Friction and Radiation, for example

Is there a single underlying causal mechanism giving rise to heat in both the case of friction and the case of the absorbtion of solar radiation?


Temperature is the measure of average kinetic energy of a materials particles. Is that what you’re asking?

No, he’s basically asking if the process by which friction causes increase in temperature is essentially the same, at some level, as the process by which absorption of radiation causes increase in temperature; that is, he is asking if they are but separate manifestations of a single physical principle.

Well, friction is particles moving across each other, changing bulk kinetic energy into particulate kinetic energy. Radiation causes particles to move faster, increasing their kinetic energy. That’s the only physical mechanism I can see that they have in common.

To be more specific, there are a number of dynamic interactions that fall under the moniker of friction, including Coulomb (static or sliding) surface friction, viscous fluid drag and aeroelastic interactions, continuum internal hysteresis (rheometric friction), turbulent diffusion, et cetera. All of these have the fundamental characteristic, as noted above, of converting the kinetic energy of a continuum or a system of continuous bodies into radiated heat energy from individual particles via interactions between adjacent molecules; that is, that the molecules form transitory bonds which are stressed by the kinetic action of the system; the system then radiates heat in excess of equilibrium by either imparting the dynamic motion on the molecules (in the case of fluids) or causing them to radiate photons in the (mostly) infrared range as the electrons involved in the interaction fall back into a reduced energy state. Conceptually, it is like a slingshot, transferring the energy stored into bands into the kinetic energy in the ammunition. Thus, this is a kinetic interaction governed by the first law of thermodynamics, and always produces some amount of energy in the form of non-recoverable heat as described by the second law.

Radiation emitted due to absorption of solar (or any other photon source) radiation is a different mechanism entirely; photons, which are bosonic carriers of the electromagnetic force, directly impinge upon the body in question. These energies are absorbed at the spectrum of wavelengths found in sunlight and the photons are destroyed in the process of elevating electrons to higher energy states. Because the system now has energy coming in, it must also have energy radiating back out to a lower temperature reservoir (the ambient environment, also a consequence of the second law of thermo). However, the spectrum of energy coming out of the body has a characteristic based not on the energy spectrum of the input but strictly on the amount of energy and the radiative characteristic emissivity and black body behavior as dictated classically by Kirchoff’s law of thermal radiation and the Rayleigh-Jeans law (suitable for most engineering thermal radiation problems) or more precisely (and appropriately at ultraviolet and shorter wavelengths) by the quantized Planck’s law of blackbody radiation.

It should be clearly and explicitly noted that heat itself is not a thing; that is to say, what we call “heat” describes a quantity of energy stored in some medium which in an isentropic system lacks any innate direction or momentum vector. When you throw a baseball, the kinetic energy of the baseball is not heat because it has a specific bulk direction and momentum which is easily determined. If you have an adiabatic (perfectly insulated) box that is full of hot gas at equilibrium, on the other hand, although there is plenty of energy in the box it has no overall momentum or direction in the system within the box; until punch a hole in the box giving what is in effect a low temperature (or pressure) reservoir for the energy to be directed to, the amount and direction of kinetic energy within the box is stochastically the same everywhere in the system. This is heat; energy without direction. The heat is not invariant, though; the energy can be given a direction via a heat engine, albeit only by pissing away some amount of the heat energy away in the process.


Clarification request. When the sunlight hits my skin, my skin gets hot. Why is that? And in what does the “hotness” consist in this case? I take it some particles have a lot of kinetic energy which taken all together adds up to no inertia in any direction. Which particles are those?


Radiant heat is electromagnetic radiation that carries energy. When that energy is transfered into a body (by body I mean a hunk of material, not necessarily a human body), it increases the kinetic energy of the particles in that body. In a gas, the kinetic energy means the molecules are actually flying around, randomly bumping into each other and the container walls (and the speed and frequency of the particles hitting said walls can be measured as a pressure). In a solid, the molecule’s kinetic energy is a vibration, and is based on how many degrees of freedom the substance has (this is based on the type of bond, amongst other things). Here’s a site that describes the different modes of vibration for a carbon dioxide molecule, which has similar dynamics to other diatomic and triatomic molecules.

Thanks for the responses so far.

Someone has said there is no single underlying mechanism giving rise to all instances of heat. I am going to reply to them by arguing that there’s no single kind of cause giving rise to all instances of heat, but there is a single underlying mechanism, to wit, a bunch of things moving around individually but not moving in any single direction when taken as a group. (Sorry for the simplistic language but I think that captures it? I’m trying to word it in a way which makes it clear why it’s appropriate to call it a “mechanism.”)


Heat is the same. Radiation is a type of heat transfer. Friction is a type of heat generation. Once a material gets hot (relatively), there’s no fundamental difference based on if it was heated via radiation, conduction, convection, or rubbing.

I think the “mechanism” you’re looking for is the molecular vibration. That doesn’t happen until after the heat has been transfered or generated, though.

I feel it only prudent to re-emphasize the fact that heat is not a “thing” that can be uniquely identified or separated from the medium in which it is measured. It isn’t a substance, or an innate property of a substance like mass; it is a derivative (and relative) property, which becomes clear when you go into the mathematics of thermodynamics because everything is described in terms of differential equations. Talking about “instances of heat”, except as abstracted from reality in a purely conceptual context, is a semantic null. One might as readily speak of an “instance of momentum”, which makes no sense except in the context of an object being pushed or thrown.

The most general definition that can be given of heat is that it is randomized energy that is transmitted from one area to another strictly on the basis of a difference of temperature. Heat as we typically experience is particles in motion (hot air, scalding water, et cetera), or particles that are constrained in some kind of solid continuum but which “vibrate” (sometimes literally vibrating, like a the resonance of a piezoelectric crystal). Heat could also be non-polarized photons bouncing around in a mirrored box, or a group of interacting fields, or stochastic motion in any other medium that interacts via the electromagnetic force.


Prudence is generally wise, but to be a bit frank I haven’t understood why you have felt the need to be “prudent” to quite this extent. :wink:


How’s this for an answer satisfying the OP’s suspicions:

Think of magnets that are epoxied into the springy part of a mattress from which all the fabric and rubber has been removed. And let’s make the springy part out of some nonmagnetic metal. If you move another magnet along the surface of the mattress skeleton, it will push and pull on all of the magnets, mostly with very tiny forces but sometimes much bigger as it gets close to certain magnets. This movement jostles the springy frame and, eventually, sort of randomizes into an indescribably complex movement of all the magnets. Imagine sliding two matress-magnet assemblies past one another. Same thing, in both assemblies. We’re not letting them touch, we’re just moving them around close enough to influence each other. This is a model for friction, in which the magnets are atoms, and they influence each other at a distance but preferentially at the smallest distances. There’s a curve describing the energy in a system of just two atoms influencing each other, the Lennard-Jones function (Lennard-Jones was one guy BTW).

OK, now, move the assemblies further apart, and really jiggle hell out of one of them. The mutual influences will still work. You could think of it as the jiggling one radiating energy and the other one absorbing it, or you could still think of it in terms of individual magnets sharing joint influence over each other. This second way of thinking of it is very hard to use in calculations, but in principle is equally valid.

In both cases, action at a distance, transmitting mechanical force from magnet to magnet, making magnets jiggle, and the jiggles randomizing into a complex distribution of tiny movements in an assembly, is a pretty physically correct mental picture, that is common to friction and radiation.

The biggest cheat here is that I used the skeleton of a mattress for a semirigid mounting mechanism to keep the magnets suspended near each other. It would have been more accurate if we could use a kind of magnet that attracts others like itself until they reach a certain closeness, and then started repelling instead, so I could have had a loose assembly of magnets without any springs between them, their attraction-repulsion curves being the only springs in the system. Atoms and molecules are more like that. There are different names for the forces, such as the van der Waals force, the hydrogen bonding force, the ionic bond force, and so forth, and chemists set great stock by distinguishing between them, but they are all Coulombic attraction/repulsion, and generally they are all present in different degrees.

I just love you guys!

I think the mattress analogy would be better if they were to actually touch in the friction example, because the materials actually touch in the friction case. I don’t really like it, though, because how would one use it to describe conduction and convection?

>the materials actually touch in the friction case

No, they don’t. Or more precisely it depends on what you mean by touch, but on the atomic scale at which you would have to consider the problem, particles don’t have distinct surfaces and don’t act as if they do anyway. When they get closer, they stop attracting and start to repel, but you can keep pushing them closer and closer if you want, and the force just gets much bigger, maybe as the 6th power of the inverse distance.

In other words, although the ancients had a hard time swallowing the idea of action at a distance, it turns out that ALL action is action at a distance, and the appearances of “solid objects” and “touching” are large-limit approximations or, if you like, illusions of our macroscopic world.

I actually first started thinking about this analogy to explain conduction. In the case of conduction, you grab the magnets on one face of the assembly and start jiggling them with rapid short random movements, with (I think) a Boltzmann distribution of square roots of velocities. Pretty quickly magnets on all the other faces will also be jiggling.

Convection is also pretty easy to explain, if you imagine getting magnets jiggling on an assembly and then transporting it to another location where the resident magnets jiggle less. It will get them jiggling harder.

Free, buoyant convection relies on an ambient gravitational field and on thermal expansion. The magnet matress assembly model isn’t specific enough to explain thermal expansion, but the Lennard-Jones curve is. It’s not symmetrical. A distribution of distances between atoms in a solid or liquid will have an average distance. Note, though, that if you change the dispersion, the width of that distribution, the skirts of the distribution will not move equally because the curve isn’t symmetric. Typically, jiggling harder won’t get their closest excursions much closer, but it will get thier furthest excursions a fair bit further. This is because the curve is very very steep as you get closer than the atoms like, but it gets flatter and flatter the further you get. So, the average atomic distance is an increasing function of the dispersion of the distribution of atomic velocities or kinetic energies (which two things are just different powers of one another).

>I just love you guys!

This kind of brouhaha really gets you going, doesn’t it?

If they’re not touching, it’s not friction. By definition. Saying that is like saying that cars in a crash don’t touch each other, either.

Ah, this brings back old memories: [thread=299054]Why can’t my hand go through my desk?[/thread]

Napier is correct in the domain that he addresses; when you get down to the level where individual molecules or atoms individually interact, the concept of “touching”, and indeed, even discrete particles becomes an abstraction. The interactions actually occur between the fields that comprise the atoms; or rather the fields that comprise the electrons in the outer orbitals of the atom since the inner electrons and nucleons of different atoms don’t interact under anything like normal conditions. At this point the probabilistic nature of quantum mechanics comes into play, and the gross concepts that allow us to consider a material to be a continuum with definable boundaries no longer work.

However, for any problem on any scale that we would be able to directly experience, the decoherence of such a large system allows the concept of “solid objects” and “surface friction” to work just fine. The thermodynamic concepts of heat and entropy (the diminishment of identifiable states of order) take on entirely new formulations at the quantum scale, and trying to discuss these in polite company inevitably results in someone spilling their gin & tonic, so we won’t go there this evening.