1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Why hot objects glow

  1. Mar 30, 2012 #1
    This is from a physics textbook

    All objects radiate energy continuously in the form of electromagnetic waves due to thermal vibrations of their molecules. These vibrations create the orange glow of an electric stove burner, an electric space heater, and the coils of a toaster

    Is the orange glow caused by electrons descending an orbit and releasing a photon? If so, why are the electrons doing that? Maybe when a slow electron bounces into a fast electron it gains energy and hence the electron ascends to an excited state.
     
  2. jcsd
  3. Mar 30, 2012 #2

    Rap

    User Avatar

    Its the atoms and molecules colliding with each other that raises the electrons to higher energy levels. The higher the temperature, the more energy is likely to be transferred in a collision. The photons that are emitted can then be absorbed by other particles, raising electrons to a higher energy level. These excited particles can then release photons, possibly at different energies than the incident photon.
     
  4. Mar 30, 2012 #3
    You are confusing "spectral line emission" and "thermal radiation". Spectral line emission is the light given off when an electron transitions between atomic levels. Thermal radiation is the light given off when entire molecules bump into each other due to their thermal motion and slow down. Spectral line emission is highly material-dependent, whereas thermal emission is highly temperature dependent.

    Everything glows, not just hot objects. Most things glow in the infrared frequencies, which human eyes can't see. When an object gets hot enough, it glows in the visible light frequencies, which we can see. This "glowing" is called thermal radiation, meaning electromagnetic radiation created because of thermal motion of the molecules. All the molecules in an object are constantly bouncing around randomly, what we call thermal motion. The temperature of the object is the average kinetic energy of these molecules bouncing around, but its just an average, some are moving faster and bouncing harder, and some molecules are moving slower and bouncing softer off each other. A higher temperature means on average the molecules are moving faster. Each time two molecules collide, some of their kinetic energy is lost to a bit of light that gets emitted: thermal radiation. Because there is a broad distribution of speeds to the molecules, there is a broad distribution of frequencies of the emitted thermal radiation.
     
  5. Mar 30, 2012 #4

    Hurkyl

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I wanted to add to this particular statement, to diminish the possibility of confusion. Such an object is still glowing in infra-red frequencies along with the visible frequencies. This is why, for example, incandescent light bulbs get hot: most of the energy they output are infra-red frequencies, which readily get converted into heat.
     
  6. Mar 30, 2012 #5

    Drakkith

    User Avatar

    Staff: Mentor

  7. Mar 30, 2012 #6
    Hmm, I don't agree. Thermal radiation can be thought of as a statistical limiting case of the full emission model, which takes into account all sorts of collision, excitation, de-excitation, emission, and absorption processes. Thermal radiation is not a process in itself, and doesn't explain _how_ the emission was generated. Rather, it's the statistical result of many processes together, which "magically" give a result similar to a blackbody curve.

    It's wrong to treat spectral line emission and thermal emission as two completely different types of emission, since thermal emission will usually include some amount of spectral line emission, as well as continuum processes like changes in vibrational and rotational states, bremsstrahlung, and many other processes.
     
  8. Mar 31, 2012 #7

    Rap

    User Avatar

    I'm not confusing spectral line emission with thermal radiation - under normal circumstances, much thermal emission is due to spectral line radiation. When you say "some of their kinetic energy is lost to a bit of light that gets emitted", what is the exact mechanism you are referring to?

    It is the result of a particle absorbing some of the kinetic energy by having an electron move to a higher energy state, then dropping down and emitting a photon. Ok, in a metal, the electrons are not bound to any particle, but they still have a set of energy states, and a collision will put them in higher energy states, and when they drop down, they emit a "thermal" photon.
     
  9. Mar 31, 2012 #8

    Drakkith

    User Avatar

    Staff: Mentor

    Charged particles emit electromagnetic radiation when they are accelerated. So when the interact and speed up or slow down through collisions and such, they emit some type of light.

    That is only one way. Electrons and ions are always moving around and vibrating and changing orientations and such. As I said above, when they get accelerated they emit light. This is on top of electrons dropping from higher energy levels in their orbitals, so the effects add.
     
  10. Mar 31, 2012 #9

    K^2

    User Avatar
    Science Advisor

    Transition radiation can also be seen as radiation due to acceleration. If you compute <a> for a transition between states, you'll get a non-zero value, while for any pure state, <a>=0.
     
  11. Mar 31, 2012 #10

    Rap

    User Avatar

    Well, thinking about it, I guess there are many ways. In a metal there are free electrons that get accelerated by collisions. In a neutral material, there can be line radiation from individual atoms, but there can be molecular vibrations too, which are not necessarily electrons jumping from energy state to energy state.
     
  12. Mar 31, 2012 #11

    Ken G

    User Avatar
    Gold Member

    Although the OP was focused on the details of the processes involved, I think what we don't want to lose sight of is the fact that the whole beauty of thermal radiation lies in the fact that it doesn't matter at all what mechanisms are creating the light, or whether we treat those mechanisms classically or quantum mechanically. Those kinds of possibilities and choices have dominated the thread so far, but isn't the main point that we don't care? As long as there are many different processes and timescales involved, you have a kind of "mixmaster" of light-generating mechanisms, and so the thermodynamical concept of maximum entropy comes into play and you have to get the same answer whether you use classical or quantum mechanics to describe what is happening. This "mixmaster" is in turn what "causes" the radiation field to be thermalized, at the appropriate temperature. So the fact that the metal in a lightbulb filament is undergoing processes that are completely different from the processes that hydrogen gas in the Sun is undergoing makes no difference-- if you heat a light bulb filament to 6000 K, it acts just like the surface of the Sun does, when it comes to making visible light.
     
  13. Mar 31, 2012 #12

    haruspex

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    As some have remarked, it's not enough for molecules to bump into each other (or for atoms to vibrate within a molecule). Electromagnetic radiation can only arise/be absorbed when there's a net acceleration of electric charge.
    This is why N2 and O2 are transparent to IR as well as to visible wavelengths. When these diatomic molecules vibrate, there's no net acceleration of charge. Each atom is neutral, so vibration won't do it, and visible light is not energetic enough to ionise them.

    CO2 and H2O, however, are polar, so internal vibrations do constitute charge acceleration. Quantum effects constrain the frequencies and modes, so emission/absorption bands are fairly narrow. These get broadened by Doppler effects.

    For any material, at sufficiently high temperatures ionisation occurs. Since the energy of a free electron is not constrained to specific values, the spectrum is continuous above the energy required for ionisation. Below that there are band gaps.

    A 'black body', freely emitting and absorbing at all wavelengths, is purely theoretical, but a rich mix of compounds in the surface might achieve a decent approximation.

    Wrt the original question:
    As explained, it isn't true that all objects radiate all the time.
    Metals will because they have free electrons.
    Salts will because of their ionic bonds.
    Most complex chemicals and their mixtures will because they contain salts or polar molecules.
     
  14. Apr 4, 2012 #13
    The OP did not ask about black bodies. He asked about thermal radiation. While blackbody radiation is a theoretical limit that is never perfectly attained, thermal radiation just means the electromagnetic radiation emitted by a body that is largely temperature dependent. True, different substances will have different thermal radiation spectra, that does not make the radiation any less real, or any less tied to the temperature.
     
  15. Apr 4, 2012 #14
    I like this description (although most objects emit thermal radiation without being true blackbodies). My original point was that thermal radiation is a collectivist phenomena, where many molecules are involved, as opposed to spectral line emission which involves a single transition in a single atom. Yes, thermal radiation includes spectral line emission but only is one step in the process. I oversimplified things on purpose to try to address the level of the OP, but my original comment still stands that the OP should think of thermal radiation as the result of the interaction of many molecules throughout an object and not one electron making one transition in one isolated atom.
     
  16. Apr 4, 2012 #15
    I see what you are getting at, but it does matter whether you approach thermal radiation classically or quantum mechanically. Treating it classically leads to the ultraviolet catastrophe.
     
  17. Apr 4, 2012 #16

    Ken G

    User Avatar
    Gold Member

    I mean "treating" the processes that create the light, not the light itself (the latter does need to be treated quantum mechanically, but the thread was worrying about the treatments of the light-creating processes). The UV catastrophe is avoided entirely by treating the light in a quantum way, even if the processes that create the light are treated classically (indeed, no treatment of those processes is needed at all, other than the assertion of the concept of temperature of the source, which suffices to do all the required statistical averaging). But you are right that the light itself must be treated as comprising of quanta of energy, or else one does have a big problem at high frequency! Interestingly, the mistake of thinking that what was crucial was the quantum nature of the processes that make the light, rather than the light itself, was why Planck did not appreciate the need for light to be a particle, and was what left the door open for Einstein to get the Nobel for the photoelectric effect rather than just giving it to Planck (who got it for the quantization of energy in the light-creating processes rather than in the light itself).
     
    Last edited: Apr 4, 2012
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Why hot objects glow
  1. Why does hot air rise? (Replies: 66)

  2. Hot objects (Replies: 10)

Loading...