Arne Bengtson

Glow discharge optical emission spectrometry: moving towards reliable thin film analysis–a short review

Glow discharge optical emission spectroscopy (GD-OES) is briefly reviewed, with particular reference to topics relevant to the application field of near surface and thin film analysis. The special needs and requirements for thin film analysis, in contrast to coating and bulk analysis, are pointed out. A task list is developed which shows the requirements of further developments to the technique and the fundamentals. The state-of-the-art is presented in measurement technique, GD source control and design, the effect of traces of molecular gases, correction and quantification procedures, contributions of modelling and, finally, reference materials for thin film analysis.

Surface analysis with the glow discharge lamp. State of the art and prospects for future development

A brief history of the development of glow discharge optical emission spectrometry (GD-OES) for in-depth profiling is given and the basic operating principles are described. Some of the most important technical applications of GD-OES are discussed. The state-of-the-art concerning quantitative surface analysis with GD-OES is presented and two different approaches to this problem are described. The prospects for further development of the methods for quantitative surface analysis are also discussed.

Further improvements in calibration techniques for depth profiling with glow discharge optical emission spectrometry

Quantitative in-depth profiling with glow discharge optical emission spectrometry is discussed. Recent experimental work on the variations in emission intensity with the discharge parameters is presented. An improved empirical intensity expression, which has been incorporated in a quantification model, is proposed. In particular, the proposed model compensates for variations in the emission yield (emission intensity per sputtered atom). Some examples of quantitative in-depth profiles, obtained with the proposed calculation model, are presented.

Microsecond pulsed glow discharge optical emission spectrometry — investigation of temporal emission characteristics

A high voltage, microsecond regime, pulsed glow discharge has been applied to a standard Grimm type source on a LECO SA-2000 multichannel optical spectrometer. The spectrometer system has been equipped with fast time-resolved electronics in order to study the temporal response of the elemental emission lines. The aim of the research was to investigate possible gains in signal-to-background ratio (S/B) and detection limits compared to a conventional direct current (dc) mode of operation. It was found that the high instantaneous pulse power in itself gives approximately a factor 2 increase in S/B. By utilising time-resolved detection, gains in the range 5–10 are achieved. For nitrogen, it was found that the background from leakage of air could be reduced to approximately 1/3 compared with the dc mode. Another interesting observation was that certain nearly non-conducting materials, which cannot be run in dc mode, could be run in the microsecond pulsed mode.

Development of a standard method for quantitative depth profile analysis of zinc-based metallic coatings by direct current glow discharge optical emission spectroscopy

A method for the determination of the thickness and composition of zinc-based coatings on steel has been developed. The method is based on direct current glow discharge optical emission spectroscopy (GD-OES). Six different types of coated materials supplied by five major European steel producers have been used to develop and evaluate the method. The types of coatings are hot-dipped zinc, electroplated zinc, electroplated ZnNi, Galvanneal (electroplated and annealed ZnFe), Galfan (hot-dipped Zn95Al5 alloy) and Aluzink (hot-dipped Zn43Al55Si2 alloy). In the research phase, the range of suitable analytical conditions for these materials was established. Based on these results, a preliminary standard method was established. The method is based on calibration of the emitted light per unit weight of each element. It is recommended that Certified Reference Materials (CRM/s) are used for calibration, and a range of alloy types can be used (multi-matrix calibration). An interlaboratory comparison based on the preliminary standard method was carried out, involving 12 laboratories from eight countries, using nine different types of glow discharge spectrometers. The results show that the method works very well for accurate and rapid determination of the major element (Zn) and the coating thickness. The concentrations of alloying elements also could be determined very accurately, but the interlaboratory variations were generally too high. Based on these results, several amendments to the standard method have been suggested. Copyright

Emission yield for quantitative depth profile analysis by glow discharge optical emission—the influence of discharge parameters

The emission yield in glow discharge optical emission spectroscopy, defined here as the integrated emission signal at some defined wavelength (spectral line) per unit sputtered mass of the elementi, was investigated for the Si 288.2 nm spectral line. The purpose of the investigation was to evaluate the influence of the discharge parameters current, voltage and pressure. Three samples were used, a steel, a brass and an aluminium. The major part of the investigation was carried out with a direct current glow discharge, but comparative measurements on the same samples were also carried out with a radiofrequency glow discharge. The measured emission yields were fitted to various functions of the discharge parameters by non-linear regression software. The results show that the parameters current and voltage have a considerably larger influence on the emission yield than the pressure. The emission yield was found to increase with the current and decrease with voltage and pressure. For the investigated Si 288.2 nm line, it is concluded that the influence of pressure is nearly negligible within a normal operating range of discharge parameters.

Developments in glow discharge optical emission spectrometry. Invited lecture

The historical background of glow discharge optical emission spectroscopy (GD–OES) is briefly discussed. The Grimm-type source is described, including some of the unique analytical characteristics. State-of-the-art analytical figures of merit for bulk analysis are RSD 0.1–1% for major and minor analytes and detection limits in the range 1–10 ppm. For surface and depth profile analysis, GD-OES is a technique with the capability of profiling 100 µm thick surface layers, yet the minimum information depth is just 1 nm. Recent developments in methods for the quantification of depth profiles are presented. The current trends in the development of new glow discharge sources is discussed, with emphasis on rf technology for non-conducting materials. Some speculations regarding future trends in the technical development of GD-OES are given. It is argued that pulsed operation in combination with time-resolved detection should be utilized in order to improve further the analytical figures of merit.

The hydrogen effect as a function of discharge parameters - investigations of the variation of emission yield in glow discharge plasmas

It is well known that traces of hydrogen in the glow discharge plasma enhance or suppress the intensities of other spectral lines, in this work termed “the hydrogen effect”. In this work, the focus is on the variation of the hydrogen effect as a function of the discharge parameters, for the purpose of developing effective correction methods in quantitative depth profile analysis. A simple, but effective experimental method was employed: The intrinsic contamination of the source leads to an increased amount of hydrogen in the plasma at the beginning of the discharge, which is rapidly reduced with sputtering time. Thus, by recording the intensity depth profiles of multi-element bulk reference materials, the influence of hydrogen on several emission lines can be observed. The decay or increase in intensity, correlated to the intensity of the hydrogen emission line, gives the desired information. Systematic investigations by varying current and voltage in a considerable range provided a good matrix of data to find a model describing these variations. Furthermore, renewed investigations of the variation of emission yields (EY) with discharge parameters have led to a model which has been used to correct the obtained data for the hydrogen intensities, so as to represent the actual content of hydrogen in the plasma. The information extracted from these studies is expected to be of great importance for routine depth profile analysis, as the hydrogen correction is a relevant tool in software quantification algorithms. Existing correction models for the hydrogen effect do not take variations in the discharge parameters into account. This work clearly shows that it is necessary to implement such variations for accurate correction.