Use of ion-energy distributions for the identification of species and production mechanisms in low-pressure DC discharges
Mass spectrometry finds widespread application in the analysis of ion distributions in different kinds of plasmas. 1 In DC discharges, the ions reaching the cathode are usually collected by means of a sampling orifice and focused with a set of electrostatic lenses onto the entrance of a mass filter. Mass spectra of ions coming directly from the plasma are not affected by the fragmentation problems at the detector that plague the mass spectrometry of neutrals, 2,3 but the recorded mass distributions are not entirely free from ambiguity. In particular, ions with the same mass/charge (m/q) ratio, but with different chemical composition, will contribute to the same peak. This can be a problem especially for discharges in gas mixtures that can lead to a variety of different products. Additional insight can be derived from the measurement of ion-energy distributions. These energy distributions are largely determined by the acceleration of the ions in the sheath region between the plasma and the cathode, but they may also carry specific information about ion generation mechanisms within the plasma or about collisional processes in the sheath, which can in turn be of help for the identification of different ions with the same m/q ratio.
In recent works, 4,5 we have investigated the ion distributions in low-pressure hollow cathode discharges of H 2 and of its mixtures with CH 4 and N 2 . The experimental setup was already described in the mentioned references and only a brief description will be given here. The plasma reactor was a cylindrical stainless steel vessel (the cathode) with a length of 34 cm and a diameter of 10 cm with a central inner anode. Three plasma precursor mixtures (H 2 /N 2 (5%); H 2 /CH 4 (5%) and H 2 /CH 4 (5%)/N 2 (5%)), with a total pressure of 0.02 mbar, were considered. A DC power supply of ³400 V, 150 mA, connected to the anode through a ballast resistance of ³580 were used in the N 2 /H 2 discharge. In the discharges containing methane, the current dropped by about 20% and the voltage rose in the same proportion, probably due to a change in the characteristic resistance of the plasma. Typical residence times in the reactor were ³0.1 s for CH 4 and N 2 and ³0.4 s for H 2 . A Balzers PPM421 Plasma Process Monitor (PPM), with a cylindrical mirror energy analyzer and a quadrupole mass filter, was used for the detection of ions. Within this arrangement, the ions extracted from the plasma are decelerated and focused with a set of electrostatic lenses onto the entrance of the energy analyzer formed by a semicylindrical capacitor. Only the ions whose entrance velocity is in an adequate relationship with the potential difference between the plates of the capacitor will pass through the filter (see Ref. 6 for more details). These ions are then mass-filtered with a quadrupole and ultimately focused on an electron multiplier. The PPM was installed in a differentially pumped chamber connected to the reactor through a 100 µm diaphragm. During operation, the pressure in the detector chamber was in the 10 7 mbar range. A global account of the ion chemistry in these discharges, which is dominated by protonation reactions, is given in Ref. 5. In all cases, hydrogenic ions H x C (with x D 1, 2, 3), and in particular H 3 C , were found to be dominant. Besides these hydrogenic ions, N 2 H C , NH 4
C and CH 5 C were also recorded in significant amounts.
Ion-energy distributions, f(E), are mostly characterized by a narrow (FWHM <2 eV) maximum for an energy, E m , close to the value of the anode-cathode potential, which indicates that most ions are accelerated from the plasma edge towards the cathode through a largely collisionless sheath. In general 'wings' of variable magnitude appear at the basis of the narrow f(E) maximum and in some of the minor ions, the energy distributions are broader. Illustrative examples are shown in Fig. 1. In the following, we will discuss the likely causes of the different f(E) shapes observed.