Author Topic: Freezing Speed of Hot Versus Cold Water  (Read 10747 times)

0 Members and 1 Guest are viewing this topic.

Offline Nominal Animal

  • Super Contributor
  • ***
  • Posts: 7192
  • Country: fi
    • My home page and email address
Re: Freezing Speed of Hot Versus Cold Water
« Reply #50 on: February 23, 2022, 05:38:42 am »
Quote
We conclude, somewhat sadly, that there is no evidence to support meaningful observations of the Mpemba effect.
Did you read the article?  The experiment was done by a single person, Henry C. Burridge, who furthermore redefined the effect:
Quote
A reasonable start to analysing the problem is to consider the process in two stages; first, cooling the water to an average temperature of 0 °C (or enthalpy equivalent thereof), and second, freezing the water to form solid ice. In so doing any effects associated with the supercooling of water are entirely contained within the second stage. We restrict our definition of the Mpemba effect to the first stage of the process, i.e. the process of cooling a sample of warm water to 0 °C in less time than it takes to cool a sample of water, which is notionally identical except that it is initially at a lower temperature, to 0 °C.

There is nothing reasonable in that approach, nor is it "notionally identical".   I do wonder how the reviewers let that pass, actually.

It is obvious that the phase change is a crucial part of the phenomena, because of conservation of energy (and how that relates to the latent heat and/or non-Boltzmannian distribution of energy in the degrees of freedom of each water molecule).  Indeed, one of the key points is to have a thermal bath well below the freezing point of water to even observe the effect.  (Why is this a requirement?  Consider the case where the rate of heat flow has to exceed some limit, before the effect can occur.  A typical flask or vessel does not have sufficient surface area [i.e. sufficient water skin to volume ratio] at something like 0°C to -15°Cfor the heat flow to suffice for the effect to occur.  To increase the rate, one can either increase the surface area to volume ratio, or simply use cooler heat bath, typically -20°C or colder (freezers).

Even if the heat bath is required to be much colder than the freezing temperature (and increasing surface area to volume ratio does not suffce), does not negate the existence of the effect; it'd just tell use something much more interesting about the phenomenon.

I shall assume all your other "counterarguments" are bullshit of the same sort, and will ignore you too.

There is nothing more I can tell people that actually matters, since the issue –– when and in which materials does the Mpemba effect occur –– has been sufficiently proven in my opinion, and arguments that are based on redefining the word to mean some other phenomenon are utter bullshit not worth discussing.
« Last Edit: February 23, 2022, 05:42:20 am by Nominal Animal »
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Freezing Speed of Hot Versus Cold Water
« Reply #51 on: February 23, 2022, 08:25:54 am »
I do wonder how the reviewers let that pass, actually.
That's the exact problem with peer reviewed science.

There are competing groups of nerds coming up with weird ideas and each patting their own guys on the backs to secure further public funding :P
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7453
  • Country: pl
Re: Freezing Speed of Hot Versus Cold Water
« Reply #52 on: February 23, 2022, 08:32:15 am »
As for the effect itself, I'm afraid I don't care enough to read a dozen papers and try to figure out which side is more wrong and whether perhaps both are talking about the same thing in different terms ;)

My earlier objection was purely to your own "layman explanation" of the thing. Based on your response I believe that you believe the alleged effect to indeed be a combination of the two factors I mentioned: reduced heat capacity (faster dT/dt for equal energy flow) and reduced thermal conductance of water at the interface with the container (is this what they meant by "something something heat diffusion something skin supersolidity something"?).

So I guess you can say that two bodies of water at equal bulk temperature release energy at different rates, but it's not by magic, but by thermal insulation and lower temperature of the outer layer of water due to some stupid factors.

Maybe it's possible, who cares. Post a video for us, the lazy crowd ;)
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Freezing Speed of Hot Versus Cold Water
« Reply #53 on: February 23, 2022, 09:05:50 am »
If you control everything other than the temperature of the water then once the water reaches the same temperature as the cold water, it can only take the same amount of time as the cold water to freeze. Claiming otherwise is claiming that the water has a memory effect.

That's...exactly what he said though.  o.o

So, if I understand this effect correctly, it would indeed make snow faster (nevermind the expense of heating all that water), because the transient heat capacity is lower, heat transfer is normal, and as a result it freezes faster.  BUT, it would make a rather dense pack, or just not freeze very well at all, because the excess heat still has to come out at some point, sooner or later, and, I guess it's going to be mushy, or partially frozen or something until then?

How does that work anyway, does it actually freeze solid and the ice retains some anomalous heat?  Even though it's a completely different phase?

And that should mean, you can shock-cool water, and then it will warm back up over some time frame as that excess heat returns to the normal sensible mode.  Which is a very strange thing indeed for a substance to do, but not thermodynamically inconsistent at least.  (Remember: despite what the name might seem to imply, thermodynamics is only concerned with end states at infinity: "dyn" refers to the fact that it's the exchange of heat at all, not that it's the dynamics (time dependence, kinetics) thereof!)

I suppose an easily dissociated (low energy), but kinetically unfavorable, chemical could exhibit the same sort of response.  Don't know of any offhand, but I'm sure something like that exists...

Which, hot-ice packs are kinda like that, but that's due to heat of crystallization and happens suddenly on nucleation, not due to a kinetically hindered reaction that happens gradually.


Quote
This is basically the same as claiming that a capacitor initially charged to 100V will discharge to 1V faster than one initially charged to 20V when you put the same value resistor across them. When the 100V capacitor has discharged to 20V it is in the same initial state as the 20V capacitor, therefore it can only take the exact same amount of time to go from 20V to 1V.

Doesn't it, though?

We need to develop the model a bit first though.  Consider the case of dielectric absorption.  We can discharge the capacitor immediately, delivering less energy/charge than it contains in total, the rest of which comes out more slowly.

This matches one aspect of the observation, but is also repeatable, from any voltage -- it's linear and time-invariant.  So we need some dependencies here.

If we make the ESR of the dielectric absorption dependent on voltage, we can have it so it charges quickly at high "temperature", but on "quenching" the main capacitance down to low voltage, it retains a lot.  (Maybe we make the resistance proportional to terminal voltage, for example.  Or more generally, some function like that.)

But this has the effect of merely varying the "heat capacity" with time and temperature.  To reproduce the effect of normal cold water having normal heat capacity, while quenched (previously-hot) water has low, we need another state dependence.

We could have a pair of fixed capacitances, modeling the anomalous plus remainder "heat capacities", plus an auxiliary C off to the side which represents the internal state.  C_aux might be arbitrarily small (i.e. just for modeling purposes, or not even connected to the component terminals at all -- a common SPICE motif, modeling auxiliary functions with respect to ground, connecting them to the main subcircuit by dependent sources), and its voltage would be coupled to the main terminals via resistor (possibly dependent again).  This could then have the effect that, for previously-cold water, C_aux is also "cold" so the dependent resistance is low and thus the total, nominal "heat capacity" is rapidly sensible; but if C_aux has been allowed to charge (i.e., it's coupled to the terminal voltage), then the remainder "heat capacity" becomes loosely coupled and is not rapidly sensible, thus reducing the transient heat capacity, until C_aux has discharged enough that the two main capacitances become tightly coupled again.  Note that, depending on how fast the C_aux dis/charges, nominal or anomalous heat capacity might be sensible at high temperature, or just over shorter time scales; depends on relative values.

Probably the same thing can be arranged from dependent (variable) capacitances as well, given suitable disambiguation of whether those should conserve charge or not (in SPICE for example, capacitance can just be some dependent source, no need to conserve charge; in reality, charge must be conserved and a reduction in capacitance causes an increase in terminal voltage), and possibly a fixed coupling resistance; maybe it still has to be dependent as well.

In any case, this can be drawn up fairly easily, given a differential equation for the system, and then you can have a capacitive sort of model to play with.


This thread is making me really depressed, to be honest.  So much "I don't believe you, because it is not how I believe things to be", instead of just checking out the peer-reviewd articles and studies by actual chemists and physicists.  Just "Because I cannot believe you, you must be wrong" type of arguments, and references to people who constructed a test setup where the phenomenon does not occur, and assume that must mean that the phenomenon does not exist.  No true counterarguments, just throwaway claims of "this violates that", without any basis for such claims beliefs.  None of this violates any of thermodynamics or anything else; I've even explained the most likely mechanism exactly how and why this happens above (with links to the underlying articles that supports that argument).

It is unfortunate; I sympathize.  Unfortunately I don't have nearly enough reading into this subject to give much of a technical review of the papers linked; they have problems, to me, but not enough to say whether that rises to the level of supporting or contradicting the claim.  To wit: the one with simulations, begs asking whether the observed effect is realistic in nature -- and if it has anywhere near such a time constant as to be relevant (min vs. fs is an extraordinary claim, to be sure!).  Or if they're simulating anything relevant at all, which, I'm just not familiar enough with the subject to understand what exactly they're concluding from it.  Which is itself a valid criticism: it seems not well enough explained, at least at the depth of reading I did.  Which is far from a fatal error, it's a common shortcut within a field; but obviously, that also limits its audience to those within that field.  And the other paper, I think is just too abstract to relate to something so [seemingly] ordinary as water?  I didn't gather much from it, it's mostly an equations of state kind of thing?

So, I'm open to the phenomenon, and mechanism; but like I said, it's an extraordinary claim, and spanning 17 orders of magnitude with one mechanism requires a lot more motivation and explanation than I've seen here.  For comparison, consider the decay of singlet oxygen: an extremely prohibited transition (due to quantum spin transition rules), so it persists for a relatively long time (seconds), typically being catalyzed by intermolecular interactions, or impurities, which break the symmetry and ultimately extract that energy, allowing decay to the ground (triplet) state.  Such a remarkable and (nearly) "iron clad" quantum prohibition would seem necessary here -- but neither paper is presenting such a claim (and, I'm not aware of such states in bulk water -- again, not that I'm any kind of expert on that!).

Tim
« Last Edit: February 23, 2022, 09:07:56 am by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline m k

  • Super Contributor
  • ***
  • Posts: 2651
  • Country: fi
Re: Freezing Speed of Hot Versus Cold Water
« Reply #54 on: February 23, 2022, 11:00:05 am »
I do wonder how the reviewers let that pass, actually.
That's the exact problem with peer reviewed science.

There are competing groups of nerds coming up with weird ideas and each patting their own guys on the backs to secure further public funding :P

There are also cognitive dissonances like confirmation bias.
One good example is Vera Rubin and Dark matter.

In this case there are also those ice-plus and ice-minus "organisms" clearing things.

That higher speed of battery charge by more parallel chemistry is not very old either.

BTW,
that video includes pretty simple and pretty statistical thing.
Advance-Aneng-Appa-AVO-Beckman-Danbridge-Data Tech-Fluke-General Radio-H. W. Sullivan-Heathkit-HP-Kaise-Kyoritsu-Leeds & Northrup-Mastech-OR-X-REO-Simpson-Sinclair-Tektronix-Tokyo Rikosha-Topward-Triplett-Tritron-YFE
(plus lesser brands from the work shop of the world)
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #55 on: February 23, 2022, 02:56:42 pm »
So, if I understand this effect correctly, it would indeed make snow faster (nevermind the expense of heating all that water), because the transient heat capacity is lower, heat transfer is normal, and as a result it freezes faster.  BUT, it would make a rather dense pack, or just not freeze very well at all, because the excess heat still has to come out at some point, sooner or later, and, I guess it's going to be mushy, or partially frozen or something until then?

How does that work anyway, does it actually freeze solid and the ice retains some anomalous heat?  Even though it's a completely different phase?

Either the low heat capacity condition related to the O-H bond length would have to both survive the phase change and then relax slowly enough that the heat could be dissipated to the environment without any melting or there would have to be an alternative form of ice, like Ice-9.  Or something I haven't thought of. 

Quote
So, I'm open to the phenomenon, and mechanism; but like I said, it's an extraordinary claim, and spanning 17 orders of magnitude with one mechanism requires a lot more motivation and explanation than I've seen here.  For comparison, consider the decay of singlet oxygen: an extremely prohibited transition (due to quantum spin transition rules), so it persists for a relatively long time (seconds), typically being catalyzed by intermolecular interactions, or impurities, which break the symmetry and ultimately extract that energy, allowing decay to the ground (triplet) state.  Such a remarkable and (nearly) "iron clad" quantum prohibition would seem necessary here -- but neither paper is presenting such a claim (and, I'm not aware of such states in bulk water -- again, not that I'm any kind of expert on that!).

The main issue I have with the theory is with applying the claim broadly without any broader experimental evidence.  I also have an issue with the conflation of skepticism with stupidity, but that's another matter. 

A very simple possible explanation of Mpemba is available based on supercooling.  If you supercool water, it may not always freeze quickly once it starts simply because the crystallization releases heat to the remaining liquid and additional heat rejection is required.  If your freezer isn't very cold--say -10C--and you have a sample of water that will supercool to -5C before freezing starts, then the latent heat of fusion will be transferred slowly, proportional to the 5C difference.  If you also have another sample of water that is crystallizing at 0C, then that system will be transferring the heat twice as fast based on the 10C difference.   So all you need is a mechanism by which one sample is susceptible to supercooling and the other is not, the right ambient temperature and some luck.  This would explain why the legend exists, but the phenomenon is not reliably observed experimentally, at least not in the ice-cube tray model.  This is why I suggested that any experimental measurement should be continued until equilibrium is reached and certainly not stopped at 0C.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #56 on: February 23, 2022, 03:34:16 pm »
So, I'm open to the phenomenon, and mechanism; but like I said, it's an extraordinary claim, and spanning 17 orders of magnitude with one mechanism requires a lot more motivation and explanation than I've seen here.

Actually, no. What it needs is an experimental procedure, clearly and precisely defined, so that any worker in any laboratory can reproduce the effect consistently.

So far, this does not appear to have been done. Where experimenters have claimed this effect, others have not been able to reproduce the same numbers.

Until there is a real, reproducible, observable phenomenon it is right to remain skeptical.

Computer simulations and theoretical models explaining how such a phenomenon might happen are interesting on an intellectual level, but they all remain hypothetical until they are able to give a precise specification for an experiment that will demonstrate in practice what the equations predict.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Freezing Speed of Hot Versus Cold Water
« Reply #57 on: February 23, 2022, 06:00:17 pm »
Actually, no -- it needs both!  An experiment is fine, but teasing out conflating variables can be tricky.  It's one thing to point at an experiment doing the thing and say "ha, see?", but conclusive proof requires both angles.

For example, point x-rays at it -- if it's bond lengths, then the diffraction pattern sure as hell ought to change.  It'll change anyway due to temperature (I would think?), but there should be an aberration in the distribution if it really is energy stored in bond lengths.

Or also, would that change density somewhat?  Enough to measure?  It's effectively a metastable phase of liquid water, that must count for something?

Another example of experimental ambiguity is just saying "oh, it freezes faster".  Well, what, it reaches 0C faster?  Or it drops below 0C (fully solid phase) faster?

A very simple possible explanation of Mpemba is available based on supercooling.  If you supercool water, it may not always freeze quickly once it starts simply because the crystallization releases heat to the remaining liquid and additional heat rejection is required.  If your freezer isn't very cold--say -10C--and you have a sample of water that will supercool to -5C before freezing starts, then the latent heat of fusion will be transferred slowly, proportional to the 5C difference.  If you also have another sample of water that is crystallizing at 0C, then that system will be transferring the heat twice as fast based on the 10C difference.   So all you need is a mechanism by which one sample is susceptible to supercooling and the other is not, the right ambient temperature and some luck.  This would explain why the legend exists, but the phenomenon is not reliably observed experimentally, at least not in the ice-cube tray model.  This is why I suggested that any experimental measurement should be continued until equilibrium is reached and certainly not stopped at 0C.

...As exemplified here.

Which is something else XRD would show, phase fractions.  With the ice phase giving sharply defined rings or whatever.

If the toss-a-pot-of-hot-water-on-a-cold-day trick is indeed an example of this effect, then it would seem the former is relevant -- given that the "control" doesn't freeze much or at all!

Of course you have to be careful that the ionization doesn't also cause nucleation, where supercooling might be involved.  But it would be nice to show it's not a supercooling phenomenon but something different entirely.

Sort of related: https://arxiv.org/ftp/arxiv/papers/1107/1107.4795.pdf

And, any other low-impact analyses would be effective, too; maybe Raman something or other, would be more "hands free" than ionizing radiation?  Or, there's a zillion other probing methods I don't even know about, probably some neat things to try among them...

Tim
« Last Edit: February 23, 2022, 06:05:06 pm by T3sl4co1l »
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #58 on: February 23, 2022, 06:53:58 pm »
Actually, no -- it needs both!  An experiment is fine, but teasing out conflating variables can be tricky.  It's one thing to point at an experiment doing the thing and say "ha, see?", but conclusive proof requires both angles.

Why would you say conclusive proof requires an explanation of a mechanism? For much of history, people knew that certain things happened, but they didn't know why. For instance, everyone knew for millennia that if you seared meat it got a brown color with a pleasing flavor. The mechanism for why it happened was not known, but the observation was nevertheless an incontrovertible fact. Only later was it discovered that it is due to a chemical reaction between amino acids and reducing sugars, and today we can call it the Maillard reaction.

Turn it around, and consider general relativity. Einstein made predictions based on a mathematical model, but those predictions were only a hypothesis until confirmed by observation. Without the real world observations, general relativity would remain an intellectual curiosity, rather than a testable theory.

The Mpemba effect can only happen if water or another substance exhibits thermal hysteresis. You can come up with a theoretical mechanism to produce such hysteresis, but then it should be possible to construct an experiment with the right elements to observe the effect in reality (as can be done with magnetic hysteresis). Without such experiments the effect is only "real" in the mathematical, theoretical sense. It is not real in the observational, experimental sense.

 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Freezing Speed of Hot Versus Cold Water
« Reply #59 on: February 23, 2022, 07:18:10 pm »
Yes, exactly -- I mean as a well proven, understood, scientific theory.  As opposed to a phenomenon (observation lacking explanation), or hypothesis (explanation lacking observation).

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 
The following users thanked this post: timenutgoblin

Online Siwastaja

  • Super Contributor
  • ***
  • Posts: 9333
  • Country: fi
Re: Freezing Speed of Hot Versus Cold Water
« Reply #60 on: February 23, 2022, 07:33:36 pm »
Why would you say conclusive proof requires an explanation of a mechanism?

It is absolutely nececessary to rule out coincidental unexpected correlations; if you have just measurement results and no explanation for them, you are not proving any theory, just replicating the experiment and it's well possible it demonstrates something completely different yet you have no idea.

I think it's quite revealing you don't understand such basic requirement of scientific research.

So yes, you need both. Only then it is valid science.
« Last Edit: February 23, 2022, 07:39:45 pm by Siwastaja »
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #61 on: February 23, 2022, 07:56:59 pm »
It is absolutely nececessary to rule out coincidental unexpected correlations; if you have just measurement results and no explanation for them, you are not proving any theory, just replicating the experiment and it's well possible it demonstrates something completely different yet you have no idea.

The causation/correlation issue only arises when you don't directly control the causation part.  If I take 1000 cuts of meat from 100 different animals and hang them over fire by a stick, I can safely conclude that the fire causes them to cook without understanding how or why.  It's true that not understanding 'why' might cause a mistaken conclusion to be made that does not allow me to predict the result in a slightly different experiment-- for example is the cooking due to the heat or the light?  I take the philosophical position that we never actually completely know "why" or "how" anything happens, we just have models that are good enough to allow us to make predictions with a certain amount of reliability.

If I can control various aspects of water in an experiment--how it is heated/cooled/handled and so forth, I can see if and when the Mpemba effect occurs and then make a catalogue of the various conditions where it does and doesn't seem to manifest itself, such as the water sample temperatures, the container characteristics, the freezer temperature, etc etc.  This catalogue might show me, for example, that the Mpemba effect is observed as long as the hot water is at least 80C, the container is metal but not wood or clay and only if the freezer temperature is between -5 and -15C.  That list of conditions where the effect does and does not occur would constitute scientific knowledge without ANY model or knowledge of the underlying processes, or even the atomic or molecular nature of water itself.  This is how science worked for thousands of years--there weren't many models and they were all completely wrong anyhow for the most part, except perhaps for simple mechanical devices.  Nowadays some think we have all that figured out, but I'm not so sure.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #62 on: February 23, 2022, 08:15:38 pm »
Why would you say conclusive proof requires an explanation of a mechanism?

It is absolutely nececessary to rule out coincidental unexpected correlations; if you have just measurement results and no explanation for them, you are not proving any theory, just replicating the experiment and it's well possible it demonstrates something completely different yet you have no idea.

I think it's quite revealing you don't understand such basic requirement of scientific research.

So yes, you need both. Only then it is valid science.

I think you are introducing something new here (the need for a scientific theory) that is not part of the original question. The question at hand is: "Can such an effect be observed by anyone, anywhere, in a reproducible manner?" In other words, is there even a phenomenon that needs explanation?

In this case I contend that no such phenomenon has been conclusively demonstrated in a laboratory, so there is not even a reason to seek a mechanism.

On the other hand, if someone hypothesizes a mechanism, then it should be testable by experiment. But I contend there is no more than anecdotal evidence or hearsay at the moment, with no reproducible, experimental procedure to demonstrate it.

Surely, if there was an experimental procedure that reliably produced the effect, then it would be documented somewhere? You or I could read the experimental method, and given an adequate laboratory we could reproduce it for ourselves. It would be in high-school science classes. There would be YouTube videos by the likes of NileRed or Veritasium showing it. Where are they?
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #63 on: February 25, 2022, 12:48:26 am »
Quote
but the ultimate question for me is whether I should use hot or cold water in my ice cube tray.

It's useful to remind us of the original proposition, but then why do you arbitrarily say:

Quote
f you observe the Mpemba effect, then at the end after they are cooled to say -20C

Why -20? Isn't the question merely which freezes faster? Otherwise you could pick -127C, and microwaved super-heated water to start, and all of a sudden the conditions for your 'ultimate question' are just as much rubbish as you complain his (alleged!) edge cases are.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #64 on: February 25, 2022, 02:24:17 am »
Why -20? Isn't the question merely which freezes faster? Otherwise you could pick -127C, and microwaved super-heated water to start, and all of a sudden the conditions for your 'ultimate question' are just as much rubbish as you complain his (alleged!) edge cases are.

Because that is approximately how cold my freezer is?

You can pick any numbers you like, and as I said elsewhere, it would be appropriate to try different combinations.  But -20C isn't what is important, it is the entire dataset as the cooling and freezing occurs, time and temperature.  The point is simply that you need to observe that long enough and far enough down that you can be sure that the phase change is complete.  Then you can go back and look at the data to see when you deem the sample 'frozen'.  If that point is clear and indisputable, great.  If not, you have more questions.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #65 on: February 25, 2022, 12:58:49 pm »
Quote
The point is simply that you need to observe that long enough and far enough down that you can be sure that the phase change is complete.  Then you can go back and look at the data to see when you deem the sample 'frozen'.

But in doing that you are merely picking some semi-random point at which you say "should be done now". The samples could have been completely frozen much earlier, or even later, and you won't know exactly when, but your confirmation bias will ensure the result you seek is apparent.

And in the end you won't be able to say why the repeatable effect shouldn't happen, only that it hasn't for your crude and unrepeatable experiment. OTOH, NA has quoted sources that, if the effect is real, can explain precisely how it can be real.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #66 on: February 25, 2022, 02:46:19 pm »
But in doing that you are merely picking some semi-random point at which you say "should be done now". The samples could have been completely frozen much earlier, or even later, and you won't know exactly when, but your confirmation bias will ensure the result you seek is apparent.

And in the end you won't be able to say why the repeatable effect shouldn't happen, only that it hasn't for your crude and unrepeatable experiment. OTOH, NA has quoted sources that, if the effect is real, can explain precisely how it can be real.

What do you mean by 'semi-random point'?  I think that most would agree that the completion of the phase change would be indicated by the temperature starting to fall again after a plateau.  How would you determine at what point the samples are indeed 'frozen'?



A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #67 on: February 25, 2022, 04:25:06 pm »
Quote
What do you mean by 'semi-random point'?

It's not completely random since it's limited to domestic refrigerator capabilities. Water freezes at 0C, so why not pick that? OTOH, if there are impurities then it will be a bit different, so how are you going to cope with unknown quantities of those?

Further, the lower ambient temperature would reduce the time available to observe the effects (and increase the likelihood of partial freezing). Shove the ice cube tray in liquid nitrogen and you'd be hard pressed to tall the exact time to the ms when it all froze. Use ambient of -5C and you have much much longer to determine things (and you won't have a crusty shell and fluid interior).

So, why -20C. No real reason other than that's what your freezer is set to. It's a semi-random figure.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #68 on: February 25, 2022, 05:31:03 pm »
Water freezes at 0C, so why not pick that?

Do you mean use an ambient (freezer) temp of 0C or terminate the experiment at 0C?  Either way, how can you determine that freezing has occurred?

Quote
So, why -20C. No real reason other than that's what your freezer is set to. It's a semi-random figure.

I haven't proposed that -20C is the correct or only temperature to use.  In fact I proposed a possible mechanism by which the Mpemba effect might be observed some times and not others and also that experiments  with varying freezer temps might reveal that.  Did you read that? However, -20C, in addition to being approximately what my freezer is set to, is also cold enough that supercooling is unlikely and you can be fairly sure that the water is completely frozen if you allow it to come near equilibrium.  So if I were going to conduct a set of experiments, in the first round at least I would use -20C as my coldest test and then successively warmer temps up to near 0C.  Unfortunately, both my freezers have a lot of food in them so I won't be doing this anytime soon.

Quote
Shove the ice cube tray in liquid nitrogen and you'd be hard pressed to tall the exact time to the ms when it all froze.

In that case it might actually not turn into conventional ice.  Ice-9 is fiction, but there is another amorphous state that occurs when you cool water so fast the the ice crystals don't have time to form before things stop moving.
« Last Edit: February 25, 2022, 05:34:34 pm by bdunham7 »
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #69 on: February 25, 2022, 06:42:07 pm »
Quote
Do you mean use an ambient (freezer) temp of 0C or terminate the experiment at 0C?

Terminate.

Quote
Either way, how can you determine that freezing has occurred?

Well, presactly. That's the problem, isn't it? And you not only have to determine that freezing has occurred, but when it occurred.

Quote
I haven't proposed that -20C is the correct or only temperature to use.

It was implied when you said: "but the ultimate question for me is whether I should use hot or cold water in my ice cube tray."
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #70 on: February 25, 2022, 06:50:11 pm »
Well, presactly. That's the problem, isn't it? And you not only have to determine that freezing has occurred, but when it occurred.

That is a problem with a pretty well established and accepted solution.  Freezing is complete when the temperature starts dropping again at the end of the plateau.  If that is a sharp curve, you know the point fairly precisely from graph of your data points.  But if you terminate the experiment at 0C, you can't possibly pinpoint the time when the water is completely frozen, at least not with a graph of temperature.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #71 on: February 25, 2022, 08:08:37 pm »
Quote
Freezing is complete when the temperature starts dropping again at the end of the plateau.

When your temperature probe drops you know the entire volume is frozen? Not just the part that your probe is embedded in?

 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 12537
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #72 on: February 25, 2022, 09:04:19 pm »
From an experimental point of view it is important that the sample is in a completely uniform state at the beginning and end of the experiment, so that its internal properties are everywhere the same. A practical way to achieve this is to have the sample well mixed. If you allow the sample to freeze, it becomes impossible to have it well mixed, and therefore impossible to know when it is uniformly frozen. For this reason, experiments should not allow the sample to freeze. They should go from liquid at one temperature to liquid at another temperature.
 

Offline PlainName

  • Super Contributor
  • ***
  • Posts: 7508
  • Country: va
Re: Freezing Speed of Hot Versus Cold Water
« Reply #73 on: February 25, 2022, 09:29:34 pm »
Quote
For this reason, experiments should not allow the sample to freeze.

I know what you're saying. Just thought it was funny, given the context  8)
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8175
  • Country: us
Re: Freezing Speed of Hot Versus Cold Water
« Reply #74 on: February 25, 2022, 09:30:36 pm »
When your temperature probe drops you know the entire volume is frozen? Not just the part that your probe is embedded in?

That's an issue for some experimental finesse and probe placement, but the general theory is that the temperature remains constant until the whole thing is frozen because any remaining liquid is still rejecting heat as it freezes.  OF course any real physical sample will have a gradient, but nobody said experimental physics was easy, even for simple things.  But this is not some new thing I'm proposing....

A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf