# Hot/warm water can freeze faster that cold water **URL:** \ **Category:** Cecil's Columns/Staff Reports\ **Created:** [May 19, 2006, 3:37pm UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413 "2006-05-19T15:37:11Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![Al\_Al](https://avatars.discourse-cdn.com/v4/letter/a/8baadc/32.png) [@Al\_Al](https://boards.straightdope.com/u/Al_Al)\ **Post date:** [May 19, 2006, 3:37pm UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/1 "2006-05-19T15:37:11Z")
Warm/hot water can freeze faster than cold water only in older refigerators, before auto defrosters, where warm water would melt the accumulated frost, which acts like an insulater, and hence get closer to the coils removing heat faster. ---
**Author:** ![C\_K\_Dexter\_Haven](https://avatars.discourse-cdn.com/v4/letter/c/b2d939/32.png) [@C\_K\_Dexter\_Haven](https://boards.straightdope.com/u/C_K_Dexter_Haven)\ **Post date:** [May 20, 2006, 11:35am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/2 "2006-05-20T11:35:04Z")
Welcome to the Straight Dope Message Boards, **Al&Al** , we’re glad to have you with us. When you start a thread, it’s helpful to other readers to provide a link to the Column or Staff Report that you’re commenting on: helps keep everyone on the same page, and avoids searching time. In this case, I presume it’s [Which freezes faster, hot water or cold water?](http://www.straightdope.com/classics/a2_098b.html) Since that’s actually a column by Cecil rather than a staff report, I’m moving your thread to the appropriate forum. No biggie, you’ll know for next time, and, as I say, welcome! ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [May 21, 2006, 6:57am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/3 "2006-05-21T06:57:09Z")
I like that explanation. It’s some good creative thinking. Of course it rellies on several factors, primarily that the water you’re working with is a lot more than just an ice cube tray. We can’t say for sure how the popular lore got started, but I doubt it was with putting in bowls. However, I had another idea, but along similar lines (which, I repeat, I find quite creative). Perhaps it’s that the hot water melts for itself a nice indentation in the ice, greatly increasing the contact area. This would work for all water amounts but would still require a frosted-up fridge. ---
**Author:** ![Majikthise](https://avatars.discourse-cdn.com/v4/letter/m/bb73d2/32.png) [@Majikthise](https://boards.straightdope.com/u/Majikthise)\ **Post date:** [May 28, 2006, 5:41am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/4 "2006-05-28T05:41:59Z")
A thorough article on this phenomenon, complete with references, can be found at [this site](http://math.ucr.edu/home/baez/physics/General/hot_water.html); it makes up a part of the Usenet Physics FAQ. > [@](#): > > The phenomenon that hot water may freeze faster than cold is often called the Mpemba effect. Because, no doubt, most readers are extremely skeptical at this point, we should begin by stating precisely what we mean by the Mpemba effect. We start with two containers of water, which are identical in shape, and which hold identical amounts of water. The only difference between the two is that the water in one is at a higher (uniform) temperature than the water in the other. Now we cool both containers, using the exact same cooling process for each container. Under some conditions the initially warmer water will freeze first. If this occurs, we have seen the Mpemba effect. Of course, the initially warmer water will not freeze before the initially cooler water for all initial conditions. If the hot water starts at 99.9° C, and the cold water at 0.01° C, then clearly under those circumstances, the initially cooler water will freeze first. However, under some conditions the initially warmer water will freeze first – if that happens, you have seen the Mpemba effect. But you will not see the Mpemba effect for just any initial temperatures, container shapes, or cooling conditions. There are several factors which may help to explain how the effect works, when it does happen. One possible factor is evaporation: the hotter water might freeze sooner, but you might end up with a smaller amount of ice. Another possibility is convection currents: with water that starts at a higher temperature, convection currents can help to maintain a temperature gradient with warmer water at the surface. This “hot-top” condition means that it loses heat faster than if the heat were distributed uniformly throughout. Basically all the explanations have to do with the fact that a single number (the average temperature) is not sufficient to describe the state of a container of water. ---
**Author:** ![gonzomax](https://avatars.discourse-cdn.com/v4/letter/g/e8c25b/32.png) [@gonzomax](https://boards.straightdope.com/u/gonzomax)\ **Post date:** [May 29, 2006, 3:19am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/5 "2006-05-29T03:19:14Z")
Ive heard this for years and in a way it defies logic. If one tub is 120 and the other 60 ,the one at 120 has to get to 60 and that takes time. A kind of Zenos paradox thing. I guess it is true I been told so for years,but it has to be a temperature range to work. If ones at 62 and the othrer at 60 is it good.?If ones at 200 and the other at 40 . The story has to be incomplete. ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [May 29, 2006, 1:50pm UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/6 "2006-05-29T13:50:23Z")
The Mpemba effect is slight, and i think it only works at temperatures near boiling. Anyway, it certainly cannot be the explanation for the popular lore. I think a lot of physics guys repeat it as if it were, but that simply cannot be the case. I think my theory that warm ice cube trays sink into the ice a tad and vastly improve their contact area is the most relevant one. Note that they don’t have to sink much. However, if you use cold water then the contact between the tray and the very uneven frost is almost negligible. Boosting that should have a very significant effect. One worthy of lore. Also, the temperatures here are not almost boiling vs nearly boiling, but rather cold versus warm. ---
**Author:** ![Majikthise](https://avatars.discourse-cdn.com/v4/letter/m/bb73d2/32.png) [@Majikthise](https://boards.straightdope.com/u/Majikthise)\ **Post date:** [May 29, 2006, 4:22pm UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/7 "2006-05-29T16:22:35Z")
> [@Alex\_Dubinsky](#): > > The Mpemba effect is slight, and i think it only works at temperatures near boiling. A recent (December 2005) article on the Mpemba effect can be found at [[physics/0512262] Hot water can freeze faster than cold?!?](http://arxiv.org/abs/physics/0512262). Page 8 of the pdf shows a graph of freezing time versus initial temperature for several experimental setups in a freezer. For most of them, the maximum freezing time peaks at an initial temperature of somewhere around 60 to 80 degrees C. These results agree with what you’ve said: The effect is slight, and does not seem to manifest itself at tap-water temperature ranges. However, a classic version of the experiment involves placing buckets of water outside in the winter, where some people have found that water at 40 C freezes fastest. (I found some discussion of this from New Scientist magazine, which is referenced from the Usenet FAQ.) The difference in the results suggest that the Mpemba effect is probably caused by more than one mechanism. > [@Alex\_Dubinsky](#): > > Anyway, it certainly cannot be the explanation for the popular lore. I think a lot of physics guys repeat it as if it were, but that simply cannot be the case. > > I think my theory that warm ice cube trays sink into the ice a tad and vastly improve their contact area is the most relevant one. Note that they don’t have to sink much. However, if you use cold water then the contact between the tray and the very uneven frost is almost negligible. Boosting that should have a very significant effect. One worthy of lore. Also, the temperatures here are not almost boiling vs nearly boiling, but rather cold versus warm. I think a major reason for why it is part of popular lore is simply that it has been around for so long. Descartes wrote about the effect in the 1600’s, and it was also mentioned by Roger Bacon and even Aristotle before that. But I agree that it probably wouldn’t be so wide-spread in the lore if there weren’t some way for ordinary people to get results from hot tap water. In addition to your melting-frost theory, I’ve also seen an explanation involving freezer thermostats: If you put a large quantity of warm water into a small freezer, it might trigger the thermostat to activate the cooling coils, which stay on until the freezer reaches a lower temperature. The lower temperature in the freezer would then freeze the water faster. Of course, this effect wouldn’t show up in a good scientific experiment, which would control for variables like freezer temperature. ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [May 30, 2006, 3:12am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/8 "2006-05-30T03:12:54Z")
Aristotle and Descart, you say? Wow. Hmm, and the 40 degree buckets too. Alright, melting front and thermostats might do the trick in modern times but the mbepa effect does seem a valid origin of the lore (although i don’t think we should call something that has many different causes and forms to be one effect). Actually, i don’t buy the thermostat thing. I can see how turning it on might mean that it won’t stop until the air becomes colder than when it was first activated, but I don’t think it would become that much colder, and the time during which it was actually warmer inside the freezer would be longer than the time it was cold. Basically, if you’re putting in so much warm water that the air temperature rises significantly (ie the air is not cooled fast enough by the ice), then i don’t think it will ever freeze faster. Only if you put your warm water right under the thermostat to in effect trick it, will such a scheme work. Anyway, I like your point about how long and prominently this effect’s been observed. Maybe freezers actually have nothing to do with it, and people spread it just via urban legend. With 40C buckets outside in the winter, I think the only good explanation could be that convection prevents the top from freezing over. This convection would have to be very slow, so that a single molecule should turn around just once. At the top it cools to near 0 and sinks to the bottom to stay there. Of course, it’s pretty frustrating to argue such things without watching the process yourself (and observation is likely to yield readily an answer in this case). > [@](#): > > Of course, this effect wouldn’t show up in a good scientific experiment, which would control for variables like freezer temperature. That’s why the scientific method is actually inappropriate for most things that are not physics or chemistry. Holding all variables constant only works when the functions you’re working with are very simple and regular so that you can expect to extrapolate your limited results to more general principles. Seeking very good precision is only relevant when you already have a clear idea of what types of mathematical funcitons will exactly describe your phenomenon, and you need that precision to select just the right one. Such things do not apply to more amorphous things like sociology, biology, psychology, etc. That’s why a single experiment used to be able to give proof to physicists while a study now yields hardly more than more uncertainty. It is why you don’t just make sure all the people you use for a study are the same, but try to get as many different ones as possible. Obviously we’ve abandoned the concept of the control in that respect. Now we need to realize that we shouldn’t stop there. Of course not holding variables constant makes obvious the increased demands of being able to process all that data. But using the “scientific method” actually makes this step more difficult than less. It’s much harder to parse a dozen studies each saying a different thing than a dozen studies which say nothing on their own but yield a mass of data that may be combined for analysis. Also, we need to somehow move past the principle of having a small team do a study and then give us conclusions. There are no conclusions they alone can give us. And worst part of all, they only publish their dumb conclusiong and not the actual valuable data (especially data uncolored by their misconceptions). ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [May 30, 2006, 3:16am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/9 "2006-05-30T03:16:32Z")
Oh, btw, water near 0C I think is denser than water at 10C (at least it’s denser than 4C). So that’s why it wouldn’t sink and would just freeze over, with ice being a good insulator. ---
**Author:** ![Alex\_Dubinsky](https://avatars.discourse-cdn.com/v4/letter/a/dbc845/32.png) [@Alex\_Dubinsky](https://boards.straightdope.com/u/Alex_Dubinsky)\ **Post date:** [May 30, 2006, 4:29am UTC](https://boards.straightdope.com/t/hot-warm-water-can-freeze-faster-that-cold-water/357413/10 "2006-05-30T04:29:56Z")
Er, I meant less dense. WHY CAN"T WE HAVE EDIT BUTTONS, FOR GOD"S SAKE??? I hate how I have to weigh bogging down the thread with making sure i don’t sound like an idiot.