1. Education
You can opt-out at any time. Please refer to our privacy policy for contact information.

Discuss in my forum

Yes, hot water can freeze faster than cold water. However, it does not always happen, nor has science explained exactly why it can happen.

Although Aristotle, Bacon, and Descartes all described hot water freezing faster than cold water, the notion was mostly resisted until the 1960's when a high school student named Mpemba noticed that hot ice cream mix, when placed into the freezer, would freeze before ice cream mix that had been cooled to room temperature before being placed in the freezer. Mpemba repeated his experiment with water rather than ice cream mixture and found the same result: the hot water froze more quickly than the cooler water. When Mpemba asked his physics teacher to explain the observations, the teacher told Mpemba his data must be in error, because the phenomenon was impossible.

Mpemba asked a visiting physics professor, Dr. Osborne, the same question. This professor replied that he did not know, but he would test the experiment. Dr. Osborne had a lab tech perform Mpemba's test. The lab tech reported that he had duplicated Mpemba's result, "But we'll keep on repeating the experiment until we get the right result." (um... yeah... ) Well, the data is the data, so when the experiment was repeated, it continued to yield the same result. In 1969 Osborne and Mpemba published the results of their reseach. Now the phenomenon in which hot water may freeze faster than cold water is sometimes called the Mpemba Effect.

Why Hot Water Sometimes Freezes Faster Than Cold Water

There is no definitive explanation for why hot water may freeze faster than cold water. Different mechanisms come into play, depending on the conditions. The main factors appear to be:
  • Evaporation - More hot water will evaporate than cold water, thus reducing the amount of water remaining to be frozen. Mass measurements lead us to believe this is an important factor when chilling water in open containers, though it isn't the mechanism that explains how the Mpemba Effect occurs in closed containers.

  • Supercooling - Hot water tends to experience less of a supercooling effect than cold water. This makes it more likely to become solid when it reaches the freezing point of water.

  • Convection - Water develops convection currents as it cools. Water density usually decreases as temperature increases, so a container of cooling water typically is warmer on top than on the bottom. If we assume water loses most of its heat across its surface (which may or may not be true, depending on the conditions), then water with a hotter top would lose its heat and freeze faster than water with a cooler top.

  • Dissolved Gases - Hot water has less capacity to hold dissolved gases than cold water, which may affect its rate of freezing.

  • Effect of the Surroundings - The difference between the initial temperatures of two containers of water may have an effect on the surroundings that could influence the rate of cooling. One example would be warm water melting a pre-existing layer of frost, permitting a better cooling rate.

Test It Yourself

Now, don't take my word for this! If you are doubtful that hot water sometimes freezes more quickly than cold water, test it for yourself. Be aware the Mpemba Effect will not be seen for all experimental conditions, so consult the references in this post to see what might work best for you (or try making ice cream in your freezer, if you'll accept that as a demonstration of the effect). Let me know how it turns out for you.


June 14, 2011 at 4:29 pm
(1) DonaldW. says:

Could it be that as the steam rises it gives more cooling surface?

November 15, 2011 at 6:13 pm
(2) Joey says:

I tried the “Mpemba Effect” and i gave me the wrong result. cold water freezed faster than the hot water every time i tried it. I think the Mpemba effect works but only rarely.

November 15, 2011 at 6:14 pm
(3) Joey says:

I tried the experiment but it never worked. i don’t understand why and i’m looking forward to a reply!

March 21, 2012 at 8:04 am
(4) Roger says:

if the hot water came from the fawcet “tap” it will contain more solutes than the cold water. Test both water samples to see if this is true. If it is true the hot water will freeze at a different temperature than the cold water because anything added to water changes it’s freezing and boiling temperature. The solutes in the hot water will spread the water molecules apart and also give attachment positions for the ice growth. Very pure distilled water will freeze at a different temperature.

April 5, 2012 at 12:53 pm
(5) Hannah R. says:

I tried it in my freezer and it turns out that, for me, the hot water froze faster than the cold water. I’m doing a science fair thing for my school and this is what I wanted to do it on. Anyways, the thing is is that the hot water froze faster for me.

May 31, 2012 at 8:00 pm
(6) Brian says:

Hot water has expanded more than cold water. When water freezes it expands. Hot water is already expanded so it forms the solid form more readily than the cold water.

June 27, 2012 at 9:39 am
(7) Arek says:

maybe because the temperature difference which shock the water? between 100 degrees temp. of hot water and -10 degrees temp. of the freezer we have 110 temp. difference, but if you have 50 temp. of water and freezer is -10 degrees temp. the difference is only 50 temp. difference, so if you have 100 degrees temp. hot water and -10 degrees temp. freezer it should freeze the same as you have 10 degrees temp. of cold water and -100 temp. of freezer, which in both cases the temp. different is 110 and the water should freeze in the same period of time

June 27, 2012 at 2:01 pm
(8) c says:

All wrong. Just give up. You are all wrong.

June 27, 2012 at 7:13 pm
(9) Critical thinker says:

Dr., did you ever give this so called mystery 5 mins of thought or are you just trying to put up some physics folklore? You mean humans can create complex chemical molecules, super colliders, map geneomes, etc. but can’t expalin such a simple mystery? People, please think.

June 28, 2012 at 5:05 pm
(10) Dennisis says:

It is inconsistent because the refrigeration is inconsistent.
You must use the same water in the same refrigeration condition.
Hot water makes the refrigerator think it is getting warm and increases the output when hot water is added to it.
The opposite occurs when you put “dry ice” in your freezer. The system thinks it is too cold and shuts off……

June 29, 2012 at 4:34 am
(11) Edwin Zong MD says:

To prove my points, you can do following experiment.
There is a perfect narrow range of degree; the water (relatively high active water molecules) will form ice the fastest. If you go beyond or under that perfect range of degree, the timing for ice forming will delay.
If I am wrong, sink me to the pacific bottom.
To a certain degree, the more active water molecules (hot water) have higher probability to interact with other water molecules to form lattice structures. It means faster in term of time. However, the probability will decline when water molecules are greatly active (e.g. steam), the greatly active water molecules will counteract the chances of forming rigid structures.
The phenomenon is common in real life. If you see a group of sluggish kids, it will take a while to move them in an organized team.
On the other hand, a group of active kids can form a team faster than sluggish kids. However, if those kids are hyper active, it will take much longer time to move them in an organized team.

June 29, 2012 at 5:21 am
(12) Ashok Kumar says:

This is because hot water radiate energy more quickly to the surrounding than cold water. It lower the temprature of hot water quickly and became solid.

June 29, 2012 at 5:36 pm
(13) Jack says:

If a container of hot water is put in a freezer (after the hot water from the kitchen sink is allowed to run until it is the same temperature as the water heater setting which varies from 110 degrees – 140 degrees), it raises the temperature above the setting of the freezer inside until the compressor comes on and causes it to freeze quicker.

If a container of cold water is put in a freezer (after it is allowed to run for the same length of time as the hot water) will have a considerably lower temperature and not turn the compressor on as quickly as the hot water causing it to take longer to freeze.

July 1, 2012 at 3:06 am
(14) vijay says:

This may be because of less latent heat of evaporation, or may be due to more radiation of infrared ray from hot water due to which it cools down and freezes more quickly than cold water.

July 3, 2012 at 2:11 am
(15) fauzia says:

The hot water has water molecules far apart, when it is slightly cooled, it becomes very easy for the water molecules to form hydrogen bonds and form ice. It freezes.
whereas it is difficult for water molecules in cold water to be pushed further apart to form hydrogen bonding!!

July 4, 2012 at 3:31 am
(16) solomon owusu says:

In order to explain the Mpemba effect, it is useful to make the assumption that as temperature decreases molecules approach stability. let us also assume that there is no molecular loss to the surrounding (since loss of molecules could account for fluctuations in the results). It therefore stands to reason that higher energy molecules (in this case, hot water) turn toward stability more rapidly than lower energy molecules (in this case, cold water). With hot water, the molecules which are highly staggering possess higher energy and as a result speedily approaches stability whereas cold water with its lower energy leads hot water with reference to stability and is relatively slower in approaching stability.
This explanation holds true if experimental observations support the fact that cold water is less reactive than hot water when reactions take place under uniform conditions (reactions should take place in a closed system). Passionate scientists can prove this with empirical data

July 21, 2012 at 3:32 pm
(17) Doug Mason says:

I’ve know this since I was a young child. We has a back porch that was inclosed where our washer & dryer were located. The hot water would ALWAYS freeze first so we could only use a cold water wash. Then my dad wrapped our pipes and that took care of the problem.

July 31, 2012 at 3:30 am
(18) SteveM Arizona says:

The longer that water remains at a constant temperature, the more hydrogen bonds and water clusters are subsequently formed between the water molecules. Water at a near boiling temperature has highly excited molecules, which effectively prevents the formation of clusters and intermolecular hydrogen bonding. Thus, to create the organized lattice structure of ice, hot water has fewer clusters and hydrogen bonds to break than does cooler water.

A useful real life analogy in visualizing the organization of liquid water molecules into a solid lattice is what I call the lifeboat model. Consider hot water to be represented by active people on a cruise ship, walking all decks randomly throughout the ship. Cool water, with its water clusters and hydrogen bonding, is represented by meal time where some of the guests are seated in an organized fashion at their tables, while others are at various parts of the ship.

If a command was given to immediately fill all of the lifeboats (form ice), the model in which the dining guests are organized and contained in one location would theoretically take longer to disassemble their structure and rearrange into the boats than the model in which the guests are already moving about randomly, unencumbered by any given structure.

Thus, the reason that hot water freezes more quickly than cooler water is that hot water, with its excited molecules, has fewer hydrogen bonds and corresponding clusters than cooler, more organized water.

October 17, 2012 at 9:32 am
(19) charlie says:

Can someone just tell me if hot water freezes faster?

December 6, 2012 at 6:49 pm
(20) joshuaishere :D says:

for me cold water wins i put the same amount in? what the?
i search 20-30 sites they all said hot for sure!!
wt heck??????????????????

January 26, 2013 at 7:27 am
(21) cvon says:

In the past 5 years years the hot water pipe to my kitchen sink burst twice while the cold water pipe did not. It happened during extended periods of severe cold. Both pipes are exposed to the exterior wall of our house, a poor architectural design…

July 6, 2013 at 6:48 pm
(22) Jack says:

I’m surprised that this question still exists…
1. In older or poorly maintianed freezers frost builds on the inside, when you place a hot contiainer of water on the frost in the freezer it melts through the frost moving closer to the cooling elements and increases the surface area (sides of the container) in contact with the frost, which is denser than air. Both conditions provide an increase to the heat loss from the hot water container. (In addition, hot containers resting on the cooling coils does cause the system to turn on and increase the heat loss rate, this affect occures even when the container isn’t resting on the coils but to a lesser degree. The biggest factor in this case is the size of the freezer and the temperature range that it opperates within). The cold water doesn’t have any of these affects, to the same degree.

Hot water pipes freeze faster becauses they are used less often than cold water pipes. If you have a real problem and can’t provide them with insulation or warm air, try leaving the tap dripping.

I really can’t explain the cold room at the entrance except that perhaps they were placing the hot liquid in a metal pot and the cold in a ceramic. Metal looses heater faster.

July 18, 2013 at 3:40 pm
(23) Foot says:

For the vessel instance: The vessel with the hot water will set up convective currents within the fluid, thus bringing the warm water to the vessel walls where they can give up more heat due to the increased delta T. The mass of water will reduce in temperature more uniformly than a relatively cool mass. The cool water will act as an insulator to a degree in the cold water vessel, and the mass will cool more slowly as it approaches the freeze point. In short, the hot water activity allows it to cool faster and more homogeneously, on a fluid and molecular level.
In the case of pipes bursting, maybe not so much convective caused losses as orderly molecular structure afforded by cooling a warmer fluid.

September 27, 2013 at 9:26 am
(24) Derek says:

I tested it. Didn’t work…

Leave a Comment

Line and paragraph breaks are automatic. Some HTML allowed: <a href="" title="">, <b>, <i>, <strike>

©2014 About.com. All rights reserved.