Non-specific absorption, or background absorption, occurs when radiation from the hollow cathode lamp is attenuated by molecular species or solid particles in the light path, that either absorb or scatter the energy from the hollow cathode lamp. Molecular absorption can occur when the atomizer is not hot enough to decompose all matrix components in the sample. The remaining molecules will then absorb light from the hollow cathode lamp. This molecular absorption and scatter is added to the atomic absorption giving a falsely high signal. In flame analysis, the background absorbance is generally less than 0.05 absorbance. Nonetheless, there are some practical analytical situations, particularly at low UV wavelengths, where background absorption can occur. However, in graphite furnace analysis, the background signals can exceed
2.0 absorbance and accurate correction is obviously important. In practice, analyte signals which are small should not be measured in the presence of large background signals. These high absorbances reduce the amount of light energy reaching the detector, reducing the signal-to-noise ratio and degrading the precision and accuracy of measurement. A number of background correction techniques have been developed to allow subtraction of the background signal from the total absorbance measurement. The new generation SpectrAA instruments will allow measurement of analyte signals with background signals of up to 2.5 absorbance. In practice operating in the presence of these high levels of background can not be recommended. The reduction of the background and the generation of the highest atomic signal through the use of chemical modifiers, appropriate gas flows and suitable temperature programming of the graphite
furnace during the drying and ashing stages are primary goals. Background correction is a way of identifying the amount of non-specific absorption from the atomic absorption. All commercially available background correction systems employ the same basic principles: the total absorption (the sum of the atomic and non-specific absorption) and the non-specific absorption alone are
measured at two separate time intervals, separated by a few milliseconds. The atomic absorption is obtained by subtracting the non-specific absorption from the total absorption. Signals in graphite furnace analysis are produced rapidly and decay rapidly (the maximum signal rise times can be up to 10 absorbance units per second). Ideally, the total and background signals should be measured simultaneously for accurate correction. This is not practical and therefore, the 2 signals should be measured as close as possible in time. The time difference in commercial instruments ranges from 2 ms to 10 ms–the larger the time difference, the greater the
error in the background correction (Refer to the Technical Sales Arguments section in the Sales Manual for further detail on the extent of these errors in background correction).