to the degree we’re talking about here)
3) Ice freezes from the outside in

Imagine for a moment, a block of ice thus:

Clear ice drawing

Why red pen, Ben? Ice is Blue, everyone knows that.

The outside of that block is frozen solid and impermeable (well, impermeable-ish – it will expand when pushed, but won’t let gas through), but the inside is still liquid. As the liquid freezes, it expands into the ice shell. This places a negative pressure on the liquid inside the block, meaning bubbles are going to form. These bubbles aren’t necessarily dissolved gases – they could just as easily be water vapour. Remember, Zero Degrees*** isn’t just the state-change point of water into ice – it’s the Triple-Point, so water-vapour can exist here and will certainly form when the liquid water is exposed to a lower-pressure system.

That’s what I reckon anyway.

Commercial ice-making machines use either controlled directional freezing (this video is an excellent demonstration of this) , or, for smaller cubes, a cold-plate ‘layering’ technique that ensures that the freezing surface constantly has liquid water venting over it to ensure there’s somewhere for the dissolved gases to go (back into solution). It’s well worth a look inside the ice machines they use in fast-food restaurants / bars if for nothing other than curiosity.

It’s worth noting that I tried the directional freezing method at home before settling on this one, and it works. Read Camper English’s Alcademics Blog all about it here. Now read the rest of the articles on there. They’re ace.

The reason I ultimately ended up discounting this method for my own personal use is that it takes a pretty long time to generate just one ice-ball. If you’re after a load of them for a party, it’s just not feasible. Likewise I discounted this (incredibly clever) method because I haven’t got room for a massive coolbox in my freezer (and I don’t think many people do).

I was after a solution that was practical, can be executed using close to no space in the freezer, and didn’t require any equipment. And I think I’ve got it.

So. Here we go. You get pictures too, you lucky lot:

Clear ice pot

Get yourself a pan. A clean pan. Really clean. Now clean it again, and wipe it out properly with Kitchen Towel. It, erm, has to be clean is, I suppose, the general message.

Clear ice filtered water

Fill it up with filtered water****. You want about 3 litres. Carefully. Try not to knock any air-bubbles into it. I’m filling this from a Brita Filtered water tap I’ve got at home, but water from a Brita Filter Jug is just as good.

Clear ice boiling water

Bring to the boil, and simmer for about 3 minutes. This will knock any dissolved gases out of solution.

Clear ice bowl

Carefully tip the water into a stainless steel bowl, and leave it to cool to room-temperature. You can cover it loosely with a tea towel to stop anything falling in, but don’t cling film it. We want to allow any residual dissolved gases to escape.

Clear ice container

Very carefully transfer approximately 1.5 litres of the water to a plastic freezer box. You want to minimise any splashing at this point, as you’ll re-incorporate those pesky dissolved gases.

Clear ice freezer

Sling it in the freezer.

Clear ice block

After somewhere between 2 and 5 hours depending on how buff your freezer is, you’ll have a block that looks something like this. Look carefully, you can see the central reservoir of water. The ice has formed perfectly clear, and pushed all the dissolved gases (and water-vapour if that’s what we reckon’s going on) into the reservoir. Here comes the clever bit:

Clear ice picking ice

Remove the partially frozen ice block from the freezer, and make a hole in it. Drain out and discard the reservoir-water, then re-fill the hole with more of the boiled water from earlier (remember you had 1.5 litres left over?).

Clear ice bowl

For the best results here, you’ll want to pop the stainless bowl in the freezer for about half an hour just to take it down to zero degrees. This stops the additional water we add from dissolving too much of the outer shell. Don’t be tempted to re-freeze the shell before tipping the chilled water in, as the difference between freezer temperature and the zero-degree water will shatter the fragile shell. I know whereof I speak.

Now pop the block back into the freezer and give it another couple of hours. Then pop it out, and repeat the drain/refill process. Now leave the block overnight to set solid.

Clear ice block

When you remove the block from the freezer, you can clearly see two ‘attempted cloudy spots’ which have been pushed progressively further inwards by our drain/re-freeze procedure. This, incidentally (and I’m sure you all thought this earlier. Dincha. Dincha?) is the reason that just freezing a big block and leaving it alone isn’t good enough – the inch or so that’d be left around the outside of the block isn’t thick enough to provide us with a meaningful chunk. The reason that the block looks cloudy at this point incidentally is that by pouring in fresh water to the top of the void in the ice, we create a new, very thin layer of ice on the top of the block that needs chipping off. You can see that top ‘strata’ more clearly a bit further down.

Clear ice breaking ice

This is the bit you’ve been waiting for – split the block in half using an ice-pick (we’ve got some cool ones here).

Clear ice chunk

As well as that clearly-defined ‘strata’ of new ice on top, you can see that the cloudy void created is both very small, and very well-defined.

Clear ice pick

Now comes the fun bit. Chip the block apart (working around the cloudy void) with your ice-pick into hunks that are big enough to be impressive, whilst being small enough to fit in your glasses. You can see from this image that there’s really very little waste. This includes the middle bit, and all the ancillary chipping…

Clear ice in container

And there we go. A big tray of lovely, clear, long-lasting, pure-tasting ice. Winner.

Clear ice in glass

I picked a chunk at random so that you can see the quality – it’s pretty darned close to perfect (perspective/refraction through liquid’s a bitch, but that chunk is actually as big as the glass is wide). There’s a tiny bit of haze on one side which is from one of the attempted cloudy spots. I could have been more thorough with the chipping process, but to be honest if there’s a tiny bit of haze left on the outside of a chunk, it’ll disappear pretty quickly due to the increased rate of melting in that sector of the chunk.

There we go then.

If you’re having a crack at this at home – don’t forget to tweet some photos to us @masterofmalt – we’d love to see how you get on, and remember, you’ll need one of these chunks for this week’s #MasterofCocktails recipe – The Highball. You can tune in at 6pm on Sunday night on twitter, or alternatively catch up on this very blog on Monday, when we’ll publish the full recipe.



* I’m almost certainly on some sort of spectrum or have something that has an acronym of some kind, but opening that whole analytical box of frogs isn’t something I’m particularly keen to do today.

** Not necessarily true. I did at one point give very careful consideration to taking a suitcase, some tin foil and a load of bubble wrap up to London, and simply collecting a chunk. In fact I may still do that. Stop judging me.

***Okay, 0.01 degrees c, not Zero. Pedant.

**** Filtered water, not mineral water. You want impurities out, not minerals (for which read impurities) in.

Categories : Cocktails, News

12 comments on “How to make clear Ice for cocktails”

  1. Dan Skinner says:

    I spent WEEKS before Christmas trying to work out the clear ice method!!
    This way seems pretty cool (geddit?), but if you can stretch to a purpose made bit of kit, I had one of these sent over from the states…
    Works on the theory of ice forming from the top down (think of a frozen lake), so all impurities and air bubbles are pushed downwards.

    I can tell you it works great !
    I can also promise I have nothing to do with the company, they were a kickstarted (though I was too late for the offering).

    Hope that helps some 🙂

  2. They look pretty good Dan – using the same directional freezing method as the Alcademics blog Ben’s highlighted above, but without the need to have a whole pan of water in your freezer or messing on with balancing wires.

    Not cheap though, even before shipping – and it still only gives you one ball at a time.

  3. That’s a very, very nice bit of kit. Very clever indeed.

    Might get in touch and see if they fancy sending some over to the UK…

  4. Dan Skinner says:

    Yep, not cheap, but I was literally going nuts trying to make clear ice balls !!
    Theres a brilliant Japanese contraption made from big blocks of aluminium that melts the ice into a ball, but you still need to make the clear ice first…and they are stupidly expensive too – look great though. The kit from Wintersmiths makes one ball a day – fortunately I only drink at the weekend, so I have a bag of balls ready made in the freezer now.

    I would offer to post an ice ball out to you Ben, but I dont think that would work….if you`re around Merseyside some time though, you bring the single malt and I`ll provide the ice ball 😉

  5. Dan, Peter,

    I’ve used those Aluminium ball-makers before, there’s a *big* problem with them that no-one ever seems to talk about – their thermal capacitance.

    They’re *awesome* for the first couple of balls, but once you get to the point where they’re a bit chilly, they simply stop melting the ice until they’ve managed to suck in enough warmth from the environment (which take ages, obviously).

    They’re very cool if you’re only after a couple (or a few), but if you’re having a gathering and want to make a load of drinks they’re as good as useless.

    Update on the Wintersmiths solution – I had a chat to them a couple of weeks ago and they’re sending a sample over, so fingers crossed…


  6. Dan Skinner says:

    Aww Ben !!
    I was hoping for a visit with you toting a bottle in hand !!

    Those aluminium ball makers are what I was refering to, the problem is with creating the clear ice to use in them…and of course the temp as mentioned.

    Ben, are you getting a sample in to test and maybe stock for sale in the future?

  7. Maaaaaaybe.

    Watch this space, that’s all I’m saying 🙂

  8. Jeff says:

    I have the Cocktail Kingdom sphere-maker, and it works very nicely. Tip for making multiple spheres in a row: Just run the two metal cylinders under some warm tap water. They have such a high thermal conductivity, it doesn’t take long to have them ready to make a new sphere.

    My problem has been getting large enough chunks of clear ice to put in the CK sphere-maker. I have one of these:

    but the thickness of the block of clear ice that it makes is less than the diameter of the CK sphere (55 mm). I end up with a sphere minus a cap. Not aesthetic, though one could argue that it might even be a superior shape to a full sphere because part of whatever shape ice you have floats above the whisk(e)y, where it is exposed to room-temperature air. I haven’t done the finite element analysis to see what the optimal shape to minimize melting is in that situation… it may well be that a "flat topped" sphere is better because it doesn’t stick up as far into the thermal boundary layer just above the booze. On the other hand, any edges have got to be bad for melting. But I’m getting off-track…

    The website for the Wintersmith Ice Baller says it produces 60 mm spheres of clear ice. Looks like it operates on the same princple as my polar tray. So I’m thinking the thing to do is get an Ice Baller to make a clear ice near-sphere, and then pop it into the CK device to get that machine-tooled sphere of perfection.

    Yes, I have a problem too…

  9. I think the rule of thumb is that in a closed system, the closer you can get to a sphere, the better (hence a curved top would be better than a flat one), but my guess is that the difference is going to be minimal enough to warrant ignoring in comparison to the interaction with the liquid.

    What thickness of clear ice do you get from the Polar tray?

  10. Kyle says:

    Needless to say this doesn’t happen in your freezer as it would at the triple point. If it was, the water would also boil at said conditions.

    The reason that water sublimates is because the partial pressure of water in your freezer is very low, thus it sublimates.

  11. Mike says:

    I get exceptionally clear ice balls at 6 degrees F using "directional freezing".
    <a href="; rel="nofollow"></a&gt;

Leave a Reply

Your email address will not be published. Required fields are marked *