# Diffraction Gratings

Hi, I recently made a spectroscope using a pringles tube. I basically cut a very narrow slit at the metal end and at the plastic lid end I cut a circular shape out of a CD so that it fits inside the lid (I had to strip off the silver lining of course). I have some idea of how it works but would greatly appreciate some clarification and any extra information. Take the case of a sodium lamp that gives discrete spectral lines. As I understand it the light comes from the sodium lamp through the narrow slit and hits the CD inside the lid. The grooves on the CD reflect light at different angles depending on the wavelength which therefore split the light up. Since the silver surface is stripped off these reflected waves can transmit through into your eye. A couple of things I don't quite understand:

1) From what I have read, the reflected light rays can interfere constructively or/and destructively. So would you therefore not necessarily be seeing all the line spectra as some may have interfered destructively between the CD and your eye (if they were half a wavelength out of phase).

2) It also says that what you are seeing as line spectra are actually images of the slit. Is that right? Surely the light just comes through the slit, there's no image being formed is there?

mfb
Mentor
The grooves on the CD reflect light at different angles depending on the wavelength which therefore split the light up.
It is diffraction, not reflection.
So would you therefore not necessarily be seeing all the line spectra as some may have interfered destructively between the CD and your eye (if they were half a wavelength out of phase).
You see exactly the points where they are in phase. Under different angles you don't see light of this specific wavelength.

2) It also says that what you are seeing as line spectra are actually images of the slit. Is that right? Surely the light just comes through the slit, there's no image being formed is there?
Well, you still have the finite size of the slit and the light source that will give a finite width/height of the light patterns.

Drakkith
Staff Emeritus
The light coming in through the slit forms a narrow, rectangular 'column' approximately the same size of the slit. When this column of light hits the CD, the grooves separate the different wavelengths and the column, as a whole, spreads out. So the line spectra look like the slit only in the sense that the length and width correspond to that of the slit. So a very narrow slit vies you narrow lines, while a wide one gives you wide lines.

It is diffraction, not reflection.
You see exactly the points where they are in phase. Under different angles you don't see light of this specific wavelength.

Well, you still have the finite size of the slit and the light source that will give a finite width/height of the light patterns.

Thanks for the answers. I thought the light hits the grooves and different wavelengths reflect at different angles? So how is the light splits up if it not reflected. How is diffraction defined?

davenn