The influence of operational modes on sputtering rates and emission processes for different sample matrices in rf-GD-OES

A study was carried out to select the most convenient pair of the three key interdependent plasma operating parameters in a radiofrequency glow discha rge (rf-GD) (generator output power, dc-bias developed in the sample surfac e and pressure) to be fixed in order to measure intensities by optical emis sion spectrometry (OES) or sputtering rates with adequate precision. The be st operating conditions in order to obtain values of emission yields unaffe cted by the nature of the sample matrix are also given. Three different mod es of rf-GD-OES operation were studied and emission intensities of selected elements (Fe, Ni, Zn, Cu) in eight different matrices were measured. Resul ts showed that a mode in which generator output power and dc-bias voltage a re fixed while pressure is allowed to vary provided the worst relative stan dard deviations (RSDs) for the emission intensities measured (higher than 1 5%). The other two approaches studied (either generator output power and pr essure fixed with the dc-bias varied or pressure and dc-bias voltage fixed with the generator output power varied) provided better performance (lower than 4%) in both cases. Concerning the sputtering rates, the two modes, wor king either with variable dc-bias or with variable power, gave good precisi on for three replicates. Finally, these two modes seemed also to provide re liable values for the emission yields of each analyte emission line under s tudy in the different reference samples investigated. Comparison of results observed by keeping constant the pressure in a rf-GD and those achieved by dc-GD in its usual mode of operation (variable pressure, maintaining fixed the voltage and current) shows similar behaviour for both sources when com paring the trends followed by the emission yields when modifying the matrix .