The influence of controlled addition of either hydrogen, nitrogen or oxygen (in the interval from 0.5 to 10% v/v) to an argon radiofrequency glow discharge (rf-GD) with optical emission spectrometric detection has been investigated in terms of analyte emission intensities, sputtering rates and analyte emission yields. Considering that plasma characteristics may vary greatly between conductive samples and insulators, both sample types (austenitic stainless steels and a homogeneous glass, respectively) have been employed in this study. Analytes investigated were silicon, iron, nickel and aluminium in the steel, while silicon, sodium, calcium and magnesium were selected in the glass.

It was observed that the addition of the assayed molecular gases gave rise to a decrease of the sputtering rates, when comparing with pure argon, at any percentage of the molecular gas investigated. For hydrogen–argon or nitrogen–argon mixtures it was found that the percentage of sputtering rate reduction, with respect to pure argon, was highly similar when comparing the results from both types of samples (e.g., the sputtering rate decrease for 1% H₂, as compared with pure argon, was 33% for stainless steel versus 28% for glass). However, when adding oxygen to the argon rf-GD, the sputtering rate decrease, as compared with argon, was much stronger in the conductive matrix than in the glass.

Concerning the emission yields, selective enhancements were obtained with the addition of hydrogen (e.g., at the Si I 288.158 nm and the Mg I 383.829 nm lines) or nitrogen (e.g., at the Al I 396.152 nm and the Mg I 383.829 nm line). However, a rather systematic increase of emission yields was found in the presence of 0.5–2% of oxygen in an argon matrix as compared to pure argon.