VII. Build it!
Put the components together.
VIII. Play with it!
Look at the spectra without a cuvette in place. Put in a cuvette and see what changes. See the various diffraction orders. If you don't like having the spectrum dispersed left/right, try rotating the grating and disperse the light at some other angle. Take pictures of the spectrum with your camera. See how you can vary the fraction of the frame that the spectrum fills. Is the spectrum too bright? Too dim? Is the room too bright? Can you tell where any particular wavelength is in the spectrum (the spectral plot suggests 460 nm and 560 nm should be discernable. Are they?)? Does the spectrum look anything like what you'd expect from the plot in §IV?
What will we use for a sample? Sitting in Urbana, I have no idea! We will need to find something that has a distinctive visible spectrum and that we can easily dilute to give transmittance between 10% and 95%. In our testing, we have found food coloring or sugary drinks with food coloring work well, but pH indicators or colored metal ions should be OK. We'll figure it out "on the fly."
IX. Trial: see how well 100% T can be reproduced
Either with no cuvette present or with just water in the cuvette, take 3 or 4 spectra. Later, we'll see how reproducible these data are.
X. Trial: see what 0% T looks like
0% T is what one measures when the light source is blocked (or turned off). So either turn off the LED or block its light and take 3 or 4 spectra to get the signal when there is "infinite absorbance."
Is there a problem here? Yes, as a matter of fact, there is. In a "real" spectrophotometer, when the light source is turned off, there really is very little light in the instrument (just leaks from the outside world, phosphorescence of any long-lived species, and any glow from semi-conductors or heated surfaces). But in this instrument? In a lighted room, there is a HUGE amount of light from the environment. If you take the instrument into a darkened room, the camera wants to use flash, another source of stray light. If you convince the camera not to use the flash, how can it focus if there's no light?
How will you deal with this stray light problem? Argue with your neighbor and see if you can figure out an equation to use with raw camera measurements to compensate for stray light.
XI. Get photometric data
Obtain a spectrum useful for measuring I0(λ). Then for each of 3 (or more) dilutions of the sample, obtain spectra from which you can measure I(λ). Name the pictures/files so you'll know which is which later.
Sample compound: ___________________________
Reference spectrum filename: ______________________________________
Sample Spectrum 1: ______________________________________ Concentration ______________
Sample Spectrum 2: ______________________________________ Concentration ______________
Sample Spectrum 3: ______________________________________ Concentration ______________
Will we have an unknown to analyze? I don't know ... . IF we have an unknown, run it:
Unknown Spectrum: ______________________________________ Concentration ______________
You might want to take multiple exposures in case some are blurry, badly exposed, or some such.