Compounds designated hydrides in QASoft are binary combinations of hydrogen and other atoms, excluding carbon. These compounds all have great detail in their spectra, exhibiting many individual lines of 0.2 cm-1 width. Identification presents no problem, but quantification can be a problem. Spectrometers that do not fully resolve the spectral structure will cause apparent deviations from the absorption law. One must then take care to base quantification only on weak spectral features.
Uncertainties of absorption coefficients are greater for these hydrides than for any other group of compounds in this collection. These compounds tend to interact readily with the vessel walls. Acidic vapors such as the halogen acids are taken up on metal surfaces. The alkaline vapors, such as ammonia and hydrazine, are taken up on glass. When the sample is disappearing during the measurement it is not easy to assign the correct value of ppm-meters to the spectrum. A way to make the measurements more reliable would be to set up a flow system and establish an equilibrium between adsorption and desorption at the walls. Then a careful measurement would need to be made of the flow and of the rate of introduction of material into the flow. An analyst who needs to be quite certain of his calibration should use such a flow system. If his calibration differs from ours, he can then make a correction to the ppm-meters number of the digitized reference spectrum.
In the present work, the samples of HCl, HBr, and water appeared to be stable in the absorption cell and the initial manometer reading was used to calculate the ppm-meters number. The samples of ammonia and hydrazine were found to be decreasing in concentration during the measurement. For these compounds the ppm-meters numbers were obtained by observing the decrease in absorbance, extrapolating back to the time when the manometer reading was made and then calculating the partial pressure that existed during the actual recording of the spectrum.
HF presents special problems because it reacts with glass as well as with the reactive metals. It also fumes in air and therefore must be handled completely dry. The HF spectrum was actually measured in a two centimeter long copper cell. A full atmosphere of HF was first used to produce a very strong but stable spectrum. Then the HF was diluted with oxygen and a weaker spectrum with the desired modest degree of absorbance was recorded. The partial pressure for the weak spectrum was then calculated by comparison with the strong spectrum, using weak lines on the edge of the band.
The water spectrum presented here is a low-noise composite spectrum created by splicing together segments from several water spectra recorded with differing concentration-pathlength products. The longest pathlength used in the water measurement was 288 meters. The strongly absorbing regions were taken from a spectrum at a short path and the weakly absorbing regions were taken from long path spectra. The resulting composite spectrum can be used over wide ranges of concentration. All the plotted water bands presented here are taken from one composite digitized water spectrum. See SPECTRA, chapter E. H2S is listed under SULFUR COMPOUNDS.