I searched for "mass" and "dens" and didn't find anyone mentioning this, so: Note that hot water is less dense than cold water. Therefore, an "equal volume" of hot water contains less mass than the same volume of cold water. The article mentions 35°C vs 100°C; going by https://www.usgs.gov/special-topics/water-science-school/sci... and interpolating, it looks like the density difference would be about 3.6%. That seems worth mentioning, even if it's not a large effect.
Also, if some of the water starts at 100°C, then presumably some of it will evaporate, and one should measure how much remains... The article does mention this possibility, though it doesn't evaluate or quantify it.
A more mundane explanation is that hot water evaporates faster than cold, decreasing its volume and thus the time it takes to freeze.
[Edit - Cannot delete my comment now. As 'ketzo' points out below my point is not same as OP's. OP points out that hot water is less dense. Maybe less dense liquid cools faster]
The first seems compelling to me, less mass means less over all energy to freeze.
The second less so. Hot water has a bigger surface area, which makes heat loss fast at first. When it gets to cold water's temp it should have the same surface area as the cold water making the advantage disappear.
It's true for first-order phase transitions but not for second-order phase transitions. The article actually talks about it further down:
> the Mpemba effect could happen through a related mechanism that Raz has previously described with Lu in systems that undergo a second-order phase transition, meaning that their solid and liquid forms can’t coexist at the same temperature. Water is not such a system (it has first-order phase transitions),
It would be pretty wild if they were immediate! But.. if you keep it moving at say -1, will it still eventually freeze? I struggle to imagine but don't know why really. Maybe it means something like that. Not that I know why that would be unique to water either.
Water isn't strange. It's state changing is what's interesting. That's where the most energy is converted. From latent hidden energy instead of specific heat.
I got into this rabbit hole once and remember having the same thought, I checked some paper and they used the same mass as measured when putting water into the freezer. AFAIR it was also tight, closed container so evaporation shouldn't have a big impact (checking mass after sure seems like a good idea).
Intuitively hot water should be much slower to freeze. It takes quite a bit of time for it to get to the same state that the cold water is at the start. After that time, the way we look at it, it's at the same point in experiment as if we have just put the colder water in. So we seem to still be missing something.
> After that time, the way we look at it, it's at the same point in experiment as if we have just put the colder water in. So we seem to still be missing something.
Unless there is very substantial mixing, the container of hot water will never be at the same state as the container of cooler water.
We did this in high school, with 3 open-topped 10L buckets, one with 25C water, one with 95C water and no mixing, and one with 95C water sitting on a magnetic mixer. They were not in a freezer, but outside at ~-25C.
The freezing order was 95C unmixed, 25C, 95C mixed. We also had a few temperature probes in the 95C unmixed bucket, and found that about halfway through, the temperature of the water in the middle about 1/3rd of the way from bottom was sitting at 0C, while the temperature of the water at middle at the top was still >70C.
Our final findings were that:
1. The buckets cool mainly by evaporation, which is only happening at the top, and which is much faster for hotter water
2. Water of different temperature stratas can be surprisingly stable even in a small container
3. In the unmixed bucket a stable, slow flow of water down the sides and up the middle forms. New hot water emerges from the middle, gets cooled and pulled to the sides, and then pulled down to the bottom. This circular flow is probably caused by the fact that water at the sides get initially cooled a bit and so it sort of biases the flow.
But now there's people saying it still works with closed-top containers, so I dunno. My physics teacher clearly did this every year and knew what was going to happen. The class came up with those findings after some very specific questions by him.
Very cool finding with the flow. It explains why when you take some round container and freeze water in it, the ice will be crystal clear at the edges, but it will not be crystalized so cleanly in the middle (turbulence once edges are frozen)
Who knows, perhaps there's even some laminar flow happening at the edges :)
> Ideally it would be in a pressure regulated space.
That probably wouldn't do much to help us answer the question of what will happen in our own freezers which aren't pressure regulated. While I'm glad there are and will continue to be all kinds of investigation and exploration around the Mpemba effect what I (and many others) really want to know is if heating water first would help with real world situations like having ice cubes ready in time for an event.
I can accept that it might not always be the case, but considering that at this point we can't say for sure if the effect is real at all it'd be nice if we could get it settled that it worked often enough and potentially decreases freezing time enough to be worth the time and trouble of microwaving our ice cube trays even if all the details and underlying mechanisms aren't well understood.
I should probably just start trying it myself and see what I come up with on my own, but even imperfect lab experiments would be much more accurate and trustworthy. I regularly do things like heat a mug full of water in the microwave, get distracted, and then forget all about it for hours.
> if heating water first would help with real world situations like having ice cubes ready in time for an event.
If the mechanism is that 10% of the water evaporates away, allowing the remaining 90% to freeze smaller by virtue of its reduced mass, the take-away for making ice in a hurry is not to use hot water. It is to use cold water, but a little less of it.
A difference that large should show up in the resulting cubes though right? You wouldn't even need to figure out exactly how long they took to freeze or which tray froze first, just put a heated water tray and a cold water tray in together and compare them after waiting long enough for both to freeze. That'd give you an idea of how much more/less water you'd need to compensate.
Honestly, smaller ice cubes is probably your best bet if you need ice in hurry either way.
> That probably wouldn't do much to help us answer the question of what will happen in our own freezers which aren't pressure regulated
It would do a great deal to answer the question. Knowing the base physics (or at least the outcomes) then allows further experimentation and calculation to answer more complicated (less controlled) conditions.
> I regularly do things like heat a mug full of water in the microwave, get distracted, and then forget all about it for hours.
When we did this as an experiment in high school, we started with the same volume at room temperature. 500mL was brought near boiling, and another 500mL while the control was left out.
Density is going to factor in somewhere, sure, but when you have 100 -> 0 vs 35 -> 0, that's almost 3x more energy (given constant specific heat) you need to sink somewhere. a single-digit percentage is a drop in the ocean here.
ANother aspect would be that the hotter an atom is the faster it will vibrate. So would it be that a vibrating atom is able to transfer thermal energy quicker akin to comparing a warm start to a cold start in a way that see's the thermal transfer inertia at the aromic level has better initial acceleration over the colder atom. So even when the atom reaches the vibrational speed of the colder atoms, the momentum of thermal transfer is already accelerated. Sorry if that sounds fluffy but a bit of a theory. One which trying other liquids and testing would give more insight.
I don't think there's such a thing as thermal transfer inertia, at least at the level of individual atoms (or molecules) within a fluid. In the inertial reference frame of the fluid, each molecule just has its own kinetic energy (and maybe excitation of electrons or whatever) representing its current temperature; I'm not sure what physical state at the molecular level could capture "how rapidly the temperature had recently dropped".
At the macro level, one can come up with ideas. Convection currents within the fluid, and in the air touching the fluid, maybe; though, say, once the hot water has cooled from 100°C to 35°C over a period of—what, ten minutes?—I'm pretty skeptical that there would be so much inertia in those currents that they'd persist, and persist strongly enough to accelerate the 35° to -1° cooling, enough to beat the head start of the water that started at 35°.
Maybe the initially cool water forms some uniform layer of very-cool water at the top (which, being very cool, doesn't exchange much heat with the very-cool air above it; an ice layer would be an example of this), which is held together by surface tension or something; whereas with initially hot water, that layer is not uniform and there's more mixing (and constantly-created convection as a result)? I have no idea if any of that is realistic. If so, it would suggest that shaking a cup of cold water (perhaps after a few minutes in the freezer) would work as well as having it start hot.
... After writing the above, I saw that (a) infogulch below has the same idea, (b) Wikipedia has an anecdote appearing to confirm it[1], and (c) the article doesn't seem to mention shaking, stirring, or otherwise agitating the cup.
[1] The Scottish scientist Joseph Black investigated a special case of this phenomenon comparing previously-boiled with unboiled water;[9] the previously-boiled water froze more quickly. Evaporation was controlled for. He discussed the influence of stirring on the results of the experiment, noting that stirring the unboiled water led to it freezing at the same time as the previously-boiled water, and also noted that stirring the very-cold unboiled water led to immediate freezing.https://en.wikipedia.org/wiki/Mpemba_effect
Good instinct to try stirring. Search the comment thread (or the Wiki page for "Mpemba effect") for "Joseph Black", and you'll find something intriguing.
Agreed, if it's about 'is this the case' especially in home freezer context then it's volume that's interesting: ice cube tray sized volumes. (But also are the resultant cubes the same volume? Presumably not.)
If it's 'yes this does happen, why?' then it seems right to try same mass, to test that theory.
How about this theory. The hot water causes the freezer to run its compressor sooner. The system is nonlinear due to intentional hysteresis in thermostats, and the excess energy stored in the hot water is trivial compared to the capability of the heat pump.
Introducing hot water into a freezer rapidly raises the ambient temperature of the freezer. This causes the thermostat to click on, and the heat pump therefore runs until the ambient air drops down below the thermostat's shut off point. A lot of heat is removed very early on, and the ambient freezer temp is at its lowest going into the liquid to solid phase change.
With the cold water, the ambient air of the freezer doesn't rise nearly as much. The thermostat does n't click on until a long while later, once enough heat leaks out of the freezer. The ambient air is warmer going into the phase change.
It doesn't work if you put the two cups in the same freezer, but at least it could explain some anecdotes.
I think it might have something to do with the convection current of the hot water. Perhaps the higher initial velocity allows the ice crystals to nucleate faster?
Hot water melts snow, and the ground is much colder than the snow which isolates heat. So in your case the simple explanation is just that, the hot buckets melted the snow/ice and therefore stopped being isolated against the cold ground so froze faster.
The assumption being they were placed on snow, which I would argue is an unlikely assumption to hold up. We have much more days with freezing temperatures than days with snow lying around and even then with lots of snow there are enough places that don't have any (like a balcony).
Some more ideas. Cold water holds more dissolved air in it than hot water, and the hot water heater would have recently rendered most of those gases out. That dissolved air could negatively impact the thermal conductivity of the water.
Near boiling water is about 4% less dense than near freezing water. If you filled the buckets right to the top, there would be less water in the hot bucket.
How about this theory. The hot water causes the freezer to run its compressor sooner.
This may happen but, in a normal freezer, the action of the compressor is meaningless.
My feezer is 70% full, with probably 30kg of stuff in it. This stuff is mostly at -22C (I have a thermometer), and so, this mass is what cools new things added to the freezer, fast.
(Why -22C instead of -18C? Salt water fish is in there, and I also want a little headroom.)
Unless the new things are physically touching the other mass, then my guess is that the primary cooling effect is due to 1.) convection through the air in the freezer, followed by the 2.) conduction through the surface the new things are sitting on, followed by the 3.) IR radiation of the new thing. The existing thermal mass in the freezer interacts with the new thing via 1 (indirectly) and 3 (by absorption), but I suspect that the compressor cooling the air is a bigger effect.
All documentation I've read, and personal experience, says otherwise.
For example, most manuals urge one to not have an empty freezer. That cold mass is one reason why.
When I was a kid, we had a fridge from the 50s in a cottage. When I started drinking, we'd put 5 cases of beer (24*5) in glass bottles in there, and 2 hours, yes hours later they'd be at 1C.
Now, modern fridges/freezers literally do not have that degree of cooling power. This is on purpose, for it is more efficient to power a small compressor all the time, than a massive compressor for 10 minutes.
You may want to argue this point, and that's fine, but I am merely providing info both from the manuals of modern freezers and fridges, which I have read, and from online when looking at why they are so bloody slow to cool things.
My LG fridge manual actually says not to put warm meat in my freezer, unless the thing has loads of frozen stuff in it.
Otherwise the meat could go bad before freezing.
Oh, another wonder of modern fridges. If you buy one and put it outside, or in an unheated garage, the freezer becomes useless.
This is because many fridges have no thermometer in the freezer part, and only get a reading when the compressor comes on, to cool the fridge.
As the fridge is always cool when it is 3C outside, or cooler, the freezer never keeps stuff frozen.
Oh, I totally get the argument to keep it full. Both the colder thermal mass and less cold air to escape make it considerably more efficient. I just doubt that the thermal transfer from the mass to the newer items is as significant as the compressor kicking in. After all, the compressor is going to flood the internal radiator with liquid far below the ambient temperature of the freezer.
> My LG fridge manual actually says not to put warm meat in my freezer, unless the thing has loads of frozen stuff in it.
That sounds silly to me. In my experience ice cubes freeze within an hour and meat within a couple hours.
You're putting a lot of faith into manuals for mass produced household appliances. All that holds true to a first order, but the actual thermodynamics going on is certainly way more complex than what's captured in a consumer facing manual. When discussing the relative freeze times of a modest amount of water for the purposes of a science experiment, the transient higher order effects are going to be an important factor.
The compressor is moving the air around in the freezer, which can have an impact on how much heat is transferred. You can sit comfortably in a sauna, but your arms will heat up quickly if you wave them around. Similarly, a convection oven heats things faster than a conventional oven.
True, but I meant that fan runs when the compressor is running to blow air over the coils. That movement of air will allow more transfer of heat from the stuff in the freezer.
You've got a good point, which is that moving air greatly improves heat transfer, but your errors in terminology are confusing people. The fan that blows cold air is the "evaporator fan". The compressor fan (if there is one) blows hot air on the outside of the freezer. Here's a diagram: https://home.howstuffworks.com/freezer2.htm.
I love that theory. I would assume they use big enough freezer or with enough ice inside for that not to matter, but that would be a hilarious solution.
> the excess energy stored in the hot water is trivial compared to the capability of the heat pump
I don't know if I'd call it trivial, bringing 1 kg of water from 85C to 0C involves about 100W for a duration of one hour whereas a typical freezer removes heat at, what, 200W or 300W?
>bringing 1 kg of water from 85C to 0C involves about 100W
That looks like the amount of energy needed to heat that water.
But cooling things doesn't work the same way. Instead you want to extract that energy/move it somewhere else. The efficiency of some approaches, unlike when heating something, can hugely depend on ambient temperature. It's possible you'll be able to extract more useful energy cooling something than you'll have to expend!
>The abstract findings suggested that the components of a hotter system, by virtue of having more energy, are able to explore more possible configurations and therefore discover states that act as a sort of bypass, allowing the hot system to overtake a cool one as both dropped toward a colder final state.
> “We all have this naive picture that says temperature should change monotonically,” said Raz. “You start at a high temperature, then a medium temperature, and go to a low temperature.” But for something driven out of equilibrium, “it’s not really true to say that the system has a temperature,” and “since that’s the case you can have strange shortcuts.”
I gotta admit, that's pretty cool and unintuitive.
What still isn’t clear is what the the initially-hot system has “learned” at the moment it reaches the starting temperature of the initially-cool system. There must be some difference between the two states at that temperature if the initially-hot one is going to catch up with and surpass the initially-cool system.
From my reading, the point is that the glass of water doesn't really "have a temperature" any more, since the cooling transition means it's no longer in thermal equilibrium.
That is, you could perhaps model the bulk of water as having a temperature field, and clearly every point in that field passes through the starting temperature of the initially-cool system, but the gradient landscape is vastly different.
The point is that the state of the system consists of more than its average temperature. It's not just "catching up" to the state of the initially-cold water, it's going through other configurations.
The simple answer is that the cool system has a very uniform temperature (because it's been at that temperature for a long time), while the hot system that's cooling down is doing so with large spatial variations of temperature. In other words, all the water in the initially cool system is at 10°C, but the initially hot system has pockets of 5°C and pockets of 15°C.
It doesn't have to "learn" anything in order for there to be a substantial difference.
These temperature gradients obviously lead to convection.
Stronger convection means greater heat transfer, thus greater rate of cooling down.
But it'd be surprising if the inertia of the convection of the initially hot system didn't just gradually decline (because of friction) to almost exactly (little bit greater) the same level of convection (which the initially cool system had in the beginning) when it reaches the same average temperature.
Water currents can last a surprisingly long time. Pour one liquid into another of a slightly different color and the swirling mixing process can go on for minutes at least. Perhaps hours in the right circumstances.
Interesting theory for a specific mechanism: "local temperature difference-induced convective cooling". Could be falsified by giving both initially-hot and initially-cold samples a stir rod, expecting the cooling performance of the initially-cold sample to improve to match the initially-hot sample.
> The Scottish scientist Joseph Black investigated a special case of this phenomenon comparing previously-boiled with unboiled water; the previously-boiled water froze more quickly. Evaporation was controlled for. He discussed the influence of stirring on the results of the experiment, noting that stirring the unboiled water led to it freezing at the same time as the previously-boiled water, and also noted that stirring the very-cold unboiled water led to immediate freezing.https://en.wikipedia.org/wiki/Mpemba_effect
Wow! I guess that's the final answer then. Maybe to seal it, devise the inverse experiment where you somehow inhibit normal convection in boiled water and expect to see it take as long as the cold water.
It's found it's way into the "minimal valley" whereas an arbitrary lukewarm state might be closer to a "ridge".
Clearly, yes, if you could start at the magic state that would be ideal. And there should be experiments that use a bunch of thermometers too compare the time from the same average temp (the initially warmer one just gets the "running start").
Mathematically it is not hard to abstractly characterize what is going on. (This is not saying the actual physics is easy!!) Temperature is an equivalence class on fluid states, but the average time to transition between those states does not form a metric space. The failure of the triangle property shows that the composition of transitions induces non-uniform distributions within the temperature equivalence classes that subvert the expected transition time by which we had attempted to build a metric space to begin with.
Just like non-euclidian space in the 19th century, this is the sort of thing where the mathematics can say "yeah sure seems legit" before the physics stops saying "wait wtaf",
1. Temperature is monotonically related to energy content. A warmer system has more energy than a colder one, all else being equal.
2. To cool, a system must release energy. The rate at which a system can release energy is monotonically related to the difference in temperature between the cooling system and the cold sink to which its energy is being released. The bigger the temperature difference, the higher the rate of energy release (all else being equal).
For the Mbemba effect to be real, one of those two premises must be false. Which one is wrong?
> Temperature is monotonically related to energy content. A warmer system has more energy than a colder one, all else being equal.
In equilibrium, the temperature of the water is directly related to the energy content, with the heat capacity being the conversion factor.
But under the thermodynamic definition, the temperature of the system depends not only on its energy but also on its entropy. And one of the points the article is making is that even when we know the total amount of energy entering or leaving the system, its entropy may not be nearly so easy to measure or calculate when its state is far from equilibrium.
Sorry, but that is not true. The definition of the SI unit of temperature, the Kelvin, is given in terms of Boltzmann's constant, which has units of Joules (i.e. energy) per Kelvin.
I don't think either one is false. To me, 2 looks false if you consider the "system" in question to be the entire glass of water. But my interpretation is that in this case the glass of water is really many very small systems (pockets of unequal amounts of energy) all interacting with each other.
Let's say you're given a few hot potatoes and have to cool them down as fast as possible. You have a refrigerator but you can only keep one of them in it at a time. How do you decide which one to put in at any given time so they all reach temp the fastest? When the "system" involves many possible pairings of temperature differentials it feels very intuitive that some configurations would be better than others. So it removes the mind-bending thermodynamics law breaking aspect of it.
Admittedly, this isn't really what the article says! I honestly don't get how my above intuition squares with the whole energy minima thing, and so while useful as a thought exercise to at least help me entertain the idea, I'm not really sure it's correct?
IIRC (my thermodynamics is a long time ago), but 2 applies only when the two (cooling system and sink) are near equilibrium, which is not the case here.
That would imply that water once heated and then cooled is different from water never heated - at least for a time. I wonder if there is any evidence for that.
One difference could be that hot water is not as good a solvent for gas (think soda bubbles). So heated water would have shed most of its gas content which would lead to a different (faster?) freezing process, if freezing occurs faster than gas can dissolve back in.
Because this is literally about the time-domain solution to heat conduction and fluid dynamics (convention) with different initial conditions and boundary conditions (shape of vessel, temperatures at boundaries, etc.).
For example if there isn't uniformity of initial temperature, if there isn't uniformed of the applied cold sink, if there isn't uniformity of heat flux capacity due to conductivity of the interfacing boundary conditions, etc. There's even questions about the water purity or contaminants which definitely would change the answer.
These are not and often can not be controlled. And in practice at home, you will never duplicate even what the answer is in a control lab experiment.
It is the norm to have a range solutions that entirely depend upon initial conditions and boundary conditions. And there's also uncertainty of even knowing the values of these conditions.
Expecting a simplistic yes/no answer to many problems ESPECIALLY those involving heat transfer and fluid dynamics is like asking for a specific date for your date from your date of birth or your childhood medical history.
See also climate change - we DO NOT KNOW there is a specific date when things "tip" and ANYONE claiming there is a date let alone timeframe is a liar or so ignorant they can't possibly be right! The error of any model is larger than the estimate value itself.
You'd think people would have learned something from the 20th century about the FACT that we do not live in a 19th clock-work universe with complete predictability. We know how it can never be predictable especially if fluid dynamics is involved - you know, mathematical chaos and all.
Both means yes. It's pretty clear that in some circumstances cold water freezes faster. The interesting thing is whether hot water EVER freezes faster.
Having empirically tested this, I can answer that yes, in my circumstances, hot water can freeze faster.
Some years ago, I kept chickens here in Michigan through the winter. I wanted to leave them with access to water while I was gone at work, but after filling the tank in the morning, I'd often come back to a block of ice and thirsty chickens in the evening. My experimental procedure was as follows: On a Saturday when I'd be home all day, I filled two identical 5 quart poultry waters to the same level, one with 115F hot tap water and one with normal 47F cold tap water. Both came from our well, from a sink with an aerator, and both passed through a softener. The hot water had been heated in a tank-style residential water heater, not over the stove. The troughs were left outside of the coop in a shed (several feet apart) to prevent unpredictable chicken activity from messing with the results. I put thermocouples in both troughs and checked in every half hour to take readings.
The hot water froze first by several hours!
The cold one reached 32F first, and after hitting 47F the rate of change of the one that had been hot was similar to the cold one (maybe slightly faster, but my measurements were too coarse to be confident of that), but both spent quite a bit of time at exactly 32F. The hot one just spent so much less time at 32F that the couple of hours to drop to 47F didn't matter. Both still had a considerable volume of liquid water in the bulk tank, the part exposed to air froze first, but that's still a failure because the chickens couldn't drink. I didn't have any equipment to measure dissolved gasses in the water, or measure convection or stacking in the tank, measure the rate of evaporative cooling, measure the rate of cold air being pulled over the surface by convection driven by the hot water temperature differential, or otherwise test any other factors that might be different from two apparently identical 32F tanks of water, all I knew is that the hot water froze faster.
The result of the experiment was just to put a small electric heating adapter under the base. After that it didn't matter. Another tragic example of pragmatism over curiosity...
Great experiment, and thanks for writing it up. I'd encourage you to get your 10 minutes of internet fame by reproducing the experiment next winter with pictures and measurements. I have to say, despite your experience, I'm still doubtful. Or rather, I'm still almost certain that there must be something besides the temperature that is causing the effect you saw.
For example, let's say you started with the hot water, and then let it cool to 47F--perhaps by putting it outside in below freezing weather. You then ran another tub of hot water, and did the experiment between those two. It definitely seems impossible to me that the new hot one would freeze first, which leads me to believe there must be something measurably different between the cool tap water version and the cooled down hot water.
Why am I unable to find a single good video on the internet demonstrating the mpemba effect occuring? If people were really heating their water before freezing it in Aristotle's time, then this shouldn't require any precise technology to reproduce. The article addresses the fact that this seems very easy to test, but then doesn't explain why no one's reproduced it on camera.
Because it’s bullshit. People who observed this supposed effect don’t even bother to place both containers in the same fridge at the same time (so it’s not affected by the thermostat’s timing) nor do they weigh the ice afterwards.
What happens if you do that with cold water? Also, there's no indication of the water freezing. How would you tell if the water is freezing or not once it's landed in the snow?
Yes, the only videos demonstrations involve throwing the water into the air, but nothing in this article or the Wikipedia page about the effect suggest that throwing the water is necessary for the effect to occur. I don't even know where these people got the idea to do the experiment that way. Their result is way less interesting because it can be explained by surface area.
There was a video by Derek/Veritasium as few years ago explaining this and also walking through all the experiments that demonstrate the effect does not in fact exist.
For anyone interested... Derek quotes 5 theorised causes but rules all but 2 insignificant, leaving the primary confounding problems as:
1. Warm liquid changing the nature of the freezer (Derek mentions melting frost into a conducting layer but there's also warm water triggering the thermostat of a freezer to work harder)
2. Supercooling can prevent ice forming in calm water, even when the water is below freezing (the effect is wildly unpredictable leading to experimental problems)
Derek then quotes the following study and meta analysis from 2016 that carefully accounts for these and other problems, which finds no evidence of the effect and concludes the other studies' claimed effects are within their own margins of error (including issues like placing thermometers at slightly different locations in a vessel giving dramatically different timing results):
In 11th-grade 'Honors' Chemistry, we ran the 'Which water freezes faster' experiment outside in ~0℉.
2 glass beakers with same amount of water: 1 hot, 1 cold, stirring both at the same rate. Teacher told us the experiment was over once ice starting forming on the surface…
Cold water developed ice first.
Outside in that weather, no one waited to see how long hot takes to become ice.
In case you missed this in the article, the effect you described is a possible explanation of why the effect happens …
> Or perhaps external factors come into play: A layer of frost in a freezer can act as an insulator, keeping heat from leaking out of a cold cup, whereas a hot cup will melt the frost and cool faster.
What I don't get is why there is no mention of evaporating effect on the energy level of the system.
Evaporating water takes quite a lot of energy so the rest of the water is cooled:
> During evaporation, energetic molecules leave the liquid phase, which lowers the average energy of the remaining liquid molecules. The remaining liquid molecules can then absorb energy from their surroundings. This process can take place at any temperature because some of the molecules in a liquid will always have enough energy to enter the gas phase.
I always thought this was the reason when observing the effect when trying to pour hot instead of cold water to defrost your windscreen during cold winter months :)
> pour hot instead of cold water to defrost your windscreen
This sounds like a risky experiment, the windshield is likely to shatter with such thermal differences and the associated tensions.
If I am not mistaken, a windshield is created already with internal pressure to ensure it will shatter to pieces when the local pressure changes abruptly (via punching e windshield with a hard pointy object for instance)
It was definitely not boiling but hot from tap, so could be about 60°C, and, yeah, somehow the windshield did survive it. It's probably not that bad because you actually pour the water on the frosted windscreen so most of the heat would dissipate already before actually reaching the material.
When I was a kid we used hot water to defrost the windscreen regularly. A couple of hundred times over multiple winters. Temperatures around -20. No cracks or anything like that ever occurred.
When was that, out of curiosity? I seem to remember that the structure of the windshields (at least in Europe) change in the 70's from ones that would break with a crack, to the ones that would shatter in plenty of small pieces (due to the internal tension, and at least two layers)
I saw that in the 90s but people were definitely doing it in the 70s as well. I am not aware of any stories of cracks or shattered windscreens because of it, though I am sure that if you do that with boiling water and not-so-frozen windscreen often enough, you would get that result :)
This makes sense to me. Surface layer partly evaporates, it immediately cools down remaining surface layer and then convection spreads it around. It would enable quick "forced" cooling similar to AC and due to convection, it can continue for a long time as warm water and cold water rotate.
Quite counter intuitive. You would expect to have water cooling slow down when it reaches temperature of colder water.
Shortly after completing my second class in thermo during undergrad, this topic came up between with a friend. The particular situation was wether to put the creamer in the coffee and let it cool or wait until it cooled and then put the creamer in. The wait time being the same, which would produce the hottest coffee.
I got a little hyper focused and spent well into the night scrawling equations on a huge whiteboard and I don’t think I ever felt that I proved it one way or the other.
I remember being pretty defeated because I thought I had all the knowledge I needed to solve the problem. I guess it wasn’t so simple after all :)
A simple application of newton's law of cooling will tell me that for most normal ratios of coffee and creamer, you want to mix them first. Because the bigger the temperature delta between the coffee and the environment is, the more heat the coffee will lose per second. Mixing first will lower the coffee's temperature, and will cause it to lose less heat per second.
I disagree. But I'm no physicist. My reasoning goes like this, with the assumption that the coffee milk has room temperature:
The coffee cools faster if it's hot. So putting in the creamer immediately steals the most efficient cooling period for the coffee. There are some caveats that I don't consider that could make a difference like that the surface area of the combined liquid is larger so it transfers heat more efficiently.
This was actually a controversy some twenty years+ ago on a science show in the Netherlands ('nationale wetenschaps kwis').
The wanted answer was that putting creamer in first was better as the resulting fluid had lower temperature and thus cooled slower. But a given answer was that the creamer made a top layer that made the coffee evaporate slower, hence keeping more of the warmth as well.
During the show this answer was counted wrong. But (theoretically, I don't think an actual experiment was made) this is a factor working towards the same answer (first creamer keeps the resulting fluid warmer compared to adding the creamer later) that may actually have a larger contribution. So afterwards the consensus was that this may actually have been a better answer.
Why would you not consider that caveat? The cooling rate should be generally proportional to both the surface area and the current temperature delta between the liquid and whatever it contacts. Taken to the extreme, if you added the creamer and then poured the coffee out onto the floor, the creamer's mass and temperature would have relatively little impact compared to the surface area change.
For a non-sealed cup of hot liquid, the significant majority of cooling happens at the exposed surface with the air, through a mixture of convection and evaporation. Modeling the cooling as being proportional to total surface area would be pretty inaccurate, although there is definitely still some conduction through the cup/mug worth considering as well.
> For a non-sealed cup of hot liquid, the significant majority of cooling happens at the exposed surface with the air, through a mixture of convection and evaporation.
I don't know enough to argue with any confidence but this is very surprising to me. Ignoring for a moment the thermal conductivity of different mug materials, it seems like a large amount of energy would go toward heating the mug up to near liquid temperature rather quickly. Then you'd have at least as much heat loss between the mug and air (compared to exposed liquid and air).
The mug I imagine we're talking about is a ceramic mug, which I believe to have high thermal conductivity just based on what processor covers are made of. It also has plenty of mass.
If you're talking about an insulated mug obviously this changes. But just the fact that insulated mugs exist proves my point that a large amount of heat is lost through the mug...
Assume for simplicity that we mix half coffee and half creamer. This halves delta t. However the area will not double, since the top and bottom areas are the same. Furthermore, as another commenter pointed out, the top area is where most of the action happen.
However, if we really want to overcomplicate things we could consider the possibility of an insulating air pocket in a half-empty cup, leading to less convective losses. Consider a vacuum flask half full of hot coffee outside in a strong wind. If you fill it to the brim with creamer it might cool faster. Evaporation might become important too.
If you don't fill the thermos to the top, we can also add the small temperature impact of Helmholtz resonance from the wind blowing across the lip to make this more complicated. (Now I wonder how loud a sound needs to be to boil water...)
Please elaborate. Newton’s law of cooling would seem to suggest that the cooling rate is higher if the liquid is hotter. Of course, depending on the exact scenario, the creamer might also be warming up at the same time…
My reasoning is that the coffee will cool in a curve having high rate of temperature change at first, and slower later in a long tail until it reaches room temp. Adding the cream will "remove" a fixed amount of "heat units". The heat of the cup can be graphed as heat/time and it will look like exponential decay.
If you remove those units at the start, you've reduced the starting temp a bit, but you haven't much changed the long tail of the cooling. You essentially just started the coffee at a slightly cooler temperature, but this doesn't affect the curve much. Or to think of it another way, the change in temperature at the start corresponds to a small amount of X axis (time) on the curve.
If you add the cream later, the temperature reduction corresponds to a larger amount of time on the curve. This means the temperature will be lower than the above.
So to my intuition, cream first should yield hotter coffee
I did read the article and interestingly, I learnt we do not know a lot about phase transitions. Even more interesting, why would hot and cold water reach 0'c almost at the same time (according to one of the experiments)? If the rate of cooling were same, we would expect hot water to be tepid when cold water is 0'c.
Btw, I have seen similar experiments in really cold climates, where a cup of hot water thrown freezes immediately, but cold water does not.
Anyways, I am always one for running the experiments :)
> Btw, I have seen similar experiments in really cold climates, where a cup of hot water thrown freezes immediately, but cold water does not.
IIRC this is because the surface tension of hot water is significantly lower, so the thrown hot water disintegrates into much smaller droplets, which freeze more quickly.
if it were that easy to solve, the question wouldn't be still be uncertain (i'm talking about the original "does hot water or cold cool faster", to which this question seems related.)
My grade 6 "science" teacher gave me a C- when she sent us home to test if cold or hot water froze faster in ice cube trays. I reported accurately that the cold one froze faster, knowing that she was expecting us to report back "hot". I will never forgive!
Anecdote: a plumber in my family swore by this effect and always connected ice makers to the hot water line in new houses. Their justification was that even if their observation was wrong it was still one less thing that was likely to freeze in winter.
When was the last time you turned on the hot water an hour after anyone else used it, and had even the most sightly warmed water within one half to one third of a glass of water?
Never. Even the under-sink heaters hardly work that fast, and their water only has to travel like a meter at most.
The hot water has less dissolved gasses. In some cases it has less chlorine. I put water in a glass pitcher in the fridge because it tastes better after sitting at a lower pressure and off gassing. I agree that the water enters the ice maker at room temperature, but it may not be a bad idea regardless.
If it's heated within a closed system (i.e. mains pressure hot water), the gases are still there - they can only leave the water if you have an open tank, which is a bit old-fashioned. You can observe this by running hot water into a glass from the faucet: it appears cloudy, which is the dissolved gases boiling off.
This means, once the hot water within the pipe has cooled (within the pipe) it'll still have the same gas content as cold water.
I think that depends where you live. Regulations in my location changed so that hot had to have the same configuration as cold (and supposedly the same potability?). Old houses in the area have separate taps for hot / cold whereas newer houses built after the change have combined hot/cold.
If your water provider says it's potable I'm not going to question it, but I have yet to see a provider that makes such a claim about their hot water. It is difficult to make a statement that would apply worldwide, but where I live, it is easier to keep legionella out of cold water even if the infrastructure for hot and cold is the same.[1]
We are supplied cold water which then goes through a hot water system. My understanding was that the new configuration required the use of check-valves to prevent backflow in such a way as to reduce the risk of cross-contamination between hot/cold. I'm guessing this is now irrelevant because hot water system standards have changed to reduce risk of corrosion / infections occurring, but there is also a chance I was getting mixed up with UK regulations [0].
I don’t know why I think this, but I have it in my head that hot water heaters can have more mineral build up. This is why I always thought you should use cold water. Even though the hot was still drinkable, it might be harder water.
Yeah hot water can leach metals from the pipes that you really don't want to be drinking. This could either just taste bad, or add unhealthy amounts of lead and copper.
We are fast approaching a time when all houses have pex. At this point we've been using it in new construction for decades, and most houses with original galvanized pipes are either already replaced with pex, or getting really close.
Sometimes houses with water softeners will run non-softened water to the cold tap on the kitchen sink for cooking purposes, as some people don't want the salted water.
Hot water heaters frequently have sacrificial anode rods to prevent corrosion. The rods corrode over time and dissolve into the water. Lower cost aluminum rods have health concerns, not as much with magnesium rods.
See also Tom Scott's video on Britain's history with separate hot and cold taps and why hot taps sometimes had unsanitary cold water tanks to supply them: https://www.youtube.com/watch?v=HfHgUu_8KgA
Recommendations for Belgium: do not drink the hot water
For Geneva: you can drink the hot water
For France: you can drink the hot water but it is disgusting
All of them talk about the same effects (fisdolved oxygen, piping ,...) and come to different conclusions.
My take on that: nobody questions the potability of cold tap water so I will use that one, there is nothing to gain (possibly a shorter time to boil? that trumps the extra cost?)
This thread has revealed an interesting cultural difference, with me being in the minority that gets hot water from a remote provider, as opposed to having a heating system and heating it myself.
wouldn't the refrigerator have to work a lot harder in that case to remove the extra heat? doesn't seem beneficial from an energy use perspective, considering you have to pay to heat the water, then cool it again.
That was my initial reaction too, but upon consideration, how long does it take for hot water to get down the line? Typically, both the hot and cold water lines are full of room-temperature water until you run them long enough to heat up all of the piping between source and sink. I'm having trouble picturing an in-home ice maker having enough throughput for it to actually matter.
This is the reality of the matter, though there may be some minor benefit to "using up" hot water a bit more. Practically it's unlikely to matter either way.
Can someone explain why Newton's Law of Cooling isn't a sufficient explanation? The temperature delta is clearly larger with hot water, do one would expect it to cool faster, no?
Cooling faster and reaching sub zero temperatures faster are two different things. The hotter water will drop more degrees per minute/hour based on Newton, but at some point the two liquids will be the same temperature, and then you’re assigning some sort of thermal momentum that Newton doesn’t cover.
Would the warmer water not have currents present from convection? Those are a source of inertia. There is a case for the warmer liquid never actually reaching the same state the cooler liquid is in.
Other people have answered, but I think this is slightly more intuitive way to put it:
Newton's law of cooling states that the rate of temperature reduction (heat loss) of a body is proportional to the temperature delta between it and its ambient surroundings. We can write this as:
dT/dt = kT
Where T is temperature delta, t is time, and k is some proportionality constant. So the rate of cooling is changing as the delta temperature reduces, specifically, it is getting exponentially reducing as it gets closer to thermodynamic equilibrium (where no heat is exchanged):
T(t) = Ce^kt
C = T(0)
Which means there is a larger cool down with the hot water at first, but under this simple model, once the T(t) hits the same temperature of the cool water it's being compared to, it's cooling rate should be equivalent (and a lot slower).
Ok..but hot water must become cold water before it can freeze, no? Are you saying that certain glasses of cold water have a memory of having been hot, but others do not? How long ago must it have been hot for that to count?
This all just feels like sloppy measurement of some kind.
Calling it "memory" is a misleading way to think about the effect.
It could be better written as: exposing hot water to a large temperature gradient allows the water to change energy states in a way that it cannot do when that temperature gradient is small. Those fun new energy state changes allow the hot water to give up more energy faster and freeze sooner.
So no, the hot water in this case is not becoming cold water in the sense that you mean.
Do you have a background in physics or is this just stuff you read?
I have an undergrad degree in physics and nothing you just said sounds right at all. It all just sounds like gibberish. What kind of "energy states"? Molecules bounce off each other and vibrate around. The measurement of that kinetic energy is called "temperature".
I'm regurgitating what I understood from reading the article. Here are some relevant quotes:
> The abstract findings suggested that the components of a hotter system, by virtue of having more energy, are able to explore more possible configurations and therefore discover states that act as a sort of bypass, allowing the hot system to overtake a cool one as both dropped toward a colder final state.
> “We all have this naive picture that says temperature should change monotonically,” said Raz. “You start at a high temperature, then a medium temperature, and go to a low temperature.” But for something driven out of equilibrium, “it’s not really true to say that the system has a temperature,” and “since that’s the case you can have strange shortcuts.”
In 11th-grade 'Honors' Chemistry, we ran the 'Which water freezes faster' experiment outside in ~0℉.
2 glass beakers with same amount of water: 1 hot, 1 cold, stirring both at the same rate. Teacher told us the experiment was over once ice starting forming on the surface…
A cold system normally takes longer to warm up than a cool system. Yet recent theoretical studies have suggested that the reverse may sometimes be possible. Here, using a colloidal particle in a heat bath, we present experimental evidence for this inverse Mpemba effect. By carefully choosing the energy landscape, we can make the cold system heat up exponentially faster than the heating rate of cool systems. While similar behavior has been seen in systems that cool down—the Mpemba effect—we find that entropic effects generally make anomalous heating harder to observe than anomalous cooling.
It would be very interesting to do this experiment in a freefall environment, where you could release identical quantities of water at different temperatures as spherical bubbles into chilled air. This would eliminate the effects of the container, and of convection within the fluid on the effect. Interestingly, the article doesn't mention convection as a confounding effect, but it is one of the ways in which a "hot" system could explore more non-equilibrium states in a given time than a "cold" system.
I'm also convinced convection is the reason why it can happen under completely equal conditions (which are arguably missing from most experiments).
1. The larger the temperature differential, the faster the heat transfer.
2. The outer layer of hotter water will lose temperature very quickly and generate a convection flow inside the container, which would replace outer layers of now colder water with hot water relatively quickly, accelerating the heat transfer.
3. The outer layer of the colder water will lose temperature much more slowly, thus the convection will be less pronounced, not significantly accelerating an already slow heat transfer.
Of course all this will depend, non linearly, on:
1. The temperatures involved (1C water will certainly freeze faster than 99C water when put in a -25C freezer).
2. Shape of the containers (as it affects the convective flows and hence the heat transfer).
3. Total mass of water: I suspect for small (in range of up to decilitres) and large (beyond decalitres), cold water tends to freeze faster. The first, because convection effects are very limited for both waters, and second, because convection effects on cold water become significant enough to improve heat transfer.
4. Environmental conditions: relative humidity of air in the freezer, temperature of the freezer, is freezer empty, are the walls close or far, the actual freezer performance, etc. All of that affects the heat transfer coefficient and make it potentially non constant and not equal between the two waters.
> When the experimental parameters were tuned just right, the hot system’s particles almost immediately found their final configuration, cooling exponentially faster than the warm system
My daughter tried to replicate the effect in our home freezer for a science fair project. Apparently her experimental parameters were not tuned just right because she was not able to replicate the effect. The fair judges were very interested in her project, even though she wasn’t able to make it work.
When I was a kid I used to keep lizards as pets. I quickly noticed that if I filled up the water trays with cold water it would evaporate within a day from the hot heat lamps. Hot or very warm warm out of a kettle would last much much longer. I made a study as a science experiment for school and scored badly as the teacher said it was impossible.
Don't let the clickbait headline fool you. This article is a wonderful account of when the effect got its name after being noticed in a secondary school in Tanzania and all the fascinating science that's happened since then.
I am wondering why, with all of the many experiments done to investigate this, none of them were done with ice cream:
"Mpemba opted to skip waiting for his boiled-milk-and-sugar concoction to cool to room temperature like the other students had done. An hour and a half later, his mixture had frozen into ice cream, whereas those of his more patient classmates remained a thick liquid slurry..."
Boiled milk and sugar is not going to freeze in the same way as pure water, there are effects from the mixing of anything, but some of the substances in milk are oily so you might even have colloidal effects. It seems like if that's where Mpemba first saw it, you should start there to attempt to reproduce it?
Not scientific whatsoever but when I worked in my high school's canteen, we'd put hot water in the Slush Puppie machine to recover quickly during rushes and it sure seemed to work faster than cold water.
Hot water has potentially a much larger surface, as the molecules all have the brownian energy to escape the fluid bonds and cool the remainder down doing so.
So given a big enough surface (turn it into a foam) should allow it to dissipate the heat much faster, then a liquid, were the energy first has to do a slightly chaotic thermodynamic walk to the surface (away from gravity), aggregate with other local energy spikes into one spike big enough to eject a molecule from the bonds and allow energy to escape.
Sending soundwaves through the liquid that intersect with each other creating cavitys, aka foam would also help.
Hot isn't a temperature. Neither is cold. Presumably hot is some figure above freezing and below or up to boiling.
The scolding point of water is about 115.
Just encase someone wants to know, this can be measured in BTUs British thermal units. It's the measure of energy used per heating water till it's boiling point . It's also measurable if cooling the water to it's freezing point.
Fun fact is about state changing from 212 to 212 liquid to vapor is significantly more BTUs then the 1:1 (btu to one degree Fahrenheit) .
There are so many potential explanations for that phenomenon:
- hot tap water has more mineral contents than cold water which helps form the initial ice crystals
- hot water in a container creates a convection effect on the air surrounding it, thus creating an influx of cold air around the container, helping cool more quickly
- hot water in a container melts the ice/snow under it (if the experiment is conducted outside), reducing the insulation with potential colder ground
- thermal motion of hotter H2O molecules may help create accidental collisions that trigger the formation of ice crystals
There's way too much theory and not nearly enough experiment in this writeup.
I still don't know what happens if I put two glasses of water, one at room temperature and the other at 100 C, into a freezer and determine which one freezes first. If I don't see the hot water freezing first all (most?) of the time, there's nothing else to consider.
Maybe the controversy has to do with the lack of solid experimental data.
>>a lesson from the initial skepticism and dismissal that the schoolboy’s counterintuitive claim had faced:
>>“It points to the danger of an authoritarian physics.”
Key point there, at the very end, worth taking on board (which is different from listening to every willfully ignorant nutter).
Jesus Christ. This is super simple, we all learned this math in highschool: time taken to freeze is a function of the mass that needs to freeze and the temperature difference. If you cheat by allowing some of the hot mass to evaporate in order to lower the temperature more quickly you can potentially get the remaining mass to freezing temperature faster, but that isn't relevant to overall freezing time.
Have you read the article? Because the specifically mention that is been proved in other materials, then the doubt if water would behave in the same way...
My hypothesis: I'd say the hot water will experience faster convection which will continue even when it's cooled down because of inertia. That way the transport of heat to the outside is faster than with the cold water, which will experience slower convection.
that sounds reasonable, but consider this complimentary explanation: temperature itself has inertia, so the (formerly) hot water's temperature momentum is greater as it smashes against the 32F buffer zone, forming crystals instantly
not to mention static vs rolling friction (as in rolling boil)
Hot water molecules move more than cold water so the area exposed to cooling get more molecules than a cold water since their molecules are almost sleeping because it's cold.
That can't be the whole story. If the hot one freezes first, the two containers have to be at the same temperature at some point in time before freezing. And yet the one that started hot somehow continues to cool faster beyond that point.
Not really. It makes no physical sense to say the container "is at a temperature" when it's not in an equilibrium state. Neither of the two containers in the experiment will be at equilibrium until they are entirely frozen and uniformly cooled to the temperature of the freezer they're in, and there is no guarantee that they will at any point be in identical non-equilibrium states at any time.
How about this theory. the greater delta H heat flow induces structural changes in water molecule matrix that causes rearrangements that more quickly attain to the optimal ice lattice structure more quickly, and this 'enhanced crystallisation through increased delta heat' causes it to cool quicker. That this occurs in water and not other substances is related somehow to its 'paradoxical' expansion on cooling*. That the effect is flakey is related to the outsize effects of impurities and defects in the crystal lattice, and how these flaws can propagate their crystallisation inefficiencies.
Maybe a more thermodynamic way to state this is:
- temperature is an aggregate measure, so because of the nature of the ensemble, you can have some states at -22 C that still have some molecules of water (or groups of molecules of water connected by intermolecular bonds), at a higher temperature.
- when you start higher, you are going to end up with more of these higher temperature pockets in your ensemble, as your aggregate temperature drops through the degrees toward freezing
- these higher energy / "higher temperature micro regions", provide more possible microstates for the water crystals to rearrange, than if these intermolecular clumps of water were lower temp, (or were less numerous, as would be the case starting lower), and the higher temp clumps (or pockets, in the honeycomb of partially crystallized water) will probe the space of configurations of molecular arrangements more effectively (than if they were lower temperature, or less numerous, as would be the case starting lower).
- By being able to search more of the configuration space, they naturally are able to more quickly find optimal water crystal arrangements of molecules, where some H20 from the higher temp clumps, can slot into the existing lattice, or accrete onto the existing surfaces of forming ice crystals. So these ice crystals in a solution of liquid that has these higher temp pockets can be more effective at generating the movement necessary to find configurations where molecules, and clumps of molecules effectively fit together in the lattice, letting it find more effective crystal structures, leading to more efficient (and faster) crystallisation.
These other commenters are talking about the same thing:
It's cute a high school kid in Africa did an experiment that had a wrong answer and people are not sure exactly why.
Position in freezer, different sizes etc. And exploring why is worthy of science at all levels.
But this nut jobbery shows why science is broken at a structural level.
It's on them to make this replicable . Which they have not for 60 years
Yet science does not denounce the result after 60 years. Think about that.
There's little point talking about something as broken as "Science" but I suspect this is the same issue as why we have only recently understood syphons. Science doesn't have an answer for when multiple things are happening.
There are multiple genuine things screw up the result. But it's probably mainly one. But since the other things are legitimate science can't converge.
Also, if some of the water starts at 100°C, then presumably some of it will evaporate, and one should measure how much remains... The article does mention this possibility, though it doesn't evaluate or quantify it.