Spectroscopy is a technique that is based on a sample's absorption of a certain wavelength of light. Increasing the concentration of a substance increases the absorption of the light. This directly proportional relationship, known as Beer's Law, allows for concentrations of unknown samples to be determined by comparing the absorption to known samples. Colorimetry is a similar technique but only uses wavelengths of visible light. For substances that have color, a wavelength of the complementary color is used.
The key to using spectroscopy is the generation of a calibration curve. A calibration curve is a graph plotting absorbance versus concentration. Standard solutions of the substance to be analyzed are generated, often through a serial dilution, and their absorbances at a pre-selected wavelength are determined. A linear line of best fit is added to the plot.
To generate and use a calibration curve:
Create the standard solutions of the substance using a serial dilution.
Select the appropriate wavelength that the substance will absorb.
Generate the calibration curve with a linear line of best fit using the prepared solutions.
Measure the absorbance of the unknown sample.
Use the slope of the linear line of best fit to calculate the concentration of the unknown.
Above: An example calibration curve for spectroscopy. From: https://www.news-medical.net/image-handler/picture/2016/2/Epp3.jpg
The linear fit provides y = mx. Substituting absorbance (y) and concentration (x) into the formula provides:
absorbance = gradient (concentration)
Rearranging the equation provides:
gradient = absorbance / concentration
On the right is a calibration curve that was generated for Compound A. If the absorbance of an unknown sample of Compound A was 0.64, the concentration of the unknown is:
absorbance = 0.0443 concentration
0.0443 = absorbance / concentration
0.0443 = 0.64 / concentration
concentration = 14.4 µg / mL
Above: A sample calibration curve. From: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcR7cksOMj6LJ5H-yDbtJI09Bko8RV9luOIp9g&s