Unsicherheitsbilanzen in der quantitativen FT-IR-Spektroskopie
The uncertainty of analytical results allows to assess the reliability of the applied analytical methods. According to quality requirements for accredited laboratories (see e.g. ISO/IEC 17025) it is necessary to report the uncertainty of measurement. In order to evaluate the uncertainty of typical quantitative FT-IR spectroscopic analysis of solids and liquids a detailed study of the influence of numerous FT-, sample- and instrument-specific parameters on the measurand was performed. The trueness of FT-IR spectroscopic measurements on glasses was tested with a transmittance standard of the National Physical Laboratory (NPL) that consists of a Schott NG11 neutral density filter. The trueness of FT-IR spectroscopic measurements on liquids was checked by the comparison of the molar absorptivities of dichloromethane determined by transmission measurements with the standard values published by IUPAC. After optimisation of the measurement conditions the transmittance values of the NPL standard and the molar absorptivities of dichloromethane are in accordance with the corresponding standard values. The uncertainty budgets evaluated for these measurends makes it possible to assess the contribution of a single parameter on the combined uncertainty. By using the KBr pressed pellet technique for the quantitative analysis of insoluble solids the particle size of the analyte and the homogeneity of the pellets are the major contributions to the quality of spectra. In order to minimize the uncertainty due to the particle size a purpose of the sample preparation must be to reduce the particle sizes of the calibration sample and the analyte to a maximum size of 2 µm and both particle sizes should agree. A detailed study with calcium carbonate samples with different mean particle sizes has shown that the half-width of the analysed band depends on the particle size and is suitable to test how far the last mentioned condition is fulfilled. In order to minimize the uncertainty due to an inhomogeneous distribution of the absorbing particles in the pellets the infrared beam must pass through the whole surface area of a pellet or the mean value of the spectra measured at different positions on the surface area has to be evaluated. This work proves that the use of the pressed pellet technique can lead to correct results within the limits of the uncertainty if the sample preparation is carried out carefully and optimal measurement conditions are kept.
Document Full Text
School:Humboldt-Universität zu Berlin
Source Type:Master's Thesis
Keywords:Messunsicherheit Transmissionsstandards Schott NG11 Neutralglasfilter FT-IR spectroscopy uncertainty transmittance standards molar absorptivity neutral density filter dichloromethane calcium carbonate
Date of Publication:05/22/2002