The conduction of electricity through gases

We are all familiar with the electric spark formed when a high-voltage discharge occurs across a region of air, and also with the bright yellow light emitted by a sodium vapour lamp. Both these are examples of the conduction of electricity through gases - the differing results being due not only to the different gases but also the different pressures under which conduction takes place. (The pressure in a neon lamp is about 10 mm of mercury.)

The effects of different gases are considered elsewhere and we will consider here only the effect of pressure changes on the discharge in air.

In dry air at atmospheric pressure a voltage of 30 kV is required to produce a spark between two spherical electrodes 1 cm apart. (For pointed electrodes the p.d. is reduced to 12 kV due to the higher field at a point.) In a thunderstorm, even allowing for the moisture in the air, you can appreciate the truly enormous voltages that are required for one lightning flash. For small potential differences, a gas is an almost perfect insulator.

At lower pressures however the potential difference to give sparking is reduced. This is because the mean free path of the electrons (distance that the electrons travel between collisions) is longer and they can therefore be accelerated to higher speeds before collision with an atom; they therefore have more chance of causing ionisation.

The following table shows the mean free path (in metres) of an electron in various gases at different pressures.
The pressure is given in mm of mercury. (760 mm is a pressure of about 10⁵ Pa.)

The discharge through gases at low pressure may be investigated using a Geissler tube. This is simply a glass tube containing air, the pressure of which may be varied, with electrodes at either end. It is unsafe to use a Geissler tube with applied potential differences above about 5-6 kV, because above this voltage X-rays may be generated by the impact of electrons with the anode and walls of the tube.


Figures 2 to 5 show the appearance of the discharge in air for various pressures.

At 20 mm pressure, violet streamers pass between cathode and anode (Figure 2).

At 5 mm pressure, a pink positive column and a negative glow appear near the cathode. These two regions are separated by Faraday's dark space (Figure 3).


At 0.1 mm pressure the positive column becomes striated, the negative glow moves away from the cathode, Crookes' dark space appears and the cathode glow appears round the cathode (Figure 4). Most of the potential difference in the tube exists across the Crookes' dark space.


At 0.01 mm pressure or less, Crookes' dark space fills the whole tube and the glass fluoresces due to electron impact (Figure 5). In 1858 Plucher demonstrated that the fluorescence could be moved about by a magnet, showing that it was due to charged particle motion.

As the pressure is reduced still further the potential difference needed to maintain the discharge rises again, and below pressures of about 10^-3 mm of mercury the tube usually becomes a good insulator again.