In 1955, it was proposed that hollow cathode lamps are good emission sources for monochromatic radiation, and thus for atomic absorption spectroscopy (AAS). In looking into the topic, I've read that "continuum radiation is not preferred in AAS." Why not?
Atomic absorption lines are very narrow.
For the Beer-Lambert law to be applicable, the bandwidth of the source should be narrow in comparison with the width of the absorption peak. Otherwise, the signal-to-noise ratio and the slope of the calibration curve would be low; the resulting sensitivity would be poor.
Continuous radiation sources use monochromators; however, even good monochromators have effective bandwidths that are significantly greater than the width of atomic absorption lines.
Therefore, the use of line sources with bandwidths even narrower than the width of absorption peaks is preferred.
One of the strengths of atomic absorption and other ionizing elemental analysis techniques using optical detection (ICP-OES is the main other one I know of) is that the optical absorption and emission properties at particular wavelengths are highly element-specific.
Despite this, the reasons for desiring monochromatic, element-specific AAS light sources are as follows:
- Most detectors used in AAS / OES are broad-spectrum: they report just about any light that strikes them within their respective ranges of sensitivity, which usually span several hundred nanometers.
- While the "fingerprint" wavelengths of each element are highly specific, they often sit at wavelengths close to those of other elements.
- In AAS, any light that is passed through the sample has the potential to be absorbed by anything present in the sample.
Therefore, the use of a continuum light source in AAS would obliterate the analytical specificity afforded by the technique: anything present in the sample that absorbs light in the region of overlap of (a) the emission wavelength range of the light source and (b) the wavelength range of sensitivity of the detector would be recorded in a given measurement.