USE OF WILSON CHAMBERS WITH PULSED ACCELERATORS

E. Hayward*)

In the course of the last several years Wilson chambers have been used together with pulsed accelerators. The purpose of this article is to discuss some of the special techniques employed under these conditions, and to clarify the advantages and shortcomings that they possess in comparison with the usual Wilson-chamber technique used, for example, in cosmic-ray investigations.

Up to the present time Wilson chambers have operated successfully in $\gamma$-ray beams from synchrotrons and betatrons and in neutron beams from cyclotrons. A beam of $\gamma$-rays from a betatron or synchrocyclotron can be obtained by directing the electron beam moving around the circumference onto an internal target. The electrons will radiate, and a beam will arise consisting of quanta of all energies up to the maximum energy of the electrons. Cyclotrons usually accelerate deuterons or protons. These particles, striking a target, produce neutron beams. In both cases the beams obtained do not ionize directly. However, when these beams interact with matter, secondary particles arise that ionize intensely. Such conditions are ideal for a Wilson chamber. If a Wilson chamber is placed in a beam of $\gamma$-quanta or neutrons, then, of the several thousand $\gamma$-quanta or neutrons that pass through the chamber, only those that interact with matter will be recorded.

The most remarkable feature of using a Wilson chamber with an accelerator is that the experimenter can control the moment at which the pulse from the accelerator arrives. When measuring curvature in a magnetic field, the trajectories must be free of distortions and as sharp as possible. For this purpose

*) E. Hayward, Science 111, 349 (1950). Abridged translation by V. A. Troitskaya.

it is desirable to introduce the beam at the end of the expansion, after the motion of the gas has ceased, but before the thermal gradient existing between the gas and the chamber walls can produce convection currents. Since the ions hardly diffuse at all before droplets condense on them, this reduces distortions and increases the sharpness of the trajectories. It should be noted that in most cosmic-ray experiments the opposite situation occurs, since in these experiments the Wilson chamber is triggered by a pulse from counters after a particle has passed through it. In this case, any irregularities in the motion of the gas in the chamber during expansion affect the trajectories. In addition, there is slow diffusion of ions in the interval of time between the passage of the particle and the beginning of droplet condensation on the ions. The fact that the experimenter can regulate the moment at which the pulse appears is of great importance in cases where the chamber must operate in a magnetic field. In this case it is not necessary for the magnet to be switched on all the time. It is sufficient to switch it on by pulses, which must be synchronized with the operating period of the chamber in such a way that the field reaches its maximum value at the moment the particles pass through the chamber. This mode of operation reduces heating of the magnet and thereby makes it possible to use larger current pulses to create the magnetic field. In ordinary operation of the Wilson chamber, an electric field is produced in it for removing the ions that have arisen between expansions. Then, before the rapid expansion, this field is switched off. After the rapid expansion the droplets move toward the bottom of the chamber, and some of them partially evaporate on the way. Having partially evaporated, the droplets remain suspended in the gas, since they are still too large for the clearing field to be able to remove them. They serve as condensation centers in subsequent expansions. Usually, in order to remove them, it is necessary to carry out one or several slow expansions, during which vapor again condenses on them and they fall out. A slow expansion takes a great deal of time, since after each expansion the gas temperature must equalize with the temperature of the chamber walls. Gertner and Jeter¹ very successfully eliminated the need for slow expansions. They sharply raise the pressure in the Wilson chamber immediately after photographing. In doing so the gas is heated, and the charged drops evaporate, becoming light enough to be removed by the strong clearing field. This technique was developed to such an extent that it proved possible to take photographs every five seconds. Since the usual operating cycle of a Wilson chamber lasts a minute or more, this improvement is of great importance for experiments in which the Wilson chamber is used together with accelerators.

To illustrate the work done up to the present time, we have tried to collect examples of various phenomena produced with the aid of pulsed accelerators. The first eight photographs relate to phenomena produced by neutrons ^2, ^3, ^4, ^5, and the last four by γ-rays ^6, ^7, ^8, ^9 (see the insert at the end of the issue). Each photograph corresponds to phenomena arising after a single pulse from the accelerator. The phenomena recorded on it occurred during the passage through the chamber of several hundred and even thousands of neutrons or γ-quanta (Fig. 1). We studied the phenomena arising in the collision of particles of high energy with atomic nuclei. The photograph shows six nuclear disintegrations, or “stars,” produced during the passage through a Wilson chamber of a pulse of neutrons from a cyclotron with an energy of 90 MeV (about 30,000 neutrons). The chamber was filled with hydrogen and saturated with a mixture of alcohol and water vapors. Since hydrogen nuclei consist of one proton, their collision with a neutron leads to the appearance of a single track. The cases recorded in the photograph therefore correspond to the complete disintegration of the carbon and oxygen nuclei present in the vapor. This means that in most disintegrations at least one fragment remains unnoticed, since observations are made only of charged particles. Since the neutron has no charge, it does not create ions and leaves no track in the Wilson chamber. The most common atoms of carbon and oxygen have nuclei consisting of equal numbers of protons and neutrons (six and eight, respectively). Since observations are made only of charged particles, interpretation is usually difficult, and often simply impossible. The photograph shows six stars consisting, if counted from top to bottom, of three, four, five, two, three, and four branches. First of all it should be noted that all these trajectories are arcs of circles. This is, of course, the result of the action on the moving charged particle of the force arising in the magnetic field. Since all the particles are positively charged, they curve in one direction, namely clockwise. It should also be noted that the trajectories of the heavier particles near the end of their range show considerable deviations from a regular circle. Since they move comparatively slowly and are multiply charged, the collisions that they undergo with the gas nuclei are sufficient to cause large deviations of these particles. Therefore measurements of curvature made for such trajectories are meaningless. The thin tracks most likely belong to singly charged particles, apparently protons. The fourth star belongs to the most usual type. It consists of one track of a very fast proton moving forward, and a small region of charged droplets created by the recoil nucleus. The last star ...

at the bottom of the photograph was produced by four doubly charged particles. Apparently, in this case there is a complete splitting of an eightfold-charged oxygen nucleus, and it is quite probable that all four tracks belong to $\alpha$-particles, since the latter are among the most stable nuclear formations. The formation of such stars can be understood from the point of view that a fast neutron, striking a nucleus, knocks out a fast particle moving in the direction of the neutron’s motion and leaves the nucleus in an excited state. After a very short time the excited nucleus explodes, emitting particles of small energies in all directions. If the fast particle moving forward is not observed, this may mean that it is a neutron, which leaves no track in the gas of the chamber.

Fig. 2 was obtained\(^8\) in the study of fast particles moving in the direction of a beam of fast neutrons when the latter collide with nuclei. Instead of using the chamber gas as the target, a small carbon plate is placed in the chamber, and the neutrons strike it. This target emits a large number of particles, and some of them remain in the field of view long enough to pass through a glass absorber placed along the diameter of the chamber. This absorber helps in determining the type of particle. A simple measurement of the curvature gives only the momentum of the particle. However, if the curvature of the particle is measured on both sides of an absorber of known thickness, this makes it possible to estimate the mass of the particle with an accuracy at least sufficient to indicate that it is a $3d$ particle. The two tracks indicated by arrows have approximately equal radii from their emergence from the target to their entry into the absorber; however, after emerging from the absorber one of them has a considerably smaller radius and a noticeably increased ionization. The other particle shows no noticeable change in these quantities. The first particle is a deuteron, and the second is a proton.

Fig. 3 was obtained\(^2\) in the study of the angular distribution of protons scattered as a result of elastic collisions with neutrons of energy 90 MeV. Experiments of this kind on scattering give very important information about the forces of interaction between elementary particles. In the photograph we see the tracks of six protons that arose in the gas ($\mathrm{H}_2$) after collisions of protons with invisible, but very fast, neutrons. The energy of the neutrons can be obtained with the help of the laws of elastic collisions and simple measurements of the curvature of the proton trajectory, as well as of the angle between the direction of its motion and the direction of the neutron beam.

Fig. 4 was obtained in analogous experiments\(^5\), in which 13-MeV neutrons were used. In this case the chamber was filled with methane to a pressure of 22 atmospheres, so that the recoil protons could

arise and come to rest in the gas. Since their energies could be obtained from their ranges, there was no need to use a magnetic field.

Mesons are the newest particles in nuclear physics. This name includes all particles whose mass lies between the masses of the proton and the electron. At the present time we are well acquainted with two kinds of mesons: $\pi$-mesons and $\mu$-mesons.

$\pi$-mesons have a mass equal to 276 electron masses, carry either a positive or a negative charge, and interact strongly with nuclei. They are of very great interest, since it is supposed that they are responsible for the forces binding the atomic nucleus. A meson may be emitted in a collision in which there is an excess of energy equivalent to its mass; it may appear as one of the prongs of a star produced by a particle of very high energy. In Fig. 5 just such a case is recorded. This photograph was obtained in a beam of neutrons produced by irradiating a target with 350-MeV protons. A star consisting of four branches arose in the splitting of an argon nucleus. One of the fragments is deflected by the magnetic field in a direction opposite to the deflection of the other fragments. It should therefore be assigned a negative charge. It is assumed that in this case we have an example of the production of a negative $\pi$-meson. Its energy is 60 MeV. Although we cannot measure the ionization of this track, its density agrees with such an assumption.

When a positive $\pi$-meson comes to rest, it decays into a $\mu$-meson (210 electron masses) and some neutral particle. The $\mu$-meson thus produced always has an energy of about 4 MeV and, on coming to rest, decays into an electron and two neutrinos. The energy contained in the mass of the $\pi$-meson is distributed among the rest energies and kinetic energies of these three particles. From the conservation laws it follows that the maximum energy that can be transferred to the electron is 55 MeV, although it may also have any energies smaller than this.

In Fig. 6 a $\pi$-meson is seen entering an absorber and decaying in it. It is identified as a $\pi$-meson because its trajectory has the appropriate curvature and ionization for it to be able to stop in the absorber; a $\mu$-meson with the same curvature would have passed through the absorber.

The $\mu$-meson, which is the product of its decay, is also slowed in the absorber. The electron, whose thin track is visible in the corner of the Wilson chamber, arose in the decay of the $\mu$-meson.

If a negative $\pi$-meson comes to rest in matter, it is captured by a nucleus. In this process the nucleus receives a certain excess energy, which causes its disintegration. In Fig. 7

a negative $\pi$-meson track is shown, on which a decrease in the radius of curvature is visible as its ionization increases. The stopped meson is captured by the nucleus of an atom. The star produced in this process consists of a single track, created by the recoil nucleus; apparently neutrons were also emitted in this process.

Fig. 8 is given to illustrate how the trajectory of a meson with a small energy can be detected amid a very large background of nuclear fragments by its curvature and ionization. The ionization density along the track marked by the arrow is considerably greater than the density along an electron track. At the same time this track cannot belong to a proton, since protons moving in so small a circle always stop without completing a full revolution. The track belongs to a meson, but it is impossible to establish whether it is a $\pi$- or a $\mu$-meson. It is very important to study the production of mesons, their decay, and the ways in which they interact with various nuclei.

Figs. 5, 6, and 7 are the best photographs of artificially produced mesons obtained in a Wilson chamber up to the present time, and Fig. 8 shows why such cases are so rare.

When a beam of electrons moving in a circle collides with a target placed inside a betatron or synchrotron, beams of quanta arise. The electrons are slowed down in the field of the nuclei of the target, as a result of which quanta are emitted; these may have any energy up to the full energy of the incident electron. Thus the beam consists of quanta with a continuous energy spectrum. If a quantum has an energy greater than one million electron-volts, then, upon colliding with a nucleus, it can create a pair—an electron and a positron. Approximately $1$ Mev goes into the creation of this pair; the remaining energy is converted into the kinetic energy of the electron and positron.

In Fig. 9 a case of pair formation is seen. This photograph was obtained for the case described above, when clearing of the chamber is accomplished by producing excess pressure, in a chamber placed in a beam of quanta from a $100$-Mev betatron. The conversion of the quantum into a pair occurs in a lead plate arranged perpendicular to the direction of the beam. The electron and positron are deflected by the magnetic field in opposite directions. Measurement of the curvature of the trajectory of the pair makes it possible to establish that the energy of the quantum is equal to $25$ Mev.

Fig. 10 was obtained in work$^{9}$ devoted to the investigation of the energy spectrum of photons from a $20$-Mev betatron. The pair visible in the photograph was created in the field of an argon nucleus. The trajectories of low-energy electrons are not true circles, since such electrons are scattered in the gas.

If an electron or positron moves in a medium of heavy nuclei, it emits quanta, which in turn form

new pairs. This process continues until the average energy of the quanta and electrons has decreased to such an extent that they can no longer lose energy to pair formation and quantum emission as effectively as they lose it to other processes. After this the electrons are absorbed through ionization, and the quanta through collisions with electrons, which in turn lose energy to ionization. The best way to observe such phenomena, called showers, is to place a series of lead plates in a Wilson chamber irradiated by a beam of γ-rays (Fig. 11). This Wilson chamber was in a beam produced by 335-MeV electrons incident on an internal target. The shower was produced by several hundred photons that arose in a single synchrotron pulse. In the photograph one can see a rapid increase in the number of electrons with thickness, up to a maximum under the fourth lead plate (the thickness of each plate is about 33 mm), after which the number of electrons slowly decreases as the lower-energy electrons are absorbed. Scattering of low-energy electrons in the gas (argon) is evident.

In Fig. 12 a star is visible, produced by a photon from a 100-MeV betatron[^7]. This phenomenon is analogous to the stars produced by fast neutrons, but it is rarer. The star was apparently formed by the splitting of a nitrogen nucleus, since the chamber was filled with air. A large number of electrons and positrons forming the background is visible. Their presence is sufficient to show that the quantum much more readily forms a pair than causes nuclear splitting.

References

1. E. K. Gaerttner and M. L. Jeater, Rev. Sci. Instr. 20, 588 (1949).
2. K. Brueckner et al., Phys. Rev. 75, 555 (1949).
3. K. Brueckner and W. M. Pocodl, Phys. Rev. 75, 1274 (1949).
4. W. Hartsough et al., Phys. Rev. 35, 905 (1949).
5. J. S. Langhlin and P. G. Kruger, Phys. Rev. 73, 197 (1948).
6. G. C. Baldwin and G. S. Klaiber, Phys. Rev. 70, 259 (1946).
7. Ibid. 73, 1156 (1948).
8. G. S. Klaiber et al., Phys. Rev. 71, 649 (1947).
9. H. W. Koch and R. E. Carter, Phys. Rev. 75, 1950 (1949).