Homemade CD spectroscope - Part 2

In a previous post on home spectroscopy/spectrometry I outlined the use of CDs and DVDs as components of such instruments and showed it was possible to accurately measure the wavelength of a monochromatic light source (a cheap laser pointer) using those principles.

Here I'll elaborate on building such an instrument capable of splitting visible light into its constituent wavelengths with reasonable resolution and from no more than an empty box of cookies and a couple of pieces of CD or DVD.

It was already established in that post that there is a direct relationship between the wavelength (λ) of the light, the fineness (D) of the grating, the incident angle (β) of the incoming light, the angle (α) of the reflected light and the order of the positive interference (n) as follows:

λ = D ( sin β - sin α ) / n

It can be seen from the development in the previous post that for a given diffraction grating (constant D) at constant β (and thus constant sin β), λ is directly proportional to the distance O'C (for a given order n of spectrum). The reflecting diffraction grating will thus split up visible light in a linearly scaled spectrum.

This principle is used primarily in spectroscopes and spectrometers (the main difference between the two being that a spectroscope allows viewing of a given spectrum, whereas a spectrometer allows also measurements on that same spectrum).

It's rather self-intuitive that for the above reasoning to hold the light beam must consist of strongly parallel rays of light as otherwise it's impossible the impose the condition of constant β. Such light is referred to as collimated light and in the previous experiment the collimation problem was solved by choosing a pre-collimated source of light: laser light. But for any other form of light source collimation will have to be imposed in some way or other.

To the right is one of my homemade spectroscopes, constructed from an empty box of cookies, a piece of diffraction grating cut from a CD (or DVD-R) and an adjustable slit made from two safety razor blades. The slit provides a narrow beam of light, more or less collimated, which impinges on the piece of CD glued into place at the end of the box. The angle of incidence β is set at 60 degrees (the piece of CD angles 30 degrees with respect to the length of the box).

Here, in close up, the end of the box with the CD in place and the peep hole. By holding the slit-end of the box more or less close to any light source and looking through the peep hole, the linearly scaled spectrum of the light source can be observed. In fact usually at least two spectra can be seen: first order (n = 1) and second order (n = 2).

Here's the 1st order spectrum of a saver light bulb (NOT a traditional incandescent filament bulb), using a DVD-R as reflecting diffraction grating (D = 0.74 μm), photographed using the above design.

Most of the lines that can be seen are the emission spectral lines of the chemical element mercury (Hg) which is contained in saver light bulbs (as well as in the more conventional fluorescent strip lights) and in this sense the experiment is corroboration of quantum mechanics, which predicts that only specific electron transitions within a given atom are permissible. As a result the emission spectra of the chemical elements at modest temperatures are made up of discrete lines, specific to the element in question, much like a fingerprint.

From right to left in the above spectrum: the red line at 615 nm, a yellow/amber cluster that contains a 579/577 nm doublet, the green line at 546 nm and the far left blue line at 436 nm. The complete spectrum of mercury can also be seen at the bottom of this page using a high resolution spectrometer.

To the right the same spectrum photographed in near-identical conditions but using a piece of CD as diffraction grating (D = 1.6 μm). As predicted by theory, the spectrum is less wide (or less 'resolved') and clearly using the DVD-R as a grating is much to be preferred due to higher resolution.

But from here to high-resolution spectrometry is still a long way off. I'm currently building such a spectrometer which will feature better light management, better collimation, better light detection and thus hopefully significantly improved resolution. The instrument should resolve the major lines of most chemical elements, as well as detect the famous Fraunhofer lines in the solar spectrum, which provide proof of the chemical elements that make up our Sun.

For now I leave you with a few interesting resources to ponder:

A site that provided much of the inspiration for my posts on home spectroscopy. Very well worth the visit.

This resource provides a pdf with a hard card paper cut-out blueprint for a pocket size spectroscope, built in less than 1 hour and which you can take with you to observe the spectra of light sources around you: street lights, neon signs, light bulbs, sodium lights, etc. Ingenious.

This pdf provides the blueprint for a medium-resolution spectrometer made from hard card paper. Suitable also for a more permanent instrument in wood or plastic.