As promised (threatened?) here's my attempt to explain the basics of spectroscopy:
A
As talked about in a previous post, hot things emit radiation. This radiation has long wavelengths for objects less than a few hundred degrees C, and it's not visible to the naked eye.
This radiation is called the Infrared part of the spectrum, and you need a special camera to see it. That's how you can see criminals glowing in the dark when filmed running across gardens in the dead of night in those Police Stop! videos. When objects get really hot, the wavelengths of the radiation gets shorter, and at a certain point we can see things start to glow. If you've seen glass or metal in a furnace or Bunsen Burner, you'll know what colour it starts glowing; a deep red colour. As it gets hotter still it turns orange, yellow, and eventually blue/white. When it's completely white, most of the radiation is in the very short, (and invisible to us) wavelengths of the ultraviolet part of the spectrum.
The first important point here is that a hot body emits a continuous spectrum. In other words it produces a whole rainbow of colours without any spaces in it. However it emits the majority of its light in the part of the spectrum proportional to its temperature.
Summary: the colour of a hot body tells you its temperature.
For more, see here: http://en.wikipedia.org/wiki/Black_body
B
Point A is really great new for astronomers because of course they have no way of visiting a star for the foreseeable future, but simply by looking at a star's colour they can work out its temperature. You can do the same on any clear night; look at the stars and impress your friends by pointing out that the bluer stars are hotter than the yellow ones, and the yellow ones are hotter than the red.
Spectroscopists just take things a little bit further. By spreading the continuous spectrum of light out, they can identify the exact wavelength where most of the energy is being radiated, and using a simple formula, give you the star's temperature. Actually there are other factors to take into account but that's the basic idea.
Here's my spectrum of the star Betelgeuse in Orion. Without any processing you can see that it's brighter at the red end than the blue, which tells you it's a colder star than a blue one.
And again repeating a previous post, young stars are hot and old stars are cooler, so you can tell what point a star is in its lifecycle. Compare Sirius to Betelgeuse:
You can see that Sirius is a hot young pup, and Betelgeuse is a tired old dog.
The colour of a star tells you its temperature and its age
C
If that was all that spectroscopy could do, it wouldn't be the impressive science it is. In fact ,it's been argued that spectroscopy is responsible for 75% of all of our knowledge of the Cosmos.
How has it managed to achieve this? By a single lucky effect resulting from Quantum Mechanics. Coming soon...
As well as being a supremely useful way of gathering information from the entire visible universe, spectroscopy was directly involved in the formulation of quantum theory; which radically changed the way we understand everything.
As Iain has described, spectroscopy works because objects at different temperatures radiate different wavelengths - the general term for this is "black-body radiation", and has been known about for over a century.
However, until 1900, the formulae which should have predicted the relationship between the wavelength of the emitted light, and its radiancy showed the latter rising to infinite as the wavelength grew shorter.
This, obviously, doesn't happen in real life, a failure of prediction which was known as the ultraviolet catastrophe.
In 1900, the German physicist Max Planck proposed a bold and innovative resolution to the ultraviolet catastrophe. He reasoned that the problem was that the formula predicted low-wavelength (and, therefore, high-frequency) radiancy much too high. Planck proposed that if there were a way to limit the high-frequency oscillations in the atoms, the corresponding radiancy of high-frequency (again, low-wavelength) waves would also be reduced, which would match the experimental results.
Planck suggested that an atom can absorb or re-emit energy only in discrete bundles (quanta). If the energy of these quanta are proportional to the radiation frequency, then at large frequencies the energy would similarly become large. Since no standing wave could have an energy greater than kT, this put an effective cap on the high-frequency radiancy, thus solving the ultraviolet catastrophe.
There's more on this here:
http://physics.about.com/od/quantumphysics/a/blackbody.htm
Posted by: David | January 29, 2011 at 07:56 AM
Thanks David. Funnily enough, I'm just sitting here about to try to explain the link between Spectrosocopy and Quantum Mechanics for Part C of my opus, but it looks like you've done the job for me.
The temptation to delete your name, cut and paste your words and claim them as my own is almost overwhelming...
Posted by: Iain | January 29, 2011 at 10:54 AM