Charakterisierung und Kompensation geometrischer Einflüsse auf die Laser-Emissionsspektroskopie bei der Inline-Werkstoffanalyse

摘要

In the field of metal recycling it is of interest to sort the material by the alloy elements before it is melted. The usage of laser-induced breakdown spectroscopy (LIBS) for the sorting process of single pieces on a conveyer belt has some advantages in contrast to conventional methods regarding sample preparation, selectivity and throughput. If a LIBS based system is used for sorting, the results from the geometry between the incident laser beam and the surface of the peaces show dependence between geometry and the spectroscopic signal. In this thesis, the geometric dependence is experimentally investigated. Beside the investigation of the plasma parameters the main focus is on the selection criteria of suitable line pairs as well as establishing an optimized calibration method. The measurements show that with increasing angle and increasing defocusing in addition to a general drop of the intensity a drop of the plasma temperature and the electron density can be determined. Some of the performed calibrations show a linear dependence to the angle. Is the angle taken into account during the calibration the dependence to the angle can be reduced by the factor of 10 but the standard deviation of the process increases by the factor of 2 at the same time. Other performed calibrations have shown that some pairs of lines have an angle dependence which is 10 times lower than the standard deviation of the process so that the angle dependence can be neglected for these pairs of lines. Some line parameters were examined for these pairs of lines to find a reference to the angle independence. The only parameter that was identified is the wavelength. Pairs of lines with small angle dependence are within a small wavelength interval. But a small wavelength distance between the analyte- and reference- line does not necessarily result in angle independence.By the use of optimal pairs of lines the angle dependence can be reduced and so the measurement accuracy can be increased. This allows a recycling process without down cycling and a better use of the resources.