Accessories for Multiple Internal Reflection

As any compound in contact with the IRE produces a spectrum, clearly the only reasonable choice for a background or reference spectrum is the bare IRE. Under no circumstances should the clamp be closed on the IRE without the sample and used as a reference. Even with IREs that are used for liquid analysis, the single-beam spectrum of the clean IRE should be used for the background. For example, if the solution is aqueous and it is desirable to remove the water absorption, two separate background spectra should be recorded. One should be the absorbance spectrum of the aqueous solution referenced to an empty ATR cell, and the other should be the absorbance spectrum of water also ratioed to the spectrum of the empty ATR cell. A scaled subtraction of the water spectrum from the solution spectrum will generally produce better results than a ratio of the spectrum of the aqueous solution to that of pure water. This is because the solute displaces water from the solution, so that the amount of water that contributes to the solution and water spectra will not be identical, so they may not ratio correctly. Scaled subtraction, however, can correct for the difference provided that the solute does not cause a shift of the water band. 

In practice, the value of dp calculated from Eq. 15.4 is only an approximation of the effective penetration depth, de, which is the thickness of a sample measured by transmission spectrometry that is required to match the corresponding absorbance in a single-reflection ATR spectrum [2]. The value of de depends on the polarization 

The values of dg p and dg for a sample with a refractive index of 1.5 are given for ZnSe and Ge in Table 15.3. As a first approximation for ATR spectra measured on a FT-IR spectrometer, the effective depth of penetration is simply the average of dgp and dg g. It should be recognized, however, that the beamsplitter efficiency is different for p- and i-polarized radiation, so again, this average is just an approximation of the true value of dg. It should also be noted that Eqs. 15 and 16 are only accurate for weak absorption bands because of the effect of anomalous dispersion. 

To measure the ATR spectrum of a solid sample, it is simply clamped to one or both sides of the IRE. Care must be taken to assure that the cushioning pads on the clamps do not come in contact with the IRE otherwise, a spectrum of the pads will be added to that of the sample. Care must also be taken such that the sample does not occlude the end of the IRE where the infrared radiation enters or leaves the IRE. 

Horizontal ATR accessories can be constructed so that the IRE is mounted flush with the top of the accessory, which allows solid samples to be pressed against the IRE with a suitable weight. With the IRE flush to the surface, the sample can be considerably larger than the IRE itself. To facilitate the placement of very large samples, the IRE can be mounted parallel to the top of the instrument. Such a geometry, however, is not well suited to liquid or powdered samples. 

---