Estimation of errors in absorbance ratio measurements

Abstract Ratio measurements are commonly used to address a variety of analytical problems in environmental, forensic and pharmaceutical laboratories. In absorbance ratioing techniques, analytical chemists rely on the spectral features of the analyte(s) of interest. The absorbances at two wavelengths are monitored and the ratio of these two absorbances is computed. This ratio is then used to confirm the identity of the analyte(s) of interest, the purity of a product of the overlap of chromatographic peaks. These decisions often have far‐reaching consequences (e.g. the identification of the source, biogenic or petrogenic, of hydrocarbons in biological tissues or water). Given the cost and the liabilities associated with such decisions, it is unfortunate that these ratios are seldom reported with any statistical confidence. The purpose of this study is to delineate the parameters that affect absorbance ratio measurements. The models that can be used to estimate the statistical confidence in these measurements are derived and evaluated experimentally. The results show that these models can estimate the relative standard deviations in absorbance ratios accurately. They can also estimate the effect of signal‐to‐noise ratio and the choice of wavelengths on the precision of absorbance ratios.