Sciencemadness Discussion Board

Why Hot Water Freezes Faster Than Cold—Physicists Solve the Mpemba Effect

Finnnicus - 31-10-2013 at 03:41

I found this article on reddit.
https://www.medium.com/editors-picks/d8a2f611e853

Hopefully this isn't old news. Anyway, just thought it would be interesting!

Source for the article:
http://www.arxiv.org/abs/1310.6514

maxpayne - 31-10-2013 at 04:28

Is this effect "active" only when freezing? Any research done with graphs to describe the speed of the effect? Is there an equilibrium points (two starting different temperature points that change to one temperature point of water while cooling at the same speed)?

How does salts dissolved in water affect this effect? Is there any compound that do change the effect? etc, etc...

Trillions of serious questions must first be answered before any conclusion, math and bonding want help here or convince anyone. Experiments first!

Edit: I wish Viktor Schauberger who is known as water wizard is alive at this time. He would certainly give us some good directions.

[Edited on 31-10-2013 by maxpayne]

Varmint - 31-10-2013 at 06:24

We did this experiment in HS chemistry class.

At the time the mantra was (and still is) that the rate of change is greater due to radiation from the wamer water, but we were puzzled (justifiably) by the apparent slighly quicker transition to freezing.

Then the teacher encouraged us to weigh the products.

In each group (maybe 10 groups of 3 students?) we found the ice that started from the hot water was significantly less mass than that which started from cold. I'm sorry but I don't recall the actual values or percentages right now, but the word significant should suffice.

We were expecting heat loss of course, and were satisfied that the radiation and convection justified the rate of change arguement, we just failed to account for the significant change in mass being responsible for the "unusual" result.

I call BS on these guys, they failed to mention the key parameter, mass.

DAS

watson.fawkes - 31-10-2013 at 07:21

It is widely considered proper to have read a paper in question before criticizing it. It's not like it's hard to do so; PDF copy of the paper is on the arxiv.org link in the original post.
Quote: Originally posted by maxpayne  
Any research done with graphs to describe the speed of the effect?
See page 2 and the references in the end note from which the graphs are drawn.
Quote: Originally posted by Varmint  
I call BS on these guys, they failed to mention the key parameter, mass.
Evaporation as a possible explanation is on page 3.

Come on, people.

It's not like this paper is the last word on the subject. It's incremental progress in the field. They present a theoretical model that provide an account of a proposed process that matches known observations in a way that previous models did not. There will have to be further research to test this model against other conditions.

woelen - 31-10-2013 at 07:34

Perform a simple experiment. Using a volumetric flask, measure 100 ml of water and transfer this to a 125 ml glass bottle. Cap the bottle. Repeat this with another bottle of the same type, using the same source of water. Now you have two bottles, filled with water (and quite some air above it, this is important, because othterwise the bottles may crack in the experiment). Both bottles are filled with water of the same source and same temperature. So, you can state that within accuracy of the volumetric flask they contain the same amount of water.

Now put one bottle in a fridge with a temperature of 5 C or so. Put the other in a warm place (e.g. on top of a hot radiator at 70 C or so or in an oven set at 70 C). Next, put both bottles in a freezer with a window in its door, so that you can watch the bottles. Which one freezes first? Maybe an automatic setup, using a camera, which takes a picture with timestamps every 15 seconds or so, can be deployed.

Repeat the experiment, using the same bottles, but now reverse which bottle is made hot and which bottle is made cold. In this way the influence of the different bottles is eliminated as well.

Using capped bottles assures that no water is lost through evaporation. I personally expect that the bottle with cold water freezes first.

[Edited on 31-10-13 by woelen]

bfesser - 31-10-2013 at 07:53

Volumetric flasks are calibrated to contain, not to deliver. Adhesion of droplets of water to the surface may be significant, particularly if the flask isn't properly cleaned.

watson.fawkes - 31-10-2013 at 08:05

Quote: Originally posted by woelen  
Next, put both bottles in a freezer with a window in its door, so that you can watch the bottles. Which one freezes first? Maybe an automatic setup, using a camera, which takes a picture with timestamps every 15 seconds or so, can be deployed.
The experimental data used an electronic temperature probe to measure the temperature more-or-less continuously. This data is important, because it shows that (under certain circumstances I don't know how to characterize) the time constants for the cooling process differ; that's what the paper in the OP purports to explain.

But really, there's no need to reinvent the wheel. I haven't read all the literature, but I have to guess that there's been more than zero work on seeing a reproducible effect.

Varmint - 31-10-2013 at 09:10

Watson:

Yes, I notice the single word dismissive use of evaporation, without any research showing their loss in mass due to evap.

In other words, they dismissed the key element. This is how "global warming scientists" are created.

DAS

Pulverulescent - 31-10-2013 at 11:13

Quote:
This is how "global warming scientists" are created.

Geeez! No fucking shit, Sherlock?

Hexavalent - 31-10-2013 at 12:41

Steady on, Pulverulescent.
As a modification to Woelen's experiment, because it may be difficult to determine exactly when the water samples have frozen entirely, one could substitute the bottles for plastic graduated cylinders, calculate the expected volume change of the water between the two states, and monitor the time taken for the water samples to reach the calculated solid volume when sat in a freezer, for example. I used this technique for a school project a while ago, and it worked very well, within the accuracy of the graduated cylinder.

deltaH - 31-10-2013 at 13:30

Perhaps the Mpemba effect (if true) can be explained by simple crystallisation kinetics theory. For example, crystal seeding has a higher activation energy than crystal growth, so by freezing warmer water, you create more ice crystal seeds than the case for cold water.

Or put backwards, cold water struggles to form ice crystal seeds, some eventually do and so it freezes over from there, but seed formation may be rate limiting on an overall ice formation rate.

Still not convinced? Okay ask yourself, this effect is due to either kinetic effect or a thermodynamic reason, the latter is clearly not possible, so that leaves only a kinetic effect.

The logical choice is therefore to turn to crystal growth kinetic theory and that gives a far more simple and plausible answer than this paper, simply by applying Occam's razor on the two explanations.

The interesting thing about this, if true, is that it would mean that ice crystal seeds can form at significantly higher temperatures than ice itself.

But clearly millions of microscopic ice crystals don't go floating around in hot water or they would have been seen. However, the seed for ice in this sense does not need to be a solid ice phase per say.

Water is special because of its strong hydrogen bonding and this dominates much of it's chemistry (including crystallisation physical chemistry too).

So perhaps these ice seeds are not micro solid particles, but simply regions where higher order of hydrogen bonding exists than would otherwise occur in equilibrated fully random hydrogen bonded water.

These could be the basis of ice crystal seeds that are still 'fluidic', but of lower entropy than the bulk.

[Edited on 31-10-2013 by deltaH]

Morgan - 31-10-2013 at 14:23

What if you had two beakers each with the same amount of water and heat energy, one with all the water stirred and the other with the equivalent amount of heat but that it was set up to contain a hot layer of water on the surface? That is, very carefully add hot water so as not to mix the two temperatures in the one beaker as much as possible. I noticed the other day an engine I ran partially submerged in a small barrel heated the water enough to burn my hand yet just below this top layer the water was quite cool.
Would a higher heat gradient transfer/lose heat relatively faster, slower, or the same as the fully mixed beaker of heated water? Would they have the same heat loss curve? If you wanted, a thin glass plate could be placed on the surface to inhibit evaporation.
This experiment might also introduce another factor, the surface area at the top region of the vessel, say if a Erlenmeyer flask were used.


[Edited on 31-10-2013 by Morgan]

watson.fawkes - 1-11-2013 at 10:33

Quote: Originally posted by Varmint  
Yes, I notice the single word dismissive use of evaporation, without any research showing their loss in mass due to evap.
That's because it's in one of the references at the end of the paper.

maxpayne - 2-11-2013 at 15:12

Since this topic is about water, please allow me to point into some directions that maybe can help someone to resolve the question of the supposed Mpemba effect (I'm somewhat sure it is true).

First let me start with before mentioned Viktor Schauberger and his book The Water Wizard. This book is really great, it gives you strange but incredible informations you want find elsewhere. Highly recommended! Here is some text from the book:

"Another of its life-giving properties is its high specific heat - lowest at +37.5°C (+99.5°F). The term "specific heat" refers to the capacity and rapidity of a body to absorb or release heat. With a relatively small input of heat fluids with a high specific heat warm up less rapidly than those with a lower specific heat. How strange then, and how remarkable, that the lowest specific heat of this "inorganic" substance - water - lies but 0.5°C (0.9°F) above the normal +37°C (+98.6°F) blood temperature of the most highly evolved of Nature's creatures - human beings. This property of water to resist rapid thermal change enables us, with blood composed of 80% water,to survive under large variations of temperature. Pure accident so we are told, or is it by clever, symbiotic design?!"

"The No. 1 enemy of water is excess heat or over-exposure to the Sun's rays. It is a well-known fact that oxygen is present in all processes of organic growth and decay. Whether its energies are harnessed for either one or the other is to a very great extent, if not wholly, dependent on the temperature of the water as itself or in the form of blood or sap. As long as the water-temperature is below +9°C (+48.2°F), its oxygen content remains passive.Under such conditions the oxygen assists in the building up of beneficial,high-grade micro-organisms and other organic life. However, if the water temperature rises above this level, then the oxygen becomes increasingly active and aggressive. This aggressiveness increases as the temperature rises, promoting the propagation of pathogenic bacteria, which, when drunk with the water, infest the organism of the drinker."

Also you may want to check:
http://www.youtube.com/watch?v=ItsSYwTSnTY (Viktor Schauberger - Comprehend and Copy Nature)
There are other documentaries on YouTube about water memory, etc.. You'll find it if you get interested.

Someone might think that I'm promoting pseudoscience, but I merely want to give insight about water. It is not that this is an attack on current scientific dogmas or whatsoever, all I want is to share this for those interested.

deltaH - 2-11-2013 at 23:56

Hiya maxpayne, the first paragraph, Schauberger states a fact about the minimum heat capacity of water (I've fact checked it, see this table), so this is not pseudoscience. Going on to speculate that we operate at this temperature because of this is in my mind scientific romanticization (poetic speculation), but not per say impossible or flawed by known fact, so I wouldn't call it pseudoscience.

However, the second paragraph might well be pseudoscience in my opinion. Consider this plot of the solubility of oxygen in water as a function of temperature (also from the www.engineeringtoolbox.com):



Oddly enough, I've also used this plot on another thread here about "super oxygenated" water type 'snake oils' (miracle products).

As you can see from the plot, nothing special happens below 9C. In fact, more oxygen dissolves in water the colder you go.

Perhaps the author is suggesting that at 9C, a clathrate-like solvation cage forms around the oxygen molecule in solution that isn't stable at higher temperatures, but surely such a peculiar property deserves referencing!

Is something referenced further on in the text? I would enjoy reading that if so!

[Edited on 3-11-2013 by deltaH]

vulture - 3-11-2013 at 02:28

Quote:

How strange then, and how remarkable, that the lowest specific heat of this "inorganic" substance - water - lies but 0.5°C (0.9°F) above the normal +37°C (+98.6°F) blood temperature of the most highly evolved of Nature's creatures - human beings. This property of water to resist rapid thermal change enables us, with blood composed of 80% water,to survive under large variations of temperature. Pure accident so we are told, or is it by clever, symbiotic design?!"


Ugh. You want us to read a book by someone who is incapable of standard scientific reasoning? It's the other way round, animals evolved to take advantage of this sweetspot. Furthermore, there are countless examples of animals with body temperatures which highly differ from 37°C.

turd - 3-11-2013 at 15:32

Quote: Originally posted by vulture  
Ugh. You want us to read a book by someone who is incapable of standard scientific reasoning?

https://translate.google.com works quite OK for the german wikipedia page on Viktor Schauberger (https://de.wikipedia.org/wiki/Viktor_Schauberger ): He was a perpetual motion crackpot. Apparently he had a catastrophic meeting with Hitler: Schauberger believed to have convinced Hitler. Hitler thought Schauberger was a fraud. :)

Morgan - 3-11-2013 at 16:34

I guess some spin-offs worked. "According to medical opinion, the whirled water can relieve tensions in the neck and shoulder areas ..."
Advance to 31 minutes and ~16 seconds.
Comprehend and Copy Nature
http://www.youtube.com/watch?v=ItsSYwTSnTY#t=31m16s

Well, even Lord Kelvin was terribly wrong about flight. But from what it looks like there's a lot of "sifting" needed with Viktor Schauberger, as much as one wants to like those who try to copy nature.
http://scienceworld.wolfram.com/biography/Kelvin.html

[Edited on 4-11-2013 by Morgan]

franklyn - 3-11-2013 at 21:02

The vapor heat transfer out of the liquid is faster if it is already rapidly evaporating as when hot. The result is that you are left with less ice but quicker than if you start with cold water. Doing this in closed vessels does not prevent heat transfer out of the system. There is likely also some stratification occuring so that freezing begins from the denser cooler bottom up. Related effects are superheating where liquid water has a temperature above it's boiling point and supercooling where it remains liquid even as it's temperature is below the freezing point.

.

.

AJKOER - 18-11-2013 at 19:56

I have another source confirming Franklyn statement "The vapor heat transfer out of the liquid is faster if it is already rapidly evaporating as when hot". The reference relates to the science of freezing relating to lakes. To quote from an educational site (link: http://abacus.bates.edu/~raustin/FluxFlow/Lab2.pdf ):

"Processes that remove heat from the lake include net thermal (infrared) radiation away from the lake, flow of water out of the lake, and conduction of heat to the overlying air (if the air is colder than the ice surface). "

so conduction and thermal radiation in a kinetically more active hot water solution favors faster cooling. I will also add an evaporation effect if the vessel is not sealed as again, I would expect more water loss from a hot solution.

Another factor, to a lesser extent, that may be occurring is the change in gas content (N2, O2, CO2,..) that occurs in cold water upon heating (see graph of dissolved oxygen versus temperature at http://www.google.com/imgres?imgurl=http://www.cotf.edu/ete/... ). This could be significant as it changes the mass of the water. To quote from the same source:

"A note of caution: Temperature does not measure the amount of heat present in the water. Heat is measured in units of energy, either calories or Joules. The heat energy stored in a volume of water is found by taking the product of the water body's MASS, TEMPERATURE, and SPECIFIC HEAT."

So, for one, simply reducing the mass of cold water by evaporation or removing dissolved gases upon heating, will lower the amount of heat energy stored in a given volume of water, thereby accelerating the freezing process.


[Edited on 19-11-2013 by AJKOER]