General > General Technical Chat
Freezing Speed of Hot Versus Cold Water
bostonman:
I need to read these replies deeper and check out the links, but wanted to expand on a few things.
One person stated my graphs don't make any sense, others commented about wind, etc... I understand many factors play into temperature from the obvious differences to minute molecular physical changes. Was my experiment ideal? Absolutely not, but it was enough to get a visual on what happens. By the way, the temp loggers I used were: Elitech URC-4 Temperature Logger taken at the quickest polling rate of 10s.
The reason the data got "cut off" was to bring the items inside because it was getting late. Both cookie tins were frozen and both meters showed 32 (or nearly 32). The experiment was to see which one froze quicker and since both were at "32", they both were full of ice, etc... I brought them inside.
My experiment was to see which froze faster, but, due to too many anomalies, including the error from the meters, all I could do is see which reached "32" the quickest. From reading over the years, I expected to see a significant difference such as the cold water still being liquid and the hot water being ice.
Also, maybe this was already confirmed or rejected (again, I need to read through the responses better), but my understanding is that the greater the temperature delta, the quicker something will change to become equal, but will slow as it reaches the other temperature of the other object. As an example, if I came in from a snow storm and removed my wet coat, it will try to reach room temp quickly in the beginning, but will slow the closer it gets closer to the ambient temp (and I think ideally it will never reach the "exact" room temperature due to what occurs at the molecular level).
My graph shows this happened: the hot water cooled very quickly and slowed as it approached 32 while the cold water had a much slower decrease because it was closer to ambient. Again though, my experiment wasn't meant to be specific and exact, it was to see which cookie tin was still liquid after the other became ice.
The more I thought about things, the more I realized, there is little difference between where the hot water temperature begins. If I heated the water to one degree F hotter than the "cold" water, it would still take just as long to freeze from that temperature as if it started at 200 degrees F and decreased to 1 degree F hotter than the cold water.
My conclusion (excluding for a moment what happens at the molecular level): with a typical cold environment such as a freezer or a cold winter night, if I want to make ice quickly, starting with hot or cold water isn't going to make a significant difference as to when ice is achieved.
My graphs show that (assuming the meters are 100% accurate) that the hot water dropped to 32 while the cold was 32.1, however, both were almost a chunk of ice. I'll read through all the replies, but wanted to clarify that I'm not taking my measurements as fact, but more of a rough visual and didn't exactly get the results I expected.
Siwastaja:
--- Quote from: bostonman on February 19, 2022, 03:13:15 pm ---One person stated my graphs don't make any sense
--- End quote ---
Your graphs are completely as expected and make perfect sense: it shows that the cold sample starts freezing significantly earlier than the hot sample. It's just that your trace ends at some point during freezing. Neither sample is fully frozen at the end, because both end up at 32F plus minus tiny amount of noise.
The only thing that did not make sense are your comments; it's like you are looking at something completely different than what you posted?
Actually deciding when the freezing is complete is quite difficult. The ice does not conduct heat very well and convection stops, so once the water is partially frozen, you don't know how much of it is frozen. If the sensor is in the middle of the tin, chances are high a good indication of being totally frozen is temperature reading starting to drop again, significantly below 32F. This would overestimate the time, though, because heat transfer through the block of ice is slow. Do place the sensors consistently, in the exact same place, in the two tins to compensate for this.
You can't visually see if a block of ice is fully frozen. Try to drill a hole in it and liquid water may spurt out from the middle. State change is difficult because it happens at constant temperature: you can't see from temperature measurement how much frozen it is. The only way you surely know is to place gazillion of sensors and verify all of them measure below 32F (significantly enough to account for noise and inaccuracies in sensors, and the freezing point itself).
rfeecs:
The article from the video:
"Questioning the Mpemba effect: hot water does not cool more quickly than cold"
https://www.nature.com/articles/srep37665
--- Quote ---Abstract
The Mpemba effect is the name given to the assertion that it is quicker to cool water to a given temperature when the initial temperature is higher. This assertion seems counter-intuitive and yet references to the effect go back at least to the writings of Aristotle. Indeed, at first thought one might consider the effect to breach fundamental thermodynamic laws, but we show that this is not the case. We go on to examine the available evidence for the Mpemba effect and carry out our own experiments by cooling water in carefully controlled conditions. We conclude, somewhat sadly, that there is no evidence to support meaningful observations of the Mpemba effect.
--- End quote ---
Rick Law:
--- Quote from: bostonman on February 19, 2022, 03:13:15 pm ---...
Also, maybe this was already confirmed or rejected (again, I need to read through the responses better), but my understanding is that the greater the temperature delta, the quicker something will change to become equal, but will slow as it reaches the other temperature of the other object.
...
The more I thought about things, the more I realized, there is little difference between where the hot water temperature begins. If I heated the water to one degree F hotter than the "cold" water, it would still take just as long to freeze from that temperature as if it started at 200 degrees F and decreased to 1 degree F hotter than the cold water.
My conclusion (excluding for a moment what happens at the molecular level): with a typical cold environment such as a freezer or a cold winter night, if I want to make ice quickly, starting with hot or cold water isn't going to make a significant difference as to when ice is achieved.
...
--- End quote ---
re: "...maybe this was already confirmed or rejected...my understanding is that the greater the temperature delta, the quicker something will change..."
Yup, confirmed by various replies by multiple people already.
re: "...starting with hot or cold water isn't going to make a significant difference as to when ice is achieved..."
That is the "latent" heat. The "latent heat of fusion" of water to ice (phase change) is much greater than heat capacitance of water.
- It takes 1 calory to heat 1 gram of water by 1 degree C.
- It takes 80 calories for 1 gram of water to phase-change between ice and water (latent heat of fusion).
- It takes 540 calories for 1 gram of water to phase-change between steam and water (latent heat of evaporation).
For 1 gram of water to phase-change between water and ice, it could change the temperature of water by 80 degrees C. That is why I was so critical of the lack of definition of "freezing" when "Mpemba's observation" was made -- it takes 80x the heat exchange of dropping 1 degree verse for it to fully freeze. When big things are wrong, the details hardly matter - this is the scale of worrying about minutes when hours are wrong.
By the way, interesting to bring up one point here. Calories (with capital C used for food) is 1000x calories (lower case C used in Physics). Some "food science" people must have needed glasses when they first adopted the Physicist's unit of calories. Someone must have mixed up g and kg by mistake. So 1 Cal heats 1kg of water 1 degree in food "science" whereas 1 cal only heats 1 gram of water 1 degree. Your bottle of Coke at merely 240 Calories is really 240,000 calories! (No wonder I am getting fatter 1000x faster than I feel I should.)
Siwastaja:
Just forget about calories. They have no place in science or engineering.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version