Quantitative analysis of halogens in polymer coatings using laser induced breakdown spectroscopy

Abstract

In today’s semiconductor industry, halogenated polymer coatings are utilized as corrosion protection in microelectronic devices. In order to monitor the degradation of these anti-corrosion coatings, a spatially resolved analysis of their halogen content is required. Because methods such as Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) or X-Ray Fluorescence Spectroscopy (XRF), which would normally be used for this task, struggle with the detection of halogens, there is need for the development of a fast and quantitative imaging technique, that is able to perform measurements in the μg/g range. The method investigated in this work was Laser Induced Breakdown Spectroscopy (LIBS). The goal of this thesis was to develop a LIBS measurement methodology that enables the quantitative analysis of F, Cl, and Br in polymer coatings. Given the low excitation efficiency of halogens, it was necessary to enhance the signals, for which there are numerous potential approaches. The two options employed in this work, however, were the application of reduced pressure, which changed the plasma dynamics resulting in an enhancement of emission line intensity, and sampling the plasma radiation directly next to the plasma plume with a fiber optic cable, referred to as ‘direct sampling’, which increased the collection efficiency. These could be investigated through the successful development of the ‘direct stage’, a custom-built ablation chamber mounted on a 3D printed stage built in-house. To be able to perform a quantitative analysis, standards were prepared by dissolving Polyimide and halogen-containing benzoic acid derivates in N-Methyl-2-pyrrolidone. After optimizing all measurement parameters, calibrations were performed, and the raw data was normalized using different normalization methods. A pressure of 50 mbar significantly improved the signal-to-background ratio (SBR) of the F I 685,6 nm emission line by a factor of ~2, resulting in a background normalized F calibration with a R2 of 0,9988 and a LOD of 48 μg/g. The quantitative analysis of F can therefore be considered a success. Although the Cl measurements at 50 mbar exhibited higher sensitivity, the calibration under atmospheric pressure achieved better figures of merit. The background normalized Cl calibration at 1013 mbar obtained a R2 of 0,9925 with a LOD of 1497 μg/g. Unfortunately, the use of vacuum did not have a beneficial effect on the Br I 827,3 nm emission line intensity. Therefore, the Br calibration was carried out under atmospheric pressure leading to a R2 of 0,8615 along with a LOD of 1703 μg/g, which was the most deficient of all investigated elements. Nevertheless, this thesis shows that LIBS can be operated in a quantitative manner in the μg/g range, and that the developed vacuum method is also applicable to fluoropolymer samples from industry, as demonstrated by the analysis of a sports bag material.