Mpemba effect or not, I stand by my statement in post #16 (slightly modified here):
The Mpemba doc throws all kinds of additional factors into the equation like air bubbles, dissolved solids, added impurities, air flow over the surface etc., any of which could cause cooling to be slower or faster, but this is changing the premise, which is concerned only with unadulterated water, not water with something.
Out of interest: at what volume/shape/temperature should I be expecting a freezing time where the warm water comes anywhere near close?
Or: describe an experiment to falsify my assertion.
Experiments I’ve done so far:
Two shot glasses 7 degree C./45 degree C. Freezer @ -20 C. Result: cold water is frozen solid before ice starts to form in the warm water.
Two plastic cups: 7 C /95C, freezer @ -20 C. Result: see above.
By the way, does anybody have any details about how Archimedes made ice cubes while performing this experiment? Was he using a frost-free refrigerator, or one of the older models that did not regulate humidity in the freezer?
I think you mean Aristotle, not Archimedes. Here’s what he says:
Were you not aware that water can be frozen other than by refrigerators, such as by cold weather?:dubious:
The article on the Mpemba effect linked above specifically addresses this. Even with unadulterated water, as the linked article points out, your reasoning rests on the assumption that “average temperature of the water” is the only important factor here. They might not become identical in every way.
ETA: We’ve had Straight Dope columns using computational fluid dynamics software to solve the shower curtain mystery and the Guinness Stout bubbles question. Can somebody apply computational fluid software to this problem and see if any of the “convection currents” effects show up?
What I’m missing here are concrete directions on how to duplicate the elusive RESULTS.
‘The world is more complex than you think’ is a very deep and almost universal truth but, as we say here; Cite?
Don’t come back with links to wordy sites with 0 experimental data.
I want to see replicatable results. I want to know how to get warmer water to freeze sooner than colder water. Because in my fridge it most definitely doesn’t.
Well, if I were you, I would try one container only a few degrees hotter than the other, not 38 to 88 degrees hotter.
Yeah, I’m trying to find the details of Jearl Walker’s “Amateur Scientist” column in September 1977 (the one mentioned in the Straight Dope column), because the article from the physics FAQ mentions that he gives good starting points. Unfortunately, none of my academic affiliations have access to Scientific American back beyond 1984, so I am still searching. Also for the other cited papers from the FAQ.
The original Mpemba article quotes him with the following:
“If you take two similar containers with equal volumes of water, one at 35 °C (95 °F) and the other at 100 °C (212 °F), and put them into a freezer, the one that started at 100 °C (212 °F) freezes first. Why?”
So that might be a good starting point, although it hardly sounds to me like “cold water” vs “hot water” – more like “warm water” vs “very hot water” – but it does seem like a good place to start with an actual experiment that could be done in the home.
Two (identical) plastic cups: one @ 35 C , one @ 95 C (fresh out of a boiling kettle). Just went into the freezer.
This paper at arXiv.org, downloadable as a PDF at this link, goes into great detail about experimental conditions that result in hot water freezing faster than cold water.
Awesome, The Librarian. Science in action! Are you measuring which hits 0 °C first, or which forms ice first, or which freezes solid first? (The difficulty of measuring when something is frozen solid has stymied me so far.)
ETA: I will be at the local university on Tuesday morning, so I will be scanning a copy of Jearl Walker’s column from their physical Scientific American archives then, if nobody can find it first.
I’m going by the easily reproduced " if I can turn it over without spilling; it’s frozen".
Btw: after reading your (chorplers) link it becomes more obvious to me that mister Mpemba is full of it.
The conditions needed to reliably reproduce ( according to the article ) the effect are pretty far from anything available to a Tanzanian schoolboy.
Update: both cups still liquid:
Having read the whole article: I’ve also read its conclusion: all else being equal: there is no Mpemba effect.
Bullshitters be Bullshitting.
Update: frost has formed in the ‘cold’ cup.
An excellent, excellent paper, and supports what I have been saying all along. It supplies a plausible reason for the difference which I was unable to provide.
(my emphasis)
Exactly.
But that’s not the conclusion that the paper comes to – that’s just one particular set of circumstances it investigated. How are you and The Librarian reading that paper and still saying “There is no Mpemba effect”? The paper’s conclusion says (bolding mine):
That paper also shows that you could get different results depending on how you define “freezing.” The rest of the conclusion points out that, assuming the hot water does NOT change the environment (by, for instance, melting through coils or the like), it always reaches 0 °C after the colder water does – so if you define freezing as “water hits 0 °C,” you may not see any Mpemba effect. But if you define it as the point where the latent heat of fusion is released, as that paper does, you can get 100 °C water to freeze before 0 °C does, based largely on the different temperatures the water actually freezes at.
To my mind, Figure 19 is fairly important here – it shows that heating a sealed sample of water, whether it is tap water or distilled water, can cause its spontaneous freezing temperature to change. The paper text says: “We believe that [the major changes in spontaneous freezing temperature] are the result of the destruction and/or production of nucleation sites (motes) in the water, as proposed by Dorsey.[sup]6[/sup]”
Now I’m reading Dorsey’s paper, “The Freezing of Supercooled Water” from 1948. It should be noted that this paper, and several other places on the Internet, cites this paper as:
N. E. Dorsey, “The freezing of supercooled water,” Transplant. Proc. 38, 246–328 (1948).
But it’s not actually in Transplantation Proceedings, which as far as I can tell doesn’t go back to the 1940’s anyway; it seems to have started in the late 1990’s. Dorsey’s paper was actually published in Transactions of the American Philosophical Society, and is available on JSTOR at least.
Supercooled water exists.
Supercooled water != Mpemba.
There is no Mpemba effect where you can reliably freeze your icecream faster when it’s hot.
All claims referring to the icecream have no factual basis.
Mr. Mpemba witnissed a container of water getting supercooled and taking forever to freeze.
Dr. Denis G. Osborne was lazy or easily impressed. A little research could have found N. E. Dorsey or other works on supercooling.
There is no Mpemba effect
Perhaps there’s a disconnect in what we’re talking about here, but clearly under some circumstances hot water will freeze faster than cold water, so yes there is an Mpemba effect, since it seems like that should be impossible. As you point out, however, it’s certainly not absolute, or even reliable, except perhaps in specific circumstances like a particular freezer.