Fundamental question about temperature

[phinds]

I read in this weeks Time (or maybe it was Newsweek) a statement that I have heard numerous times over the years and dismissed as nonsense but it occurred to me now that it has been 45 years since I took college physics and maybe I'm remembering something incorrectly, although I don't think so.

The statements that I have heard go like this: you increase the temperature of something and that causes the collisions among the molecules to speed up.

Now it has always been my belief, and this is what I would like confirmation of, that it exactly the opposite. You do something that causes the collisions to speed up and that thing that we call "temperature" therefore measures a larger amount because fundamentally what it is measuring is the speed of the collisions.

Do I have that right?

 

[JeffKoch]

I don't follow, or understand if you agree with the article's statement or not, but temperature is a measure of the kinetic energy or moving things. Higher temperature means higher average kinetic energy, so for example if you heat up a pot of water the molecules are moving more rapidly. Twice the temperature above absolute zero means twice the energy, which is 1.4 times faster average speed.

 

[phinds]

I don't follow, or understand if you agree with the article's statement or not, but temperature is a measure of the kinetic energy or moving things. Higher temperature means higher average kinetic energy, so for example if you heat up a pot of water the molecules are moving more rapidly. Twice the temperature above absolute zero means twice the energy, which is 1.4 times faster average speed.

The article is saying that you FIRST increase the temperature somehow and that causes the molecules to move faster. What I'm saying is I think it's the other way around ... FIRST you make the molecules move faster and that causes the thing we measure and call temperature to be higher because it is measuring the movement of the molecules.

I understand that the two words/phrases (temperature and molecular movement) are tightly related but my point is the I believe that one (temperature) is really just a measurement of the other (molecular movement) whereas the article is saying that you create one (temperature) somehow and that causes the other (molecular movement).

In other words the authors of these statements believe that you create temperature WITHOUT moving the molecules faster and what you have done (whatever it is that they think that is) to raise the temperature then causes the molecules to move faster.

 

[JeffKoch]

If you *heat* something, the result is higher temperature. Temperature is a measure of kinetic energy, so causing something to have a higher temperature does not result in faster motion, it is faster motion.

 

[phinds]

If you *heat* something, the result is higher temperature. Temperature is a measure of kinetic energy, so causing something to have a higher temperature does not result in faster motion, it is faster motion.

Yes, that's my point. "temperature" is just the name we give to the measurement of molecular motion. The statements that I often have seen in non-scientific magazines are based on a belief that faster motion is the RESULT of something called "temperature" which we create without first making molecules move faster, and having done whatever it is that we did it then makes the molecules move faster.

My original question should have been more in the form of "in terms of temperature is molecular motion a cause or an effect."

I see everyone here agreeing with me that it is a cause in the sense that first we make it happen and then measure it, and then call the resulting measurement "temperature".

 

[klimatos]

I see everyone here agreeing with me that it is a cause in the sense that first we make it happen and then measure it, and then call the resulting measurement "temperature".

Not really. Most physicists consider "temperature" to be an intrinsic characteristic of a substance, independent of its measurement. That substance has that characteristic whether or not anyone is around to sense it or to measure it. Creatures in the ocean depths that we have never seen have a temperature, as do stars whose light has not yet reached the Earth.

 

[sweet springs]

Hi. plinds

Please read below my primitive understanding of temperature.

Temperature is index to give direction of heat energy flow. When two things of different temperature are in conatact, heat energy always flow from higher temperature thing to lower one.

On relation with heat energy contained in things and its temperature, amount of heat energy is larger when temperature is higher as for the same thing or material. Higher the temperature, the larger running and oscillating motions of molecules or kinetic and potential energy of molecules are. In the same temperature, however, amount of heat energy in the things depends on materials and conditions e.g. under the constant pressure, volume, mass or so on. For example one c.c. of water of room temperature contains more heat energy than one c.c. of hot plasma of 10,000 C.

I hope this will help you in some way.

Regards.

 

[phinds]

Not really. Most physicists consider "temperature" to be an intrinsic characteristic of a substance, independent of its measurement. That substance has that characteristic whether or not anyone is around to sense it or to measure it. Creatures in the ocean depths that we have never seen have a temperature, as do stars whose light has not yet reached the Earth.

Yes, I have no disagreement w/ this but don't see the relevance to my point. You are NOT, I"m sure, agreeing w/ the other point of view and saying that we somehow mystically raise a temperature without increasing molecular motion and then because we have done that the molecular motion then increases?

Perhaps I should have said "temperature is that thing that would be measured if it WERE measured)", which seems to me to be the same as saying that it's a fundamental characteristic of the thing.

This whole mess seems to be hinging much more on semantics that I had thought it was, but I still feel that the writers of the statements that caused me concern did not understand that you don't raise temperature and then AFTER you raise the temperature the moleculare motion increases as a result.

 

[phinds]

Hi. plinds

Please read below my primitive understanding of temperature.

Temperature is index to give direction of heat energy flow. When two things of different temperature are in conatact, heat energy always flow from higher temperature thing to lower one.

On relation with heat energy contained in things and its temperature, amount of heat energy is larger when temperature is higher as for the same thing or material. Higher the temperature, the larger running and oscillating motions of molecules or kinetic and potential energy of molecules are. In the same temperature, however, amount of heat energy in the things depends on materials and conditions e.g. under the constant pressure, volume, mass or so on. For example one c.c. of water of room temperature contains more heat energy than one c.c. of hot plasma of 10,000 C.

I hope this will help you in some way.

Regards.

Yes, I understand this completely. Thanks.

 

[Ken G]

What temperature actually means is something quite specific. Temperature has to do with under what circumstances will heat spontaneously (i.e., without intervention) flow out from a large system in thermodynamic equilibrium. Making heat flow out of such a large system comes at a kind of "cost"-- the number of different state the system could be in will go down by some factor. Since this factor depends on the heat Q that comes out, call it f(Q). It turns out from very general arguments that we can infer the function f(Q) must be e^-Q/kT where k is an arbitrary constant named after Boltzmann, and T is defined as the temperature of the large system (the real definition of temperature is actually an empirical one, but this is the theoretical counterpart to that definition-- what T "actually is" from the rationalist perspective). So that's how we figure out T-- we look at how the number of states a large system changes when we remove Q, and call that factor e^-Q/kT, and then read off T.

So yes, T is an attibute of a system (often related to the average energy per particle in the large system, but not always-- for example, a material that is creating a laser beam may be said to be at negative temperature, despite having positive energy content, because removing heat from it actually increases the number of states the material has access to, and that's why heat comes out of it in the form of the laser beam).

In typical cases, where T relates to the average energy per particle, we can really say it either way-- we can say that increasing the average energy per particle increases T (which would be appropriate if we knew we were adding Q somehow), or we can say that increasing T will increase the average energy per particle (which would be perfectly appropriate if we knew that we were bringing the system into thermal contact with some higher T that it was going to be brought up to as a result of the thermal contact). There is a law that says a smaller system that is brought into thermodynamic equilibrium with a much (much!) larger system will come to the T of the larger system, so that would motivate the latter usage, even though in other situations your objection would be very warranted. Oftentimes, the appropriate usage of the words depends not on the definitions of the quantities, but on the context of the situation.

 

[phinds]

Ken, that's good food for thought for me. Thank you.

Paul

 

[sambristol]

Almost – for gases and liquids the 'temperature' is just a measure of how fast the molecules are traveling but the velocity is proportionate to the square root of the temperature. For solids it's little more complex as the molecules are more tightly bound so they oscillate along their bond angles.

However if I put a beaker of water on a source of heat the glass molecules will oscillate and every time a water molecule bounces into one it will get a little kick of kinetic energy so increasing its speed.

This is not the full story – I have several books on my bookshelf devoted entirely to temperature

 

[Ken G]

Almost – for gases and liquids the 'temperature' is just a measure of how fast the molecules are traveling but the velocity is proportionate to the square root of the temperature.

No, the only general meaning of temperature is the one I gave. It is just that in many situations, like free particles in a gas, this connects to the average kinetic energy of the particles. In a fluid, there are interparticle forces that could cause the kinetic energy of the particles to scale differently than proportionally to T (and ultimately lead to things like phase changes when those interparticle forces undergo some kind of long-range transition). For example, if we used springs with an anharmonic potential to model the interparticle forces, we would not get the kinetic energy scales in proportion to T. That's why it's important to understand the general meaning of T.

For those who know astrophysics, there is a wonderful example of the complete breakdown of the concept that T gives the average kinetic energy of the particles-- white dwarfs. A white dwarf star could in principle cool to T near 0, but the particles within it would still be at kinetic energies corresponding to speeds of order a percent of c.

However if I put a beaker of water on a source of heat the glass molecules will oscillate and every time a water molecule bounces into one it will get a little kick of kinetic energy so increasing its speed.

Yes, that describes the microphysics of how the liquid will reach thermodynamic equilibrium with its surroundings. What I'm saying is, even if we have no idea what is going on inside some system (say a liquid with an anharmonic interparticle potential over large T changes, like van der Waals), we can invoke that it will eventually reach thermodynamic equilibrium when in thermal contact with a reservoir at T. So we can take as the causative agent the thermal contact to assert the temperature will come to T, and then we can use that causative agent (the thermal contact) to motivate what will happen microscopically once we know what is actually happening inside the system. In a sense that's "how nature does it"-- the reason nature will spontaneously cause those molecules to speed up is simply that it seeks to reach temperature equilibrium with the reservoir, which is in turn because temperature equilibrium simply has more "ways of happening"-- it accesses more states, so is analogous to "more area on the dart board."

 

[klimatos]

This whole mess seems to be hinging much more on semantics that I had thought it was, but I still feel that the writers of the statements that caused me concern did not understand that you don't raise temperature and then AFTER you raise the temperature the molecular motion increases as a result.

Yes, I agree that the issue seems to be semantic. Also, I can find no fault with your above statement. However, I also feel that you seem to be hung up on the question of time.

My own understanding of temperature is that the change in temperature and the change in kinetic energy are contemporaneous. They occur simultaneously and neither can either precede or follow the other.

 

[JeffKoch]

Well, this all can't help but be entirely about semantics, because even if we dispense with questions about causality we are still faced with a word, temperature, that is shorthand for some characteristic of a system. We have the statistical mechanics definition involving accessible states and thermodynamic equilibrium, we have the classical gas hard-sphere-interaction definition that is probably the most common, and there are others. In laser produced plasmas for example we talk about many kinds of temperatures - electrons and ions often have different temperatures, and each sometimes has several values of temperature depending on what part of the distribution function you're looking at, and the radiation inside the plasma has a different temperature defined by what blackbody temperature best fits the spectral emission in the part of the spectrum you care about. There are brightness temperatures, color temperatures, and so on. As long as everyone understands what you mean when you use the word, all is well.

 

[phinds]

My own understanding of temperature is that the change in temperature and the change in kinetic energy are contemporaneous. They occur simultaneously and neither can either precede or follow the other.

Well, for perhaps the 3rd or 4th time in this thread, YES ... this is EXACTLY my point. My belief is that this is the correct view and that the writers of the articles feel that it is NOT the correct view and that in fact FIRST you "raise the temperature" whatever that means to them and THEN the molecules, as a result of your having "raised the temperature" start to move faster.

 

[phinds]

Well, this all can't help but be entirely about semantics, because even if we dispense with questions about causality we are still faced with a word, temperature, that is shorthand for some characteristic of a system. We have the statistical mechanics definition involving accessible states and thermodynamic equilibrium, we have the classical gas hard-sphere-interaction definition that is probably the most common, and there are others. In laser produced plasmas for example we talk about many kinds of temperatures - electrons and ions often have different temperatures, and each sometimes has several values of temperature depending on what part of the distribution function you're looking at, and the radiation inside the plasma has a different temperature defined by what blackbody temperature best fits the spectral emission in the part of the spectrum you care about. There are brightness temperatures, color temperatures, and so on. As long as everyone understands what you mean when you use the word, all is well.

Gads ... the semantic complexity of talking about this was beginning to make me regret having started this thread, but I've learned a lot and definitely part of that was your post in the quotes above. I'm getting clear that temperature is a lot more complex than I had envisioned.

Thanks.

 

[Ken G]

Well, for perhaps the 3rd or 4th time in this thread, YES ... this is EXACTLY my point. My belief is that this is the correct view and that the writers of the articles feel that it is NOT the correct view and that in fact FIRST you "raise the temperature" whatever that means to them and THEN the molecules, as a result of your having "raised the temperature" start to move faster.

I don't think anyone believes that. Instead, when someone says you would (1) "raise the temperature if you want to cause the particles to move faster" or you would (2) "get the particles to move faster if you want to cause the temperature to go up", they are just expressing some larger contextual issue about the situation. The second version would be like "hold a bunsen burner under the liquid" or "put the liquid where it will absorb infrared light from a nearby campfire". The first version would be like "place the fluid into thermal contact with a hot reservoir and get it to come to equilibrium with it." In either case, the T and the motion come together at the same time, but there is a difference in which is the dog and which is the tail. Interestingly, it is really the first statement that is more formally correct, in terms of the kinds of assumptions made by the formal temperature concept, but as you've heard the term gets used in many more informal ways that make the second statement fine too, in the appropriate context.

 

[phinds]

I don't think anyone believes that.

Well, it's possible that I am underestimating the intelligence/education of the typical Time and Newsweek science writer, but I actually think it more likely that you are overestimating it.

 

[Ken G]

Perhaps I'm just an optimist!

 

[phinds]

Perhaps I'm just an optimist!

Fair enough. Could be I'm too much of a pessimist but I recommend that you keep an eye of the track record of reporters in general to say nothing of science reporters when it comes to sticking to the facts. I believe that some of them are absolutely scrupulous in intentent but many are not.

 

[Drakkith]

Fair enough. Could be I'm too much of a pessimist but I recommend that you keep an eye of the track record of reporters in general to say nothing of science reporters when it comes to sticking to the facts. I believe that some of them are absolutely scrupulous in intentent but many are not.

I think you're making too much out of this, but that's just me!