Does movement affect H2O freezing temperature?

[Chris J]

TL;DR Summary

Many say water freezes at 32F / 0C even if it's moving. Is that correct?

When water is moving does the actual freezing temperature change or does steam etc not freeze because warmer water keeps getting brought to the surface?

Or is there something else going on that lowers the freezing point as long as the water is in motion?

There are several reasons for my curiosity but the main ones are :

The first would be fog that doesn't appear to freeze when I would expect it to. My understanding is humidity/fog is water droplets in the air rather than it being in gas form. So shouldn't it freeze around 32F / 0C? Yet it doesn't seem to freeze even well below that.

Second, I stopped covering my swimming pool in the winter and instead, let the pump run continuously until the water was very cold and all of the leaves had fallen. I then disconnect it all and drain it and pretty much that night the pool freezes. When I do this the water temperature is usually very close to 32F. One year I had the water just a few tenths above 32F and I became very nervous because I thought once the water hit 32F the entire pool would become a block of ice since it's all circulating and mixing thus no warmer water below the surface. Perhaps this was a ridiculous fear? Will water in a large tank being circulated freeze at or close to 32F / 0C or does the movement lower this?

If movement does lower the freezing point why is this? What's going on to cause it and how much of an effect does it have?

 

[jbriggs444]

One year I had the water just a few tenths above 32F and I became very nervous because I thought once the water hit 32F the entire pool would become a block of ice since it's all circulating and mixing thus no warmer water below the surface.

It will not freeze all at once. There is a latent heat of fusion. This is about 334 kilojoules per kilogram. Or 80 kilocalories. Which is the heat that would otherwise be required to lower the water temperature a further 80 degrees Celsius!

When the water starts freezing, it will need to happen at the surface. That is where heat is leaving the system.

Flowing streams do freeze. As a boy I did a lot of ice skating. One can see ripples in the frozen ice that seem to mirror the pattern of ripples on the fast moving water that had frozen. You can also see that it freezes incrementally with the ice closing in from the sides on the place where the water flows most rapidly.

It is somewhat interesting to note that the ice from the fast moving water tends to be clear while the ice from slow moving water is often cloudy. I assume that there is a good explanation for this in terms of the way that dissolved gasses are released when the water freezes.

 

[Greg Bernhardt]

Flowing streams do freeze.

Is it theoretically possible for water to be moving so fast (or churning) that makes it impossible for ice crystals to form in even the coldest temperatures? I think parts of the Niagra falls will freeze, but not the heavy fast flow sections.

 

[jbriggs444]

Is it theoretically possible for water to be moving so fast (or churning) that makes it impossible for ice crystals to form in even the coldest temperatures? I think parts of the Niagra falls will freeze, but not the heavy fast flow sections.

If you know the shear rate and the viscosity then you can compute the volumetric rate of heat production. If you know the flow volume, you can recover a heat production rate. If you know the flow's surface area, you can compute an areal heat flow rate. If you know something about the adjacent volume (overlying ice or air) then you may be able to decide whether the net outbound heat flow exceeds the rate of heat production.

But this is just me thinking about the situation from first principles with no body of evidence to cite.

 

[Chris J]

It will not freeze all at once. There is a latent heat of fusion. This is about 334 kilojoules per kilogram. Or 80 kilocalories. Which is the heat that would otherwise be required to lower the water temperature a further 80 degrees Celsius!

When the water starts freezing, it will need to happen at the surface. That is where heat is leaving the system.

Flowing streams do freeze. As a boy I did a lot of ice skating. One can see ripples in the frozen ice that seem to mirror the pattern of ripples on the fast moving water that had frozen. You can also see that it freezes incrementally with the ice closing in from the sides on the place where the water flows most rapidly.

It is somewhat interesting to note that the ice from the fast moving water tends to be clear while the ice from slow moving water is often cloudy. I assume that there is a good explanation for this in terms of the way that dissolved gasses are released when the water freezes.

Thank you for taking the time to respond!

Can you explain the line in bold? I understand latent heat.
But I don't understand the part about an additional 80C.

 

[Borek]

But I don't understand the part about an additional 80C.

Just a reference to show relative amounts of heat involved: to solidify water you need to get rid of the amount of latent heat that is equivalent to cooling water from 80 °C to 0 °C.

 

[Chris J]

Just a reference to show relative amounts of heat involved: to solidify water you need to get rid of the amount of latent heat that is equivalent to cooling water from 80 °C to 0 °C.

Ah,
So similar to turning water into steam taking an incredible amount of heat yet not increasing it's temperature.

To go from water to ice will take a huge amount of cooling, not at all the same as dropping the water temperature a few degrees.

Understood.

So my fear of the pool freezing was foolish. It wasn't even remotely close even at 32.7F or whatever temperature I measured that night. I took a sample and put it in the freezer and it started to freeze at something like 32.4F so I knew the thermometer was close enough.

It was close temperature wise, but not heat wise. I'm guessing 37,854 liters isn't going to drop that much energy easily especially if the surrounding air is only 7C lower.

 

[jbriggs444]

So my fear of the pool freezing was foolish.

It is wise to fear in the absence of certain knowledge.

 

[russ_watters]

If you know the shear rate and the viscosity then you can compute the volumetric rate of heat production. If you know the flow volume, you can recover a heat production rate.

My interpretation of the question is that it was about weather the motion itself inhibits crystal formation, not about whether the motion creates enough heat to offset the heat loss. I have thought about this before as well and wondered if that was the case. I may have to do some testing of this with a stir plate in the freezer.

 

[Chris J]

My interpretation of the question is that it was about weather the motion itself inhibits crystal formation, not about whether the motion creates enough heat to offset the heat loss. I have thought about this before as well and wondered if that was the case. I may have to do some testing of this with a stir plate in the freezer.

I wonder if setting up a small closed loop system with some small rubber tubing and a tiny pump would be a good test in a freezer. If water moving through a centrifugal pump and through some tubing at a decent velocity freezes, it would suggest movement alone isn't enough.

But I'm not sure how much if any heat that would be creating?

 

[jbriggs444]

But I'm not sure how much if any heat that would be creating?

Ideally, 100 percent of the input electrical power would eventually manifest as heat in the water. In practice, I'd expect darned near 100 percent.

 

[russ_watters]

Ideally, 100 percent of the input electrical power would eventually manifest as heat in the water. In practice, I'd expect darned near 100 percent.

For larger pumps yes, the motors can be well above 90% efficient. For smaller ones the efficiency can vary greatly. But the motor nameplate should tell. This will be a confounding factor if I do the stir plate in the freezer test because there's no good way to measure the efficiency of the stir plate drive system. What I can measure accurately is time and temperature. I'd expect agitated water to cool faster than still water even with the added heat input. What I'll want to find out is if there's more sub-cooling of the moving water than still water.

 

[Chris J]

For larger pumps yes, the motors can be well above 90% efficient. For smaller ones the efficiency can vary greatly. But the motor nameplate should tell. This will be a confounding factor if I do the stir plate in the freezer test because there's no good way to measure the efficiency of the stir plate drive system. What I can measure accurately is time and temperature. I'd expect agitated water to cool faster than still water even with the added heat input. What I'll want to find out is if there's more sub-cooling of the moving water than still water.

Motor heat aside, because I guess it could also be a submerged pump in theory....
But, assuming you're delivering 90% into the pump, where does that 90% actually go? Does it all go to friction losses as heat in the end?

 

[russ_watters]

Motor heat aside, because I guess it could also be a submerged pump in theory....
But, assuming you're delivering 90% into the pump, where does that 90% actually go? Does it all go to friction losses as heat in the end?

Yes (in a closed system anyway).

 

[snorkack]

It will not freeze all at once. There is a latent heat of fusion. This is about 334 kilojoules per kilogram. Or 80 kilocalories. Which is the heat that would otherwise be required to lower the water temperature a further 80 degrees Celsius!

When the water starts freezing, it will need to happen at the surface. That is where heat is leaving the system.

Doesn´t actually have to.
If the water is flowing rapidly, well mixed and at or below freezing point throughout, the ice crystals can form wherever they find seed surfaces... and air surface is not a very good seed surface compared to bottom. Also once formed, fine snow slush has buoyancy but since it is fine grains, turbulence will mix both ice crystals from surface and sediments from bottom into the water mass. With the result that ice sticks to surfaces and objects in the bottom, where it is called frazil ice.
Floating frazil ice and slush thus sticks to surfaces - bottom, banks, ice edges.

 

[Mike in NC]

An excellent discussion! Yet another factor is the temperature of the ground around the pool. My understanding is that below 3 feet, the ground is a fairly steady temperature, and unless we are dealing with a permafrost situation, it is going to be above the freezing point of water. Therefore, the air would have to remove heat faster than the pump and ground are adding it.

 

[Tom.G]

​

1. Yes, the Falls have really frozen over!​

You’ll have to go back in time to find when the Falls truly froze over — to 1848

https://niagara.imgix.net/images/main-images/NFSP_058.jpg?auto=compress,format&fit=max&position=50 50&q=80&w=1024&s=ca05d3f668094e4d833b82c6028e740f

from: https://www.niagarafallsusa.com/the-source/do-the-falls-really-freeze-in-the-winter/

(found with: https://www.google.com/search?hl=en&q=does+niagara+falls+freeze)

Cheers,
Tom