Cooled hollow cathode

Normally, a different lamp is used for each element. Multi-element lamps (e.g. Ca-Mg, Fe-Mn or Fe-Ni-Cr) are available, but are less satisfactory owing to the differing volatilities of the metals. Demountable (water-cooled) hollow-cathode lamps have also been marketed, but are not widely used. 

Figure 2. Neon spectra (hollow cathode emission) acquired with an SIT detector cooled to —50°C. (a) Real-time detection, 20 ms/scan, neutral density filter (ND) = 0 (b) Readout, after signal integration for 20 seconds (c) Integration equivalent to 10s scan periods. ND filter = 5 (0.001% transmission) was used to attenuate the signal. Equivalent dark spectra were subtracted from each neon...

Figure 1 Schematic representation of the cloud of cool atoms which escapes from the centre of the hollow cathode at high lamp operating currents...

While conventional or high-intensity (boosted output)4 hollow cathode lamps are usually simply operated at room temperature, electrodeless discharge lamps are sometimes cooled with a regulated flow of air maintained at a constant temperature,5 and this flow too must be optimized with respect to signal-to-noise ratio. Sometimes these sources are operated in a vacuum jacket to enhance sensitivity and/or to improve stability.6... 

Cases have been observed where the isotopic line absorption profiles completely overlap, e.g. boron-10 and -11 in a krypton-filled lamp at 249.7 nm [244]. Hannaford and Lowe [245] later showed that this was caused by an unusually large Doppler half-width induced by the fill-gas, and, if neon is used, the 208.9 and 209.0 nm lines can allow the determination of boron-10 and boron-11 isotope ratios. The 208.89/208.96 nm doublet was found to be more useful than the 249.68/249.77 nm doublet. Enriched isotope hollow-cathode lamps were used as sources. A sputtering cell was preferred to a nitrous oxide/acetylene flame as the atom reservoir, as it could be water-cooled to reduce broadening and solid samples could be used, thus avoiding the slow dissolution in nitric acid of samples of boron-10 used as a neutron absorber in reactor technology. 

The characteristic depends on the discharge gas (at a few mbar of argon i = 0.2-2 A, V = 1-2 kV at 10-20 mbar of helium i = 0.2-2 A, V < 1 kV) but also on the cathode mounting. Indeed, in a cooled cathode the characteristic is normal and the analyte volatilizes by cathodic sputtering only, whereas in the hot hollow cathode thermal evaporation also takes place, by which the characteristic especially at high currents may become normal. In this case thermal effects could even lead to a strong selective volatilization, which can be made use of analytically. The latter was shown to occur in the case of brass, as it could be demonstrated by electron probe micrographs of partly molten brass samples, the outer layers of which are less rich in Zn [467]. 

Transferred arcs with water-cooled non-consumable cathodes are illustrated in Fig. 4-44. The generation of electrons on the inner walls of the hollow cathodes is provided by field emission, which permits operation of the transferred arcs at the multi-megawatt power for thousands of hours. The electric circuit is completed by transferring the arc to an external anode, which is a conducting material, where the arc is to be applied. The arc root can move over the cathode surface, which further increases its lifetime. 

A non-linear wall-stabilized non-transferred arc is shown in Fig. 4 8. It consists of a cylindrical hollow cathode and coaxial hollow anode located in a water-cooled chamber and separated by an insulator. Gas flow blows the arc column out of the anode opening to heat a downstream material, which is supposed to be treated. In contrast to transferred arcs, the treated material is not supposed to operate as an anode. Magnetic 7x5 forces cause the arc roots to rotate around electrodes (Fig. 4-48), which provides longer electrode lifetime. The generation of electrons on the cathode is provided in this case by field emission. An axisymmetric version of the non-transferred arc, usually referred to as the plasma torch or the arc jet, is illustrated in Fig. 4-49. The arc is generated in a conical gap in the anode and pushed out of this opening by gas flow. The heated gas flow forms a very-high-temperature arc jet, sometimes at supersonic velocities. 

Difficulties occur with multiple-element hollow cathodes and only certain combinations of elements have been successful. The principal source of difficulty is due to different volatilities of elements. The more easily volatilized element is preferentially vaporized and, on cooling, deposits over the entire hollow cathode surface. The intensities of emission of other metals in the hollow cathode thus are seriously reduced. 

The energetically important parts of the discharge (cathode layer, dark space, and negative glow) as well as the sample are inside the cathode cavity. The volatilization results from cathodic sputtering and/or thermal evaporation. This depends on whether the whole cathode with its outer and inner wall is subjected to sputtering without any cooling (hot hollow cathode), the outer wall is shielded by a quartz... 

---