Atomic hydrogen III—The energy efficiency of atom production in a glow discharge

Abstract

There seems to have been a tendency amongst workers on the use of the glow discharge as a source of atomic hydrogen to regard the current or power as determining the degree of dissociation of the gas, i. e. the equi­librium H 2 ⇌2H. It is clear, however, that the discharge itself determines only the rate of production of atoms, whereas the degree of dissociation depends also on the rate of removal of atoms by pumping and by recombina­tion processes which are independent of the discharge. The two homogeneous recombination processes are those resulting from three-body collisions be­tween three atoms and between two atoms and a molecule; in addition, there is a heterogeneously catalysed reaction in which the walls of the tube act as the energy acceptor. Attempts to connect electrical conditions with degree of dissociation have been made by Crew and Hulburt (1927) and by Wrede (1929), but the above remarks show that only empirical relationships can be hoped for. In Crew and Hulburt’s experiments, the degree of dissociation was estimated by measuring the change of pressure in a closed system on passing a discharge. A correction for temperature was applied, which was based on the erroneous idea that the rise of temperature due to discharge in helium is about the same as that in hydrogen at the same pressure and power input. The method of determining the pressure depended on an empirical relation between pressure and the length of the cathode dark space in an auxiliary discharge connected to the main system ; but since the cathode dark space has not a sharply defined boundary, and the degree of dissociation is calculated from the difference of two pressures measured in this way, considerable error is possible. Furthermore, in a closed system, the rate of production of atoms is equal to the rate of recombination; and since these workers relied on a water-on-glass film to inhibit heterogeneous recombination, and as the power input was 200-1000 W, the catalytic activity of the walls must have been very variable and large (Part II). Crew and Hulburt’s curves con­necting degree of dissociation with pressure and power input cannot, there­fore, be credited with quantitative significance.