Optical spectrometer experiment

[LoD]

Hi,
I need to do an experiment to calculate Planck's constant from LEDs. Basically what I'm intending to do is a fairly simple variable dc supply->voltmeter->LED->spectrometer->diffraction grating->my eye type setup. I know how to do that, but the trouble is I'll need to track down a diffraction grating. What I'd like to know is, would it be possible to use a triangular prism instead and work out the wavelength using refraction? In theory it should, but would it in practice?
Thanks

 

[andrevdh]

Assuming that you know the energy gap $E_g$ of the LED you can construct a calibration curve with the prism by plotting the minimum deviation angles, measured with the spectroscope, for various spectral lines, against their wavelengths. Use this curve to determine the wavelength of the LED.