Principle

A copper tube is brought very close to the solid sample to be analysed. The copper tube is filled with a low pressure of argon (~600 Pa) and a voltage (potential difference) is applied between the anode and the sample. The applied voltage can be radio frequency (RF) or direct current (DC). The sample is negatively biassed relative to the anode.

Electrons leave the more negative surface of the sample towards the anode but collide with argon atoms creating positively charged argon ions and high energy metastable argon atoms. The positively charged argon ions are attracted to the negatively biassed sample surface. Along the way they have many low energy collisions with other argon atoms, losing about half their original energy. Many of the ions are neutralised by collisions but continue towards the sample.

When the argon ions (plus neutrals) strike the sample surface they impart sufficient energy (>100 eV) to disrupt atomic bonds and eject atoms and electrons. This process is called sputtering. The sputtered atoms fly away from the sample surface and coat themselves on the anode or are removed by vacuum pumps.

Away from the sample, some of the sputtered atoms have collisions with high energy electrons or metastable argon atoms and are excited to high energy states. When they then de-excites they emit a photons, which create a 'glow'. This glow is then analysed with one or more optical spectrometers.

It sounds complicated and it is. The plasma created in the source is very small and therefore difficult to measure directly and to model mathematically. The plasma has the same diameter as the inside diameter of the anode, typically 4 mm. No plasma is created between the front face of the anode and the sample as the distance there, typically 0.1-0.2 mm, is too small. It is therefore called a restricted plasma.

Immediately in front of the sample is a dark space called the cathode fall. It extends for about 0.5 mm. Here there is no glow but most of the drop in plasma potential occurs here. So most of the ions are created near the beginning of the cathode fall. After the cathode fall is the negative glow region, where most emission occurs. The negative glow region extends for about 2-3 mm away from the sample. Normally there is no positive glow region in GD-OES sources.