Boil water to cool faster?

Boiling drives off the dissolved gases, then it´ll freezer faster than the raw water.

wouldn't boiling the water waste time which could have gone into cooling/freezing the water in the first place?

Rate of cooling is temperature difference proportional. Boiled hot water will have greater temperature difference so will also cool at faster rate than one near room temperature. Rest is MST and heat removal processes.

And a higher energy contend you do neither mention nor take into consideration. Even if you have a more important rate you should also evacuate a more important energy.

It should be specified more precisely: from 2 water samples having at start same temperature the one boiled and cooled to the start temperature will freeze faster than the not previously boiled.

Tell me - where's the increased temperature differential once the boiled water is at the same temperature as (or cooler than) the one that was originally near room temperature?

Fyz

I was told that if you put water at air temp. and boiling water of the same volume in to freeze at the same time. The time difference between each freezing is insignificant. The boiling water will start to freeze first. Had to due with the sustain rate the hotter water gave off its heat.

Hi Optimusprime,

Its called the Mpemba effect. Named after a Tanzanian high school student named (who else) Mpemba.

The phenomenon occurs due to a combination of factors, such as, conduction, evaporation, convection and dissolved gasses.

Take the last point for example. ordinary tap water usually contains various impurities, including gasses, which tend to lower the freezing point of the water. When you heat the water these gasses are driven out of solution thus raising the freezing point which, of course, allows it to freeze faster.

There are, I believe, other factors not mentioned above that may come into play. I don't think they are all completely understood.

Cheers,

John

Yes, boiled water freezes quicker under some conditions - i.e. low humidity, and the cooling 'source' being only marginally below actual freezing temperature. The reason is a combination of some of those stated: the removal of gases and some impurities reduces the both the freezing temperature and the energy that has to be removed once the water is near the critical freezing temperature; after a longer delay in reaching the region of freezing, the boiled water freezes considerably more rapidly.

However:
If the boiled water is cooled from below in an otherwise warm and humid environment, the slower cooling will outweigh the advantage of more rapid final freezing; or
If the cooling environment is a long way below freezing, the difference between the cooling times again becomes more important than the difference between the freezing times

Fyz

Shyam has it right ..... It's the temp differential.

Example: 2 cups of coffee, identical conditions

one you add cool cream to ... Which one will be hotter after sitting for a few minutes????

Answer .... The one with the cream...........

Are you being mischievous?

In order for the hotter drink to become the cooler one, they must at some point be the same temperature - where's the temperature difference then?

So obviously it is not the temperature differential. For the coffee, it is the "impurity" (cream) lowering the vapour pressure. The slower rate of evaporation slows the cooling. There may also be a contribution from the increased thermal capacity - if the two cups were identically filled before adding the cream.

And... if you 'float' the cream, the effect is much more marked than if you mix it in.

Fyz

See ## 5 & 6 for a truer explanantion

Hi

That is a lot of Chemistry indeed.

You are right. Milk cools faster than water or perhaps conducts heat faster than water does. When you say boiled water cools faster, one has to define the water quality and when the cooling experiment is performed. Perhaps the condition immediately after boiling may have different dissolved gases and it will take time before gases will again get in. Another fact is that some impurities will precipitate on boiling so they may affect in either way as it is their property that will make the difference. Are we talking about a particular type of impurity here?

In any way this observation is good one and nice to see all of you putting up some efforts.

Yes, diffusion through water is not very fast, so it takes a long time for gases to be reabsorbed if the surface of the water is undisturbed.

We also need to be careful with terminology here. Boiled water evaporates very slightly more rapidly than unboiled water at the same temperature - but that is not too significant for the freezing issue - the main difference occurs when the water gets very close to freezing, where energy is released as the various solutes are expelled from the unboiled water. It is worth noting that it is not only the rate of cooling that is important - given suitable solutes (e.g. calcium bicarbonate*), you could actually find a temperature where the boiled water freezes, but the unboiled water stays liquid for ever.

Fyz

*That bit truly is chemistry - to my mind, the rest is more physics, although it is taught in chemistry classes; maybe I'm prejudiced...

It is very common to put some salt to lower the ice temperature and even lower than ice cold water (about -20C) is achieved. This clearly is your point that at near 0C such water will remain water and will not form ice. Well analyzed and good point for common use.

The effect is also known as the "log mean temperature differential". That just means that the larger the temperature difference between two sources, the faster the rate of change. The large difference also increases the initial evaporation rate of the hot water, resulting in more rapid cooling than the same volume of a cooler water.

Note that I have simplified the physics. Lots.

Surely you mean you've omitted most of the relevant physics...

Tell us - where's the increased temperature differential once the boiled water is at the same temperature as (or cooler than) the one that was originally near room temperature?

Fyz

Consider too boiling water is losing substantial mass and energy as vapor. The mass escapes the water and interacts directly with the cooled atmosphere inside the freezer. Also, the now obvious point of removal of gases (though not so much impurities, actually concentrates most ionic impurities like salts).

I have the feeling either the question was not understood or the physics are not very well applied.

Let us have a look.

1st assumption. We start with "boiling water" i.e. at normal pressure with a temperature of 100°C. So that before considering the freezing we must first eliminate the thermal energy present in the water between 100°C and 0°C. After that we can consider freezing. The other sample is not boiling water with a temperature less 100°C it will reach sooner the 0°C to start freezing since a lower quantity of thermal energy has to be taken off. We can think about freezing comparison ONLY after the 2 samples reached same temperature.

2nd Assumption we start with "previously boiled water" and cooled to same room temperature as the other sample and we now consider a possibility that one sample will freeze faster -i.e. at a higher temperature than the other.

The different reasons concerning content of gas are determinant. So that all the discussion about the higher temperature difference and the higher cooling rate and so on are not at the right place. The 2 samples have almost same rate in cooling one will "freeze" sooner because its structure allows ice to form at a higher temperature.

I had always thought that as engineers we actually made "applied physics" but some times i doubt.

Not all engineers are equally able to apply all aspects of physics (nor all physicists to apply anything at all, when it comes to that). That is part of the reason that discussion on this site is helpful - we learn from each other - and partly because, by posting our incomplete understandings, we can recognise when we've missed things.

BTW, did I omit any of your points from posts ## 6 and 14?

Phallible Fyz

Excellent point Guest. You said it very well.

Lets reset the problem- Take 2 equal size containers. Put exactly the same amount of tap water into each of them (lets not nick-pick about what kind of impurities are in the water; lets just say it's average tap water from Anywhere USA, UK, etc.)

Place one of them on a stove and bring it to 100º C then immediately place both containers into the freezer.

The question is: which container will reach a solid (ice) state first?

Start your timer.......Now!

Cheers,

John

Oh BTW, the question is not only which one freezes first, but why, and by what mechanisms (to stay on topic)?

In this case the impurity likely to be precipitated as sediments and gases to escape out as you can see in bubbling boiling water initially and then normal boiling takes place. Chlorine which is also used in the water escapes on heating.

Assuming that gases will be inhibitor of heat conduction in comparison with fluid which is water here, one can have some difference in colling rate. whether ice will form at 0C or not again depends on impurity level that exist in water before and after boiling. Gases may act as molecular separator for water and inhibit crystallization process.

I think this posting looked simple, but is very interesting one. I greatly appreciate it.

Taking the mechanisms of (for example) posts ##5 and 6, it depends on the freezer. If it is a three-star freezer, the tap water will freeze first. If it is a freezer for serving traditional ice-creams, which is barely below 0-degC, the boiled water will probably freeze first.

Fyz

I totally disagree. The rate of heat transfer will be the same so that the problem at which temperature the freezing process can start. Independently of the lowest possible temperature limit either the temperature allowing the freezing is reached or not. The temperature limit will not influence the sample behaviour.

Sir Physicist please apply physics.

hi Guest,

"Independently of the lowest possible temperature limit"

What do you mean by this? What do you call the lowest possible temperature limit?

Regards-

I mean the limiys of the cooler if it is * or ***.

What counts is the moment when the sample freezes. If the cooler can go lower it does not change the sample behaviour.

Hey nick name,

Okay, lets assume that the internal temperature of the freezer is -10ºC. I place the container of room temp (21ºC) water (container A) and the just boiled (100ºC) water (container B) into the freezer.

You said in earlier post "The rate of heat transfer will be the same".

If we assume the Mpemba effect to be true on this particular occasion, i.e., container B freezing first, then container B must lose ~100º before container A loses ~21º. Hence the rate of heat transfer cannot be the same!

Am I missing something here?

-John

Freezing depends on the own temperature of water. So that in the zone where a difference can occur the rate is the same. I doubt that a "boiling water" will freeze faster than a sample of water not "boiling". We have to think on 2 aspects:

- energy, water has a content of energy which is proportional to its own temperature. The cooling is a convective/conductive process depending on the temperature difference between the 2 media. In the first phase the "boiling water" will have to get rid of the energy via convection and mass transfer since vapor will also carry some energy with. Not boiling water will not meet this phase with the same intensity. since its own energy is less important. During this phase heat transfer rate is variable since even if the cold source is big enough and has a constant temperature the water sample cools and the temperature differential decreases. It is quite complex since the lower the sample temperature the lower the vapor flow and associated heat loss. assuming that the cold source is constant as temperature and that the convection will not change cooling (water /air temperature difference) is an exponential function of time.

- freezing, it is a phase change (from liquid to solid) related as well with an energy loss. This is the latent energy which maintains water fluid. When this energy was lost freezing goes on. During freezing the temperature of water does not change much thus the rate of heat transfer is quite constant. Now I think that the different other elements in water (gas a.s.o.) have an influence on the temperature at which freezing starts since the structure of water is modified by their presence. Their presence or not presence lead to the different freezing temperatures. A good example is the fact that NaCl (salt) maintains water fluid at temperatures below the freezing point of clean water.

I doubt that we have to compare "boiling water" with "not boiling water" but "boiled water" with "not boiled water". This was mentioned previously.

I'll try not to take offence, and interpret your "please apply physics" to mean "please explain the physics you are using".

First, I take the temperature of a freezer for serving traditional ice creams to be about -5-degrees-C. This is much warmer than used for most for modern recipes (although most of these are described as traditional).
Now I will assume a cooling process where the rate of heat flow is simply proportional to temperature differential. This might be relatively slower for the boiled water than it should be, but allows a semianalytic solution. We can then state the rate of extraction of the heat as = K/mass*deltaT (Joules/gramme/degreeC).

We have an additional time for the boiled water to reach the temperature of the tap water (100-degrees to say 18 degrees). Both take the same time to go from 18 to 0.

These times are respectively:
100->18: 4.2*ln((10+5)/(18+5))/K = 6.38/K and
18->0: 4.2*ln((18+5)/5)/K = 6.41/K

The preboiled water starts to freeze at 0-degrees C, I am working on the principle that the tap water supercools to -0.5-deg-C where expels the adsorbed gases and the Calcium-bicarbonate - but I am neglecting for now any additional heat that needs to be extracted to desorb the gas or Calcium Bicarbonate; the times are:
0deg freezing-> 334/5/K = 67/k
Cooling from 0 to -0.5 -> 4.2*ln(5/(5-0.5)/K = 0.44/K
-0.5deg freezing -> 334/(5-0.5)/K = 74.44

That gives total times for the freezing of 79.79/K (preboiled) and 81.29/K (tapwater). So tapwater takes longer than preboiled water in a -5-degree cooler

Using the same principles and a 3* freezer with a temperature of -18 degC, the times for freezing are 26.51/K (preboiled), and 22.17/K (tapwater) - so for this case, the tapwater freezes first.

I know the values for my assumptions may not be quite right - but they do illustrate how a small differential between the freezing points and the freezer temperature give earlier freezing for the preboiled water, whereas a larger temperature differential gives the tapwater freezing first. Incidentally, I intended to type 5*, which makes the differences more marked. However, as the numbers work for 3* as accidentally typed, that is what the calculations used.

Questions?

Fyz

The difference is in the model used not in the physics.

If the model goes from a small sample in a freezer already at its limit temperature your results are correct.

If the model considers that the volume where the sample was placed and the sample start at same temperature and cool together then the other model is correct.

Physics are allways the same but the way to look at a problem can be different.

The difference should not be in the models, but the situation being analysed.

At least I have convinced you (I think) that boiling water will sometimes freeze more rapidly than unboiled water - which you doubted originally. If the samples start at the same temperature, of course the boiled water will always freeze first. But the question says you boil it to get it to freeze faster - so in reality I should have started the model at the point when you turn the kettle on - in which case I would have needed to take more complex effects into account (or assumed a larger degree of supercooling).

Fyz

Due to your last comment I had a more detailed look at the "problem". I recognize that first I did not made any thing else than a qualitative analysis.

I made some assumptions and defined some parameters as follows:

- for one of the possible cases I considered the same as you, i.e., "the sample is small and the freezer powerful enough to assume its internal temperature as constant".

- The samples are characterized by:

- their volume "V" [m^3]

- heat capacity cp[J/kg K]

- Density ρ [kg/m^3]

- Heat transfer from sample is characterized by:

- transfer area A[m^2]

- heat convective coefficient α[W/m^2K]

- Cooling is in 2 steps:

- fluid cooling from initial temperature θ1to critical θ cr [°C]

- transfer of latent heat at constant fluid temperature θ cr

- freezer temperature is either θfr1= -5°C or θfr2= -25C° only for the example, other values can be considered.

- Both samples have same parameters

For the 1^st phase the time to reach the critical temperature is:

T1= (V*cp*ρ)/(A*α)*ln((θ1-θfr)/(θcr-θfr))

The different values are in end table.

If it is accepted that α is constant then the time to take off the latent heat is:

T2= (V*ρ*Q1)/( A*α)/( θcr-θfr) where Q1= latent heat [J /kg].

The total time will be:

T=T1+T2= (V*cp*ρ)/(A*α)*( ln((θ1-θfr)/(θcr-θfr))+Q1/(cp* ρ)/(θcr-θfr)).

Following table gives the results for different combinations of temperatures.

Combination

θ 1 [°C]

Θcr[°C]

Θfr [°C]

Ln[(θ1-θfr)/ θcr-θfr)]

1/ (θcr-θfr)

1

100

0

-5

3.0445

0.20

2

18

0

-5

1.5261

0.20

3

18

-1

-5

1.7492

0.25

4

100

0

-25

1.6094

0.04

5

18

0

-25

0.5423

0.04

6

18

-1

-25

0.583

0.0417

I do not see any inversion, but if I am wrong I would appreciate a comment.

I have not yet analysed what happens if the sample is "big" and the freezer has a limited power to expel heat.

I appreciate such contradictory discussions since as the old greeks said truth comes out from contradictions and discussions.

I found this extraordinarily hard to follow - possibly because you used formatting that got lost.

But, as far as I could judge you have used the same freezing temperature and the same latent heat for both samples. Given that we are starting with tap water, these should both be changed by the action of boiling. That is the entire cause for the change of order. If you take the simplified values that I used (0.5-degree change in freezing temperature), you will get similar results to mine. If you also allow the heat requirements to change as a result of boiling the water, that will alter the effects again. [Note that even rain water will see some change, due to the expulsion of dissolved gases]

To repeat remarks that have been made by others elsewhere in this thread:

There are two reasons that the boiled water can freeze first: one is that any calcium bicarbonate that is dissolved in the water is decomposed by boiling, and the decomposition products are virtually insoluble. Similar to common salt, the presence of the solid solutes lowers the boiling temperature; so its removal will result in the boiled water having a higher freezing point than the tap water.

The other relates to the expulsion of dissolved gases: both boiling and freezing remove the adsorbed gases - but this takes additional energy. For tap water, this energy has to be conducted away during the freezing process. Once the water has been boiled, the freezing process does not have to remove this additional heat. This can actually result in a secondary effect - for fast freezing you get a degree of supercooling.

If you ignore these effects, your model will inevitably show that whatever starts cooler will freeze first. I only took one effect into account - the reduction in freezing temperature - mainly because I did not have any convincing data for either of the effects, and I haven't the leisure at the moment to perform the requisite experiments.

There could be one other difference between our arrangements - I think you have modelled convection, whereas I have modelled the case where the vessel containing the water sits on the floor of the freezer, and conduction is dominant. That will result in changes to the detail - but it will not change the order of freezing for the -5-degree case.

Regards

Fyz

Freezing temperatures were different: 0° for boiled or boiling water and -1 ° for the other one.

The content of gas and carbonates is so small and the latent heat too important to be affected in a remarkable way. as far as I understood in your computations you also had only one value for the latent heat.

Either convection or conduction there is a similar "heat barrier" limiting the heat transfer. In the fluid heat transmission to the wall is only by convection.

I am sorry for the format I wrote the text under "Word" and copied it not knowing that it will be so messed!

By the way I never wrote that boiled water will freeze lower I totally agree that the impurities lower the freezing temperature by their presence and modification of the molecular strcture of water.

This* is what I was referring to - it appeared to me that you had recanted, but I must have misunderstood.

* "I doubt that a "boiling water" will freeze faster than a sample of water not "boiling"."

Is there any chance that you could set your calculations out so that a mere mortal could interpret them? BTW, you need to check what they look like using the "Preview Comment" to have any confidence in the result.

N.B. that at the moment I don't even know whether you believe the boiled water will freeze first in both cases, or that the tap-water will always freeze first.

Regards

Fyz

Thanks for the indication.

I shall try to put it in an more comprehensible form. For me "boiling water" is water at ≈100°C.

There are 2 kinds of "not boiling water". The 1st is coolled pre-boiled water and not at all boiled (tap)water.

What I wanted to say is that if one cools a sample of water starting from 100°C and an other from let say 18°C the 1st one will take a longer time to reach the freezing temperature which ever it is. The difference is generated by the time to take off the heat content between 100 and 18 °C.

The term "faster" is in general related to time so this is the way I used it.

If one would compare a "pre-boiled" and a "not boiled" sample, starting from same temperature, the second will freeze at a lower temperature and take a longer time to reach the critical temperature since this will be lower due to the impurities.

Where I was not precise enough was the estimation of the influence of cold source on the transfer rate. The reason is that I did only look at the sample alone and did not consider the environment and the heat transfer.

I hope I made it clear now but if any other question please feel free to put it.

not same guest