Gaseous Detectors

PHY6040

Particle Detectors

Dr C N Booth

Gaseous detectors exploit the ionisation produced by a charged particle as it passes through matter. Full details of these detectors aregiven in the handout. This page contains a summary of the key features of gaseous detectors, with links to relevant sections of the Particle Detector BriefBook.

Gaseous Detectors

When a charged particle passes through a gas, free electrons and positive ions are produced along its track. If no electric field is applied, ion pairs will recombine, and no signal is produced. By applying different strength fields, different types of detector can be realised.

If a small field is applied, the electrons drift towards the electrodes. The signal collected is just the original ionisation produced. This regime is exploited in an ionisation chamber.

When a larger field is present, the electrons are soon accelerated until they become ionising themselves. In this way amplification is produced due to secondary ionisation. At low gain, the resultant signal remains proportional to the original ionisation, and this is known as operation in "proportional mode".

As the field is increased, so does the gain, and eventually the amplification becomes non-linear. This regime is known as the "semi-proportional mode".

At still larger fields, the positive and negative charges produced in the secondary shower become so large that they screen much of the charge from the applied field. In this case, the large resulting signal is independent of the original ionisation, and the signal is said to be saturated.

As the field increases still further, recombination of ion pairs in the centre of secondary showers leads to the production of photons which can ionise other gas molecules anywhere in the detector volume. This is the Geiger-Müller effect, and produces very large saturated signals.

If the field is increased any further, then the gas is likely to break down even in the absence of primary ionisation.

Signals from Ionisation Detectors

By considering the electrodes forming a detector as the plates of a capacitor, and examining the work done in moving a charge between the plates, it is possible to show that the signal induced on the electrodes by a charge q moving through a distance δx in a field E is

where C is the capacitance of the electrodes and V₀ the applied voltage.

Ionisation Chambers

When a charged particle passes through gas, it causes ionisation. If a small electric field is applied, the electrons and ions will drift in the field and be collected at the anode and cathode. A minimum ionising particle leaves about 120 ion pairs per cm, so the signal produced is very small. With a voltage of about 100 V applied across a small chamber, the drift times will be typically 10 microseconds for the electrons and 10 ms for the heavier ions.

Ionisation chambers are therefore

insensitive to single minimum ionising particles, and more useful for heavy (slow) ions
slow, suitable for an environmental radiation monitor with a DC current indicating a steady flux of particles.

If a liquid is used rather than a gas, the higher density (and lower ionisation potential) lead to a considerable increase in signal. Suitable liquids include

liquid argon (which requires a cryogenic system)
room temperature liquids, such as TMP (tetramethyl pentane) and TMS (trimethyl silane), which must be kept extremely pure to avoid electron capture.

Proportional Chambers

In an increased electric field, the electrons are accelerated and produce secondary ionisation in an avalanche process. A gain, or gas multiplication, of 10³ to 10⁵ is then possible with a produced signal proportional to the original ionisation. A detector using this technique is known as a proportional counter. The simplest way of producing a local region of high electrical field is to utilise a cylindrical geometry with a very fine central "sense" wire acting as anode. In this way, the amplification region is surrounded by a lower field, and there is no danger of a direct breakdown between the electrodes. The maximum of the avalanche occurs very close to the wire. The electrons therefore have a very short drift distance (δx in the equation above) before being collected at the anode, and so contribute little to the induced signal. The ions, however, even though they are more slowly moving, produce a significant signal as they travel through the region of large E.

Choice of Gas

We have already noted that materials with high atomic number have the highest specific ionisation. There is also an avalanche at lower electric fields in inert gases, due to the lack of competing mechanisms for energy dissipation, such as vibration etc. Argon therefore forms an ideal gas to use. (Xenon is even better, but is much more expensive!)

However, when positive argon ions (with an ionisation potential of 11.6 V) reach the cathode, they are neutralised and often release an 11.6 eV photon. Due to the photoelectric effect, this can liberate an electron from the cathode, which may in turn lead to spurious showers. (This effect becomes a problem for gains greater than about 10³.) A quenching agent, consisting of an organic gas such as isobutane, is therefore added. This has a lower ionisation potential than argon, so when the ion strikes a neutral isobutane molecule, an electron is transferred from the organic compound to the argon. (The excess energy excites rotation and vibration of the molecule.) It is the isobutane ion which then drifts to the cathode, and neutralisation there leads to non-radiative processes including dissociation and polymerisation of the organic molecule.

A small quantity of a highly electronegative gas such as freon (CF₃Br) is also added to "mop up" any free electrons produced at the cathode. This allows operation with a gain up to 10⁷ without any breakdown of the chamber.

A typical mixture of gases would therefore be argon, isobutane and freon in the ratio of 70% : 29.6% : 0.4%. (Such a mixture is often known as "magic gas".) A slow continuous flow of the gas is required to replace the dissociated quenching agent.

Applications of Proportional Chambers

1) Multiwire Proportional Chambers (MWPCs)

A development of proportional chambers was to place many detectors in the same gas volume. This is done by positioning parallel fine anode wires about 2 mm apart between planar walls which form the cathode. If each wire is about 20 microns in diameter, an intense field will be produced in its vicinity, and each will act as an independent detector, if equipped with amplifier and readout electronics. The position resolution (perpendicular to the wires) will then be of the order of the wire spacing, though this can be improved to ~0.7 mm if the centroid of the wire pulses is used.

Resolution in the perpendicular direction can be achieved by two means.

The cathode planes can be divided into independent strips. The centroid of the induced pulses can then provide a measurement to about 0.1 mm.
If the anode wires are made of resistive material, then reading out both ends and using charge division provides a resolution of the order of 0.1% of their length.

A large number of MWPCs are in use in many high energy physics experiments.

2) Drift Chambers

In order to cover large areas with MWPCs, very many wires, and therefore many channels of amplification and readout, are required. With drift chambers, the wires are much more widely spaced. Electrons from ionisation drift in regions of low electric field before reaching the high field or avalanche region near a wire, where amplification and detection occur. The drift time is then a measure of the position of the original particle. (Note that to measure this drift time, the transit time of the original particle must be known, e.g. from independent scintillation counters.

In addition to anode and cathode wires, which may be spaced 10 cm apart, field-shaping wires ensure a uniform drift field. With a drift velocity of the order of 40 microns per ns, drift times up to 2 microseconds are possible, so drift chambers are most suitable for experiments with low rates (e.g. e⁺e⁻ colliders) or pulsed machines. The spatial resolution is limited by electron diffusion during the drift process, and is typically 0.1 mm.

Various geometries are possible for drift chambers, including planar detectors and cylindrical (with wires parallel to the axis or at small angles to provide a stereo view). In so-called "jet" chambers, there are drift cells forming sectors of the cylinder, with staggered sense wires to resolve left-right ambiguities.

A specialised form of drift chamber is the Time Projection Chamber (TPC). This contains a large drift volume (2.2 m long by 1.7 m radius in the case of ALEPH). A field of the order of 100 kV per metre drifts charge onto MWPCs at the end of the drift volume. This provides a 2-dimensional x-y readout, with the third co-ordinate being supplied by the drift time before the signal arrives at the MWPC. Spatial resolution in 3D to 60 microns can be achieved with such a TPC.

Problems with Proportional Chambers

Proportional chambers are used very successfully in a large number of experiments. They do, however, suffer from three potential problems:

Mechanical - very fine wires are strung under high mechanical tension and significant electrostatic forces, and should be kept in their nominal position with a tolerance of the order of 50 microns.
Physical and chemical breakdown of gases can lead to deposits on the cathode, which in turn can cause electrical discharge within the chamber.
There are overall rate limitations, due to the build up of space-charge (mainly due to the slowly drifting positive ions).

A relatively recent idea (from G. Charpak) is the multistep chamber. A very low gas amplification in the "active" chamber volume reduces all of the above problems.& A pulsed wire grid then admits only a small fraction of selected events into a high-gain region for further amplification.

Streamer Chambers

If the electric field is increased still further, the gas gain can increase to values as large as 10¹⁰. The chamber is then operating in saturated mode, and recombination of electrons and ions in the middle of ionisation showers leads to the production of visible streamers. In a streamer chamber, a uniform field is applied to a large volume. (In the UA5 experiment, the chambers were 6 x 1.25 x 0.5 m.) If a DC field was applied, the ionisation would spread to both electrodes, and a breakdown occur. However, by applying a short pulse (of duration ~12 ns), the streamers are localised along the track of a charged particle, being 0.1 to 1 mm long. The streamers can then be photographed through transparent electrodes (formed of fine wire mesh), resulting in pictures of particle tracks rather like bubble chamber photographs. The density of streamers along the track is proportional to the energy loss of the traversing particle, and so can be used to aid in particle identification. (Note that this is still true in saturated more provided the ionisation density is reasonably low, even though the charge within each streamer is independent of the primary ionisation which caused it.)

The best position resolution is obtained with small streamers, though these produce less light. Image intensification is then required, together with very sensitive film.

Streamer chamber photograph

Spark Chambers and Geiger-Müller Counters

With a very large electric field (and a longer voltage pulse than used for streamer chambers) the photons produced by recombination lead to secondary avalanches, and the discharge rapidly spreads throughout the whole detector volume. This is utilised in various forms of spark chamber. One example, where some degree of spatial resolution is maintained, is in flash tubes. Here independent inert-gas filled glass tubes are placed between planar electrodes. A pulse of the order of a microsecond is applied, resulting in the complete breakdown of those tubes through which a charged particle had passed. Readout can be either electrical, with large pulses of the order of a volt produced from each hit tube, or optical, when the ends of the tubes are viewed with a camera.

In a Geiger-Müller counter, a steady field is provided between a central wire and outer cylindrical electrode. The passage of a charged particle causes complete breakdown of the tube, and a large electrical signal, after which the tube is insensitive for a considerable period while the ionisation is neutralised.

Back to Detectors Page