Why do ice cubes crack in drinks? (SLOW MOTION)

Dropping ice cubes in water (and liquid nitrogen) produce interesting results. Differential expansion.

Some sound design courtesy of Alan Stewart.

[Noughtilus]

I'm Australian, and let me tell you, we use a lot of ice in summer, but I have never seen ice crack so dramatically that the two sections separate.

I suspect this is because our harsh climate and dangerous wildlife has resulted in a stronger, hardier breed of ice.

[bit_pineapple]

1:16 "And the reason is, his freezer is cold". Would have been comedy gold if the video just ended there.

[OurLifeWithFood]

This man LOOKS like science. I love it.

[xjamac]

I often noticed this effect, but I never thought about it so thoroughly. I just argued that it might be related to temperature difference, but never thought about ice expanding and stress building up in the ice. By the way, I never had an ice cube cracking in two parts when dropped into water...

The shot of ice going into liquid nitrogen is great!

[KaWouter_]

My dearest professor,

This explanation was one of the most greatest things I've ever seen. This makes me so glad that I've studied biochemical engineering. I just graduated last week!

[Bruhgono]

This video with the ice cracking sound effects was great!

[atcaleb]

So why didn't we drop the cube at the end of the video back into warm water?? Love your work on youtube!

[SardiPax]

Given ice is less dense than liquid water. Doesn't that mean that as it warms, the ice contracts rather than expands?

[stv]

When ice warms up it does not expand but contract. The Ice on the outside (-10 > 999.8 kg/m3) has a higher density than the ice on the inside (-20 > 993.5 kg/m3). Therefore the ice on the outside contracts faster than the ice on the inside, causing the inner ice to break since it cannot escape. So the reason for the cracking is contraction, not expansion as stated in your video.

[GustavTallberg]

Doesn't water ice expand when it gets colder and contract when it gets warmer and not the other way around like the professor said?

[bulwynkl]

The modes of cracking ARE different between the cold ice in water and ice in LN2. In the first case, the crack initiates in the core of the ice, in the second it's the surface that cracks.

why does this make a difference - 2 things.

1) the minimum flaw size for crack propagation inside a body is twice as big as that required to initiate a crack at a surface.
This means that the amount of stress required to create a crack is much higher in the core than at the surface.

2) the stress field inside the core of the ice cube is tri-axial - equal in all directions - versus in 2 direction at the surface - i.e. no stress through the surface.
Notice that in the first case, one crack propagates and completely breaks the ice cube. In the second, lots of surface cracks propagate into the cube, but stop about 1/4 of the way in - the cube does not actually break.

[carlosmora]

The simplest experiments are the most meaningful. Thanks for another great video.

[lovealphaomega]

Typing for my child. Hes still learning.
~I'm 7. I like your fluffy hair. I had the exact same questions last summer, so I tried this experiment with my mum. She was happy to help me and asked me what I thought. Mum just said "sounds about right to me!" "My baby is so observant, lets make lemonade". We had no idea if my answers were correct or way off. So.. today my mum was on you tube looking at other stuff when I saw your video. Asked her to play it for me. ALL of my answers were correct!~
My child loves science. He likes watching college level chemistry, tech, physics and astronomy videos since he was 3. He always gets EVERY answer correct for those "free give away" science questions lol. If he got to the videos on time, we would never see an end to the coming of the freebies & have to put him on freebie restriction lol. Glad we found this channel. It will help him to understand the terms & etc. more in depth.

[Cross]

Nice! Been wondering about this myself. Thanks for it, Brady!

[CookingWithCows]

but what happens when you put the now super cool cracked ice cube into normal water? It will probably get bigger because it freezes the water around it and then melt, but first, will it crack more?

[WilliamBoothClibborn]

Your opinion on who is the greatest genius in the BBC focus magazine is very well thought through and i am happy you did not go for an obvious one  and i completely agree with you

[Pumbear]

I love this video. It shows how science starts with a simple observation (ice cracking in a glass of water on a warm day). And then after the initial experiment it shows how the initial observation forms the basis for a new expermint that you wouldn't thank about in everyday life (ice cracking in a glass of liquid nitrogen)(on a warm day)

[wackattack]

 Water actually expands when it becomes a solid... So by melting it faster in water, the core drops in temperature, starting a expansion crack making that popping sound you hear from the cube...   
 

[yarone]

I think the conclusion might be wrong. By what they explained any thermal isolator should crack when heated or cooled.
What I think is happening is that the cubes are not frozen solid. When you put water in the freezer it quickly freezes the outside then the inside cannot expand and freeze so it gets to becomes a fluid under the melting point due to pressure. As the cube is heated the inner fluid expands quickly and breaks the outer crust.
When cooled the same happens - the inside expands cracks the outside but immediately freezes - that's why the cube stays intact and does break up like in water.
If you heat up the cube before placing it in water the inside has time to freeze and expand as the outer crust warms up. The cube stays frozen as the expansion is endothermic.
To test this theory create ice cubes that are frozen slowly or stirred during freezing and see if they break

[ManuelAntonioMercado]

The reason ice shatters "more dramatically" when submerged in liquid nitrogen is that it undergoes a phase transition at around -100°C and rearranges from its "normal" hexagonal lattice to a cubic lattice.