Why are incandescent lights white?

October 7, 2008 7:26 PM

Why does an incandescent light bulb have an broad, continuous spectra when viewed with a spectrometer? Why doesn't it have some lines like a sodium lamp? I guess I'm asking why photons come off of a black body and why are their wavelengths spread out. Bonus points for clarity and accessibility of the answers.
posted by bdc34 to science & nature (5 comments total)

The spectrum emitted by a black body(or your light bulb filament, which can be considered - as pretty much everything - an approximation of a black body) is continuous because it's a statistical phenomenon; Planck's law (valid for a black body) states that the wavelenghts emitted in thermal radiation are distributed all along the spectrum, and their distribution has a peak that moves up (to shorter wavelenghts) with the temperature.

So, at room temperature your filament absorbs and emits mostly infrared light, while with the temperature going up, visible wavelenghts start to show up, first mostly in the red, then orange (think of a light bulb connected to a dimmer), yellow and so on, and you see a coloured light which is the sum of all these wavelenghts, including a huge part of them in the infrared (most of them, actually, for a light bulb).

Gas discharge lamps, on the other hand, emit light by way of a completely different mechanism: the electrons in the gas (neon, sodium, mercury) are excited by an electric discharge and emit that energy back when they leave their excited state. Those electrons can only move stepwise between discrete energy statuses, and every time they jump down, they release a photon, the wavelenght of which depends on how high the energy "jump" was (pretty much as strings on a stringed instrument can only vibrate to a given frequency), so you get emission lines on your spectrum.
posted by _dario at 8:45 PM on October 7, 2008 [2 favorites has favorites]

Thanks for asking this question, it's something that's been bugging me for a while.

_dario, that's pretty much what I remember from A-level physics. But what physical process takes place in a solid or high-pressure gas that causes the photons to be emitted with a continuous range of frequencies, instead of electrons transitioning between discrete orbitals? Is it a result of interatomic collisions, or something else?
posted by teraflop at 9:02 PM on October 7, 2008

short answer: in a solid, atoms wiggle each other. in a gas, they do not (much).

since you're asking about black bodies, you're pretty close to the answer. as _dario describes, the lines from a gas lamp come from the electronic energy levels of the atoms in the gas. these atoms can be treated as more or less independent (but identical) entities with a few, well-defined energy levels, and therefore a few, well-defined emission lines.

with solids, however, the atoms are strongly bound to each other and there are a lot more degrees of freedom (read: possible energy levels) per atom. most thermal energy is manifest in a solid by vibrations of atoms around their equilibrium positions. these vibrations couple to nearby atoms and the whole thing wibbles and wobbles as you turn up the temperature.

so atom A may wiggle in phase with atom B, but opposite atom C. or atom A may wiggle in phase with atom C, but opposite B. or they may all wiggle in phase together. now consider an avogadro's number of those atoms and all the possible combinations of vibrational modes. lots of different energies, all slightly different.

so, as with all things, strictly speaking, the vibrational energies are not really continuous either, though for a large collection of atoms the spacing between nearby levels is very very small, so that it appears continuous.
posted by sergeant sandwich at 9:07 PM on October 7, 2008 [4 favorites has favorites]

Gas discharge lamps, on the other hand, emit light by way of a completely different mechanism: the electrons in the gas (neon, sodium, mercury) are excited by an electric discharge and emit that energy back when they leave their excited state.

And in an LED, light is emitted when a mobile electron falls into a hole. So the frequency of light emitted is the function of the energy difference for that electron between being mobile and being in that hole, which I think is referred to as the "band gap".
posted by Class Goat at 10:14 PM on October 7, 2008

the continuous range of frequencies, as Sgt. Sandwich has it, is a consequence of the fact that thermal radiation is a large number phenomenon: energy comes in, the interatomic bond between two atoms in a crystal lattice is weakened, or broken for a fraction of a second, then when it reforms energy comes out again, as electromagnetic radiation. If you apply it to large numbers, you get a continuous spectrum.

w/r/t LEDs, another factor comes into play, which is if the ability of the p-n junction to have a direct (light is emitted) or indirect (no light, as regular GaAs diodes) band gap.
posted by _dario at 11:50 PM on October 7, 2008