Clear Ice in Cook's Illustrated
Matching the Fitness Program to the Liquor Brand

Make Clear Ice Balls Using a Thermos

Most readers of Alcademics are probably familiar with this by now, but I realized that in my many (many, many) ice posts here on Alcademics I don't have a straightforward post listing how I most often make large perfectly clear ice spheres at home. So I made this post to insert into my ice project index. 

6a00e553b3da2088340240a501d5c8200bThis method scales up so if you want to make 10 large clear ice balls at a time you can do so without taking up a lot of room or buying more expensive versions of this set-up. 

The process is based on Directional Freezing (pushing the cloudy part of freezing water in one direction, in this case out the hole in the bottom of the ice ball mold). The particular thermos to use was the suggestion of a reader on Alcademics. You nerds are the best! 


Ice balls molds - You'll want the large, 2.5 inch ice ball molds that are circular. 

Thermos - The most compact shape that fits these large ice sphere molds perfectly is these Thermos Funtainers. (You can use an insulated coffee mug or other thermos-style container if you want.)


  1. Fill the thermos with water to the top. The water doesn't need to be distilled or boiled or any of that. 
  2. Fill the ice ball mold with water.
  3. Hold your thumb over the hole in the ice ball mold and place it upside-down on the thermos. The water will stay in the mold part as well as the thermos.
  4. Put it in the freezer. The ice ball mold part will freeze first because it is not insulated. As it freezes, clear ice forms first and it will push the cloudy part of the ice down out the hole in the bottom (because it's partially insulated sitting on top) and into the thermos. The thermos should only just be starting to freeze when the ice ball on top is done freezing. 
  5. Wait about 24 hours. Try not to let it go for more than a day and half or so, as you don't want the water in the thermos to completely freeze, as it expands and will deform your thermos, or even shatter if it's a glass-insulated thermos. 
  6. The ice ball mold should easily pop off the thermos and be perfectly clear or pretty darn close.

Ice balls in thermos01
Ice balls in thermos01Ice balls in thermos01Ice balls in thermos01

Ice balls in thermos11
Ice balls in thermos11
Ice balls in thermos11
Ice balls in thermos11Ice balls in thermos27
Ice balls in thermos27


  • If your freezer is set on super duper ridiculously cold it may freeze too fast for this to work (it won't push the cloudy part out the bottom but instead into the center of the sphere). You probably don't need your freezer quite that cold!
  • If you're getting more of an egg shape from your ice sphere, try leaving a little space in the mold and/or thermos. 
  • You can also "polish" your egg-shaped ice sphere into a better sphere just using a clean dish cloth, or even just rubbing it in your hands.
  • If you're opening and closing your freezer and generally jostling the set-up, it tends to be cloudier. 


Be sure to check out the many many other things you can do with ice at the Index of Ice Experiments on Alcademics



Feed You can follow this conversation by subscribing to the comment feed for this post.

Andy L

In light of, I was thinking of adapting this experiment to freeze from the bottom up; namely to put the thermos upside-down on top of the ball instead of putting the ball upside-down on the thermos. I've ordered some ice ball molds, already have the thermoses (thermi?) and I'm looking forward to trying this out.

Camper English

What would be the reasoning/advantage to doing this bottom-up?

Andy L

My thought is to avoid the timing issue (having to pull the assembly out before the thermos-water freezes through) and the inversion of the ice mold in water (also not having to fill the thermos with water).

If the seal between the mold and the thermos is good, then the air in the thermos might provide adequate insulation and allow the freezing to proceed directionally. So the air would be forced out the top of the mold, into the thermos itself (which will have a slight negative pressure as the temperature of the air drops).

This is all very theoretical; my ice molds get here in two days so I'll try it out and report.

Camper English

Thank you for the explanation. I had not considered that we might be able to eliminate using water in the thermos that way. I guess it will be a matter of whether the ice freezing in the ice ball floats to the top and plugs the exit hole.

And even if it doesn't work with air, perhaps we could only use a small portion of water inside the inverted thermos rather than filling it, which should prevent the thermos deforming or cracking when we inevitably forget to take it out of the freezer in time.

Good idea and certainly worth a try.

Andy L

First attempt came out just okay. Compared to the control ice ball it was significantly clearer, but had a spike of cloudy ice running up the center of the sphere, from the bottom to where the plug is. Both balls were filled to the brim; the control one split the mold because of expansion, the thermos one basically shot a plume of water out the top.

I'm going to try a little more carefully for the next run; modifying as little as I can but using a kitchen scale to fill it 91% capacity to account for the expansion.

Camper English

After your last response I also tried the thermos upside down trick and found the exact same thing (didn't comment earlier as a spoiler because experimenting is all the fun of this) - I did one water in the mold with the upside-down thermos, and another with some water in the upside-down thermos with the idea that the water would provide additional insulation better than air.

In both cases I too found clear outside but a tornado shaped cloudy part from bottom to top - as you say a definite improvement over just the mold but still lots of ugly air. This would mean the last part to freeze was the center, so it seems that the top where the hole is does get plugged up with ice, not leaving an escape route for the air.

While I don't think your 91% experiment will be successful as there is always air in the water, what I find with all this ice stuff is that practice is more important than theory and there are surprises to be found.

Please do let me know what you find and thank you again for thinking up this set-up/experiment!

Deke Dunne

Hey Camper!

I have tried the ice mold, thermos approach a couple of times now and am left with half clear, half cloudy iceballs. I inverted the molds, and the ice near the bottom is clear, and the ice at the top is cloudy. I tried tilting them slightly to no avail. Is this normal? Or is there a way to combat this?

Camper English

@Deke - 1. You have the hole facing into the thermos right? 2. Assuming yes, is your freezer super cold? It's possible that the freezer setting is so cold that the hole in the bottom freezes over before the clear ice from the top and push it into the thermos. Try either turning up the freezer temp or insulating the top of the thermos by wrapping a small towel or something there to see if it helps.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)