Quantitative analysis of conductive coatings by radiofrequency‐powered glow discharge optical emission spectrometry: hydrogen, d.c. bias voltage and density corrections

Abstract

Radiofrequency-powered glow discharge optical emission spectrometry (rf-GDOES INTRODUCTIONGlow discharge optical emission spectrometry (GDOES) has emerged as a major technique for surface and depth profile analysis, owing to the unique combination of fast sputtering rate, high depth resolution, excellent sensitivity and multi-element capability. Although direct current (d.c.) glow discharges sources are widely used for the analysis of metallic coatings, the more recent radiofrequency (r.f.) GD sources offer the advantage that both conducting and nonconducting samples can be analysed on the same source without changing the source. Because of this ability of r.f. sources, research is continuing aimed at using r.f.-GD sources for the routine quantitative analysis of any coating type. Recently, an algorithm for r.f.-GDOES was presented combining a d.c. bias voltage and a hydrogen correction with the emission yields of each element.1 These corrections allowed samples with different electrical characteristics to be combined in a single calibration and correct quantitative depth profiles of hard coatings on steel to be obtained stoiciometrically, even in the presence of significant amounts of hydrogen in the coating. In this paper, we apply this new algorithm to the mystery in GDOES of why there appears to be excess oxygen on the surface of metallic coatings, and then illustrate the success of the algorithm on two types of conductive coatings of major commercial interest: hard coatings and electrical coatings.For some years it has been known that qualitative and quantitative analysis of the extreme surface of metals by GDOES leads to an apparent excess of oxygen, 2 some surfaces having as much as an unrealistic 80 at.% oxygen. Initially it was thought that this might be due to a preferential sputtering artefact, but attempts to correct for the preferential sputtering of oxygen were not successful. When the effects of small quantities of hydrogen on emission yields became known, it was immediately suspected that the oxygen surplus may be due to the effect of relatively high hydrogen levels on the surfaces of metals exposed to air (A. Bengtson and R. Payling, personal communication, 2000). With the development of the hydrogen correction algorithm it is now possible to test this hypothesis.Normal 'moist' air contains significant amounts of water vapour. The air in a cool laboratory (20°C, 1 bar, 50% relative humidity) without air-conditioning might typically contain ¾0.2% water vapour. When a fresh metal surface is exposed to 'moist' air, it oxidizes by adsorbing oxygen both from O 2 and from H 2 O, forming oxides and hydroxides at the immediate surface. The formation of oxide and hydrogen from the reaction of the metal with water vapour is favoured energetically. Oxidation continues by migration of oxygen atoms into the metal selvedge (sub-surface). prepared by CVD are obtained by heating metal samples in hydrogen-rich atmospheres, and the coatings grow by chemical reaction between the metal and the gases in the oven. Su...

New aspects of quantification in r.f. GDOES

“…For hydrogen, the expressions in Eqns (4)- (8) can be multiplied by a hydrogen correction term. 9 A property noticed some years ago, 13,14 but not exploited to date, is that at constant pressure the emission yield appears not to vary significantly with varying power. As the power is increased, both current and voltage increase, and the increase in emission yield due to increasing current is offset by a decrease in emission yield due to the increasing voltage.…”

Characterization of Ti-C-N coatings deposited on Ti6Al4V for biomedical applications

“…The measurement conditions used were 650 Pa and 35 W, with a copper anode of 4 mm of diameter [20]. For this characterization, coatings deposited on Ti6Al4V samples were employed.…”

Glow discharge techniques in the chemical analysis of photovoltaic materials

“…A high depth resolution (up to 20 nm) is a feature of GD‐OES as well, a method which simultaneously detects up to 52 elements (Horiba Jobin Yvon/JY 5000 rf) at a detection limit of about 1–10 ppm . It was applied to perform high resolution depth profiling analysis on second and third generation PV devices.…”