Absorption spectroscopy is an analytical tool used by chemists. It is based on the absorption of quanta of light by a substance, due to the promotion of electrons from one orbital to another in that substance.

The absorption spectrum is characteristic for a particular compound, and does not change with varying concentration. It is usually measured on a sample of the substance in solution, although reflectance spectroscopy can also be used.

Absorption is measured as A = log(Io/It) where Io is the incident light, It the transmitted light. This quantity has units of OD (optical density) and is wavelength-dependent.

The measured absorbance has been shown to be proportional to the molar concentration of the absorbing species and the thickness of the sample the light passes through. This is known as Beer's Law, and is expressed: A = (molar absorption coefficent) * (concentration) * thickness.

The plot of Absorption versus wavelength for a particular compound is referred to as the absorption spectrum.

The absorption spectrum is measured using a spectrophotometer, where a light source coupled with a diffraction grating allows Absorption to be measured at various wavelengths.

Optical absorption spectra are measured from the near ultraviolet to near infrared.

Visible light absorption spectra can be taken in anything that is visibly clear. Polystyrene, quartz, and borosilicate (pyrex) cells are the most commonly used. UV light is absorbed by most glasses and plastics, so quartz cells are used. The Si-O moieties in glasses and quartz, and the C-C moieties in plastics absorb infrared light. Therefore, infrared absorption spectra are typically carried out with a thin film of the sample held in place between sodium chloride sample plates. Other methods involve suspending the compound in a substance does not absorb in the region of study. Mineral oil (Nujol) emulsions and potassium bromide glasses are perhaps the most common. NaCl and KBr do not have significant IR absorptions, being ionic, and Nujol has a relatively uncomplicated IR spectrum.

Spectroscopy as an analytical tool

Suppose you had two pieces of pipe. One weighs 105 grams and the other weighs 350 grams. Furthermore, you have a spring, the stretchiness of which corrsponds to these weights. Finally you have a scale or a method of marking and recording the stretch.

Put the lighter pipe on the spring, make a mark and label it "105". Put the heavier pipe on the spring and label it "350".

Now suppose you want to weigh something. Put the object on the spring and observe the stretch. Make a mark. From the ratio of the unknown to the known the weight of the object could be determined.

Please make several observations. First, the pieces of pipe are rather arbitrary. We know their weight with some precision, but the actual weight is arbirtary. Second, we know little about springs. Third, our "scale" will only work within a fairly narrow range of weights. And finally, it is obvious that our "scale" is not perfect.

Quantative analysis works in much the same way. The analytical chemist uses instruments (the spring) to extend her senses. The analysis of known standards can be compared to unknowns to give quantitative results.

A very common instrumental technique for quantitative analysis is spectroscopy. Spectropscopy works great because there is a linear relationship between absorbance and concentration. This is described in Beer's law.

The math is linear and, since zero concentration corresponds to zero absorbance, the line goes through the origin.

Example: a cyanide standard at 200 parts per million gives an absorbance with an arbitrary value of 1540. An unknown sample gives a value of 834. The math could be stated as: "if 200 gives you 1540, what gives you 834?" Since this is a linear relation and goes through the origin, the unknown is easily calculated to be 108 parts per million.

Analytical chemistry is based on these sort of simple calculations. Note that the chemist is not concerned about coeffecients or theory or chomophores or cell length or anything else. If everything is constant then a rigorous approach to the math would reveal that everything drops out of the equation except the concentration and the absorbance. And the absorbance is typically a an arbitrary number which is generated by the detector and computer.

The analytical chemist makes lots of assumputions and has to accept the inevitable errors and uncertainty.

Our example with the pipes and spring is somewhat poor because the stretch of a spring does not have a linear relationship with the weight. This example is more apropos to spectroscopy because there is a linear relationship between concentration and absorbance.