# Making Ice with no air bubbles

Is it possible to make very clear ice in a home freezer?

Tap water ice is often cloudy white, which I think comes from minerals in tap water. Making ice with distilled water is a pretty conclusive experiment, I think.

However, the ice cubes I make with distilled water still have many small air bubbles frozen into them. I have tried melting and re-using the ice in the hope that there would be less dissolved air in the water on subsequent freezings, but it did not seem to make a difference.

By examining cubes as they are freezing, you can see that the top gets cooled the most, and this skin of ice can trap air. I have tried putting the water in bit by bit. With a thinner layer of water, I expected the air might reach the top before the water freezes. There are still bubbles.

I put a container with a small hole in it above a dish. I filled the container and let the water drip out. I cooled the water in the container first, in the hope the drops would freeze rapidly. The hole froze shut before the container could empty, but the ice in the dish below still had air bubbles.

Related Other Physics Topics News on Phys.org
Integral
Staff Emeritus
Gold Member
Have you tried boiling the water before freezing it?

cronxeh
Gold Member
Maybe try applying pressure on the closed container of water and freezing it?

rachmaninoff
What if you froze the ice cubes under a very inert gas like helium or argon? I'm just guessing here.

remember, I want to do this in my freezer, so I can't put the water in a closed container with helium.

try uses a thick and small metal container, freeze the container first and pour warm water in..... the water temperature will drop fast and uniformly..... hopefully this will make a little different........ I will try tis tonight and tell you guys the result....

HallsofIvy
Homework Helper
The cloudiness is not from minerals, it is from air dissolved in the water. You can get clearer ice, not necessarily perfectly clear, by boiling the water first to drive out the air, then freezing it.

(Yes, I know it sounds silly to boil water before you freeze but---)

Q_Goest
Homework Helper
Gold Member
Water is often prepared for experiments where all the air must be removed by exposing the water to a vacuum for a period of time sufficient to get the air out.

pervect
Staff Emeritus
I found this link which confirms Halls of Ivy's statements:

Deaeration

When fresh water is heated up, air bubbles start to form as soon as the temperature begins to rise. The water can not hold the air when the temperature increase. At 100 oC (212 oF) the water begins to boil. If the water is cooled down at then reheated, the bubbles do not appear until the water starts to boil. The water is deaerated.
http://www.engineeringtoolbox.com/24_639.html [Broken]

The other thing I can think of to try is to see if you can find or borrow one of those vacuum devices that they sell for freezing food. (I'm not actually sure how well they work though, I just see them advertised all the time on TV, only ninety-nine ninety-nine ninety-nine :-)). But it looks like boiling the water in an open container should work just as well from the solubility chart.

Last edited by a moderator:
Gokul43201
Staff Emeritus
Gold Member
Boiling the water, just prior to freezing will definitely help reduce air bubbles. You really can not eliminate ALL the bubbles this way, as the water will redissolve air as it cools. You wnat to minimize the time that this water stays liquid, so I would set the thermostat on the freezer to the coldest it can get, and put the freshly boiled water in quickly. If you do certain tests in advance you could optimize the temperature you heat the water to in order to be able to make use of the mpemba effect.

Since the typical solubility of air in cold water is about 30 cc/L, and an ice-tray holds...what...300 mL or so, that's about 10 cc of air that is dissolved in it. If you can make the total volume of air available to the water of order a couple tens of cc's, then you can slow down the rate of dissolution. So what all this speculation is leading to is that a tight fitting ziplock bag, or something of that nature MIGHT help.

Final suggestion, and this may not be feasible : there are those vacuum bags that they advertise on TV for putting clothes and stuff into, that attach to a vacuum cleaner or such. Something like that would work like a miracle...I think.

Last edited:
Chronos
Gold Member
You can do this with aluminum foil and metal ice trays. Take a piece of foil more than large large enough to cover the ice tray. Form a cylinder about a half inch long in the center of the sheet using the eraser end of a pencil, or similar object. Poke a hole in the top of the cylinder using a round toothpick. Place sheet over filled ice tray [foil cylinder pointing up!]. Seal foil around edges of ice tray. Boil water in tray [set burner on medium, you want a slow boil]. Once boiling and you see steam coming out cylinder hole, remove tray from burner and pinch off cylinder closing the hole [caution, wear oven mit]. This will prevent air from reentering. Freeze to taste.

Gokul43201
Staff Emeritus
Gold Member
Chronos said:
You can do this with aluminum foil and metal ice trays. Take a piece of foil more than large large enough to cover the ice tray. Form a cylinder about a half inch long in the center of the sheet using the eraser end of a pencil, or similar object. Poke a hole in the top of the cylinder using a round toothpick. Place sheet over filled ice tray [foil cylinder pointing up!]. Seal foil around edges of ice tray. Boil water in tray [set burner on medium, you want a slow boil]. Once boiling and you see steam coming out cylinder hole, remove tray from burner and pinch off cylinder closing the hole [caution, wear oven mit]. This will prevent air from reentering. Freeze to taste.
Very nice !

This may sound wild but it is proven that Hot Water freezes faster then Cold water. Therefore it wold be more effiecient to use boiled water then freeze it becuase it will have less air and it will freeze faster because it is hot. Since the water will freeze faster it will have less time to allow air to dissolve in it as well.

Check it out!
http://math.ucr.edu/home/baez/physics/General/hot_water.html

I'm not sure if this will work...

What if you coat the boiled water with a layer of oil, and then freeze it?

Presumably the oil in the water would stop air being transfered, but if you are using these 'airless icecubes' to consume then I think there is a bit of a health risk

V-Tec said:
What if you coat the boiled water with a layer of oil, and then freeze it?
No...Remember Boiled water is less dense than oil/
The oil will sink to the bottom of the container/....producing no effect.

I tried to make a large clear ice cube yesterday, but the result is not satisfactory.

I used a 20 cm x 30 cm aluminum roasting pan, and put 4 l. of distilled water in it. I covered this pan with aluminum foil, including a vent tube as suggested. This cover, of course, was not air tight. I do not think it could be made so even using a smaller ice tray. I but the pan on my stove and boiled the water. Then I put it into the freezer on top of a tray that had a couple cm thick coating of ice.

The ice close to the edge, within 5 cm say, is pretty clear, but the center still had a lot of air bubbles and was whitish.

The air bubbles in the clearer part of the ice have an interesting shape and orientation. They are thin tubes. At the bottom of most of them is a sphere perhaps 2 or 3 times the diameter of the tube. Thus they look a little like a bulb thermometer. I would expect the tubes to be oriented vertically (the air bubble trying to rise out of the ice). Most are at something like a 45 degree angle. Some tubes at the corners are horizontal, and some even go downwards. Why do the bubbles have this shape?

Some of the tubes reach near the top of the ice. I thought if they did make it all the way to the surface, or near, so that by melting the surface a little the hole would be exposed, I might be able to fill the tubes with water. Then I could refreeze the block and eliminate the hole. Melting the surface with water does not cause the tubes to fill. Assuming the air pressure in a tube is not lower than 1 atm., the air will resist being compressed by a droplet. If the surface tension of the droplet is strong enough it might act like a little beam that gravity is not strong enough to "fracture" and cause to fall. Compare pouring water into a glass (full of air) versus a beam of ice across the top. Is this explanation correct?

What if I don't want to eat the ice, is there something I can dissolve in it to react with and remove the air?

Mk
derekmohammed said:
This may sound wild but it is proven that Hot Water freezes faster then Cold water. Therefore it wold be more effiecient to use boiled water then freeze it becuase it will have less air and it will freeze faster because it is hot. Since the water will freeze faster it will have less time to allow air to dissolve in it as well.

Check it out!
http://math.ucr.edu/home/baez/physics/General/hot_water.html
Are you sure? I've heard claims that it freezes faster and it doesn't.

Chronos
Gold Member
Interesting, try leaving the vent open.

I did not pinch off the opening of the vent after putting it into the freezer. I made a cone of al. foil and poked it up through the covering sheets. The bottom of the cone may have been under the water to start, but I think it fell over because the cone did not freeze into the block.

Are you the same Chronos on SDMB?

Chronos
Gold Member
Given I don't know what SDBM means, I would guess no. [I pilfered the name from a Pierce Anthony novel]. The foil will likely collapse onto the water tray if you pinch it off fresh from the stove.

I don't think minerals matter. I saw a program on either PBS or the discovery channel about artic aquatic life. They showed a part where people were scuba diving near icebergs and showed these particular icebergs that were completely clear. These were very large blocks too and it was weird to see 9 m^3 blocks of clear ice. They said the ice was formed at certain pressures under water.

One time I filled a 10 cm*10 cm*10 cm with water to make a large ice cube. While the center was still cloudy at the bottom it was completely clear and the clear parts were about as large as a normal ice cube. I guess you could make a very large cube at once and just break off the clear parts.

One time I tried making clear ice cubes myself by using a layered technique. I figured layers would work because any time water spilled from a tray and froze it was clear. Instead it was worse and practically gave me an entirely cloudy cube.

Gokul43201
Staff Emeritus
Gold Member
Fulhamfan, you got a cloudy cube not so much because of trapped air but more likely from lattice mismatch between the different layers of ice that you grew. Each of these boundaries between layers would act as a partial reflector, cumulatively reducing the overall transmission of light to nearly zero.

Hooleehootoo, the shape and direction of the air spicules (tubes) tell you the direction of crystallization of the ice. It looks like due to the large volume of water used (from the size of the pan, assuming you had a minimum of about 1 cm of water, that's about 6 liters) the time taken to get cold was so long that that air managed to find its way back in without much difficulty.

First the water starts freezing at the walls and bottom of the tray, which are the coldest. Then the freezing (or crystallization) proceeds inwards and upwards towards the center and the surface. As the ice forms the air dissolved air gets pushed into the remaining water till beyond some point the volume of water remaining is insufficient to dissolve the air (ie: the concentration of air exceeds the solubility limit). At this point, the air starts "precipitating" as little bubbles. The first bubbles start to form at the ice-water boundary, creating little pits or craters in the surface of the ice. These pockets of air serve as nucleation spots for more air to come out as the ice proceeds inwards and upwards, creating less and less available water. So as the ice crystallizes, , it leaves behind deep and narrow wells oriented along this direction. Non-uniformities in the local freezing rates and thermal fluctuations cause these long wells to close up at some point. The more stable and undisturbed the system is, the longer will be these tubes of air. Eventually, the last portion of water - near the center of the surface - also freezes, just after the final traces of oxygen bubble up.

If the system is perfectly undisturbed (this is impossible - air flow, vibrations from the compressor, etc. cause disturbances) then you would expect the direction of the tubes to be such that they point towards the nearest wall (since the directions of crystallization is inwards from the walls). Near the surface, a little inside the midpoints of the side walls, the tubes should be nearly horizontal and perpendicular to these walls. The closer you go to the center of the tray, the greater is the likelihood of seeing vertical tubes. As you move outwards from the center, the tubes will have a greater and greater tilt, becoming nearly horizontal as you approach the sides. Just inside the corners, you would expect to see tubes which are nearly horizontal as well as pointing towards the corners (in the plane of the tray). At the center itself you may not see tubes at all. As the freezing approaches the center, the volume of available water decreases exponentially, causing the air to bubble out very rapidly, not giving them enough time to find the ideal spot.

Is this at all close to what you see/saw ?

PS: If you had a container that was deep as well as wide, there would be a greater volume of bubble-free ice.

Last edited:
Chronos