Image above: Taking good photos of the spectra produced by the Public Laboratory fold-up paper spectrometer.
Seeing Mathew's note about taking delivery on 2000 fold-up spectrometers put me over the top and I finally made one. I printed it on black construction paper, super-glued it together, and tried two different gratings: a piece of DVD-R and a piece of 1000 lines/mm diffraction grating film that I bought from (Edmund) Scientifics several years ago (it took me an hour to find it). I tried it with the only webcam I have (Microsoft LifeCam VX-6000) with acceptable results, but then I tried to figure out how each component contributes to the quality of the spectra. So I eliminated the webcam and took really good photos of the spectra with a DSLR and Micro-NIKKOR 55mm 1:3.5 lens. I tried three different entrance slits and both gratings and recorded spectra of my compact fluorescent desk lamp. The photos were uploaded to Spectral Workbench to make spectrographs (search for "clf_cfl").
The conclusion of the entrance slit trials is that narrower is better. The narrowest slit I made was about 0.2 mm wide. The 0.5 mm slit blurred the peaks quite a bit. The slit below is cut in black heat-shrink tubing material (not a great choice for flame spectroscopy).
The conclusion of the grating trials is that the DVD-R is almost as good as the grating film. I bought a square foot of the film for less than $10 (similar to this), so I have enough for 100-200 spectrometers. The spectrum displayed by the film is brighter than the DVD spectra, and I had trouble finding the first mercury line (404.7 nm) in the DVD spectra. But almost all the standard peaks below 650 nm are distinguishable with either material.
The spectrum below was made with an Ocean Optics HR2000 spectrometer that costs $4200. All of the peaks to the left of 650 nm are more or less distinguishable in the spectrum I made with the 1000 lines per mm grating film (peaks above 700 nm are in the infrared and are invisible to my DSLR). Only a couple of the peaks are missing only in the DVD spectrum. The calibration of the two spectra above differs by up to 5-10 nm (I guess because Spectral Workbench does a linear calibration and uses peaks 2 and 5).See source
Numbered Peaks above
|Peak number||Species producing peak||Correct wavelength (nm)|
|3||terbium from Tb3+||~485 to 490|
|4||terbium from Tb3+||~543 to 544|
|6||likely terbium from Tb3+ or mercury||576.960 for Hg or ~578 for Tb|
|7||mercury or terbium from Tb3+||579.066 for Hg or ~580 for Tb|
|8||possibly terbium from Tb3+ or europium in Eu+3:Y2O3||~580|
|9||likely europium in Eu+3:Y2O3||~587|
|10||likely europium in Eu+3:Y2O3||~593|
|11||likely europium in Eu+3:Y2O3||~598|
|12||europium in Eu+3:Y2O3||~611|
|13||likely terbium from Tb3+||~625|
|14||likely europium in Eu+3:Y2O3||~630|
|15||likely europium in Eu+3:Y2O3||~650|
|16||likely europium in Eu+3:Y2O3||~661|
|17||likely europium in Eu+3:Y2O3||~687-688|
|18||likely europium in Eu+3:Y2O3||~693|
|19||likely europium in Eu+3:Y2O3||~707 and ~709|
|20||likely europium in Eu+3:Y2O3||~712|
|21||likely argon||758.9315 or 763.5106 (??)|