Full Text
OBSERVATION OF ANTIPROTONS
O. Chamberlain, E. Segrè, C. Wiegand, T. Ypsilantis*)
One of the remarkable features of Dirac’s theory of the electron is that the equations of the theory have solutions requiring the existence of an antiparticle, later identified with the positron. Extending Dirac’s theory to the proton requires the existence of the antiproton, a particle standing in the same relation to the proton as the positron does to the electron. However, until experimental evidence for the existence of the antiproton was obtained, one could doubt whether the proton is a Dirac particle in the same sense as the electron. For example, the anomalous magnetic moment of the proton indicates that the simple Dirac equation does not give a complete description of the proton.
The experimental discovery of the existence of the antiproton was therefore one of the tasks planned for the bevatron. The minimum (laboratory) kinetic energy for the production of an antiproton in a nucleon–nucleon collision is equal to 5.6 Bev. If the target nucleon is located in a nucleus and has some momentum, the threshold is lowered. Assuming that the Fermi energy is 25 Mev, it can be shown that the threshold for the formation of a proton–antiproton pair is approximately 4.3 Mev. Another (two-step) process, considered by Feldman¹, has an even lower threshold.
In investigations of cosmic radiation, several events were recorded²–⁴ that could have been caused by antiprotons, but at present these experiments do not permit unambiguous conclusions. With these initial data, we carried out an experiment whose purpose was to obtain and detect the antiproton. This experiment is based on determining the masses of negatively charged particles arising in the target of the bevatron. It consisted in the simultaneous measurement of momentum and velocity
*) Radiation Laboratory, Department of Physics, University of California, Berkeley, California, Phys. Rev. 100, No. 3, 947 (1955).
particles. Since antiprotons had to be selected against a large background of \(\pi\)-mesons, it was desirable to measure the velocity by more than one method. To date sixty antiprotons have been observed.
Fig. 1. Scheme of the experiment.
Characteristics of the individual units of the setup. \(S_1, S_2\)—plastic scintillation counters, diameter \(5.6\ \mathrm{cm}\) and thickness \(1.5\ \mathrm{cm}\).
\(C_1\)—Cherenkov counter made of fluorochemical \(0\text{–}75\) \((\mathrm{C}_8\mathrm{F}_{16}\mathrm{O})\), \(n_D = 1.276\), \(\rho = 1.76\ \mathrm{g/cm^3}\). Diameter \(7.5\ \mathrm{cm}\), thickness \(5\ \mathrm{cm}\).
\(C_2\)—Cherenkov counter made of fused quartz, \(n_D = 1.458\), \(\rho = 2.2\ \mathrm{g/cm^3}\), diameter \(5.9\ \mathrm{cm}\), length \(6.25\ \mathrm{cm}\).
\(Q_1, Q_2\)—quadrupole focusing magnets, focal length \(300\ \mathrm{cm}\), aperture \(10\ \mathrm{cm}\).
\(M_1, M_2\)—deflecting magnets, length \(150\ \mathrm{cm}\). Aperture \(30 \times 10\ \mathrm{cm^2}\), \(B \cong 13700\) gauss.
Figure 1 shows the scheme of the experiment. The proton beam of the bevatron struck a copper target, and the negative particles with momentum \(1.19\ \mathrm{Bev}/c\) moved along the orbit shown in the figure. These particles were deflected by \(21^\circ\) by the field of the bevatron and additionally by \(32^\circ\) by the magnet \(M_1\). With the aid of the quadrupole focusing magnet \(Q_1\) (consisting of three quadrupole magnets arranged in succession), these particles were brought to a focus at the counter \(S_1\), which was the first scintillation counter. After passing through the counter \(S_1\), the particles were focused again (with the aid of \(Q_2\)) and were additionally deflected (\(M_2\)) through an angle of \(34^\circ\), after which they were again brought to a focus located at the counter \(S_2\). The particles focused at the counter \(S_2\) had, within a 2% error, the same momentum.
The counters \(S_1, S_2\), and \(S_3\) were ordinary scintillation counters. The counters \(C_1\) and \(C_2\) were Cherenkov counters. Particles with the mass of a proton and momentum \(1.19\ \mathrm{Bev}/c\), arriving at the counter \(S_2\), have a velocity \(v/c = \beta = 0.78\). Ionization energy losses in passing through the counters \(S_2, C_1\), and \(C_2\) reduce the mean velocity of such particles to \(\beta = 0.765\). The counter \(C_1\) registers all charged particles for which \(\beta > 0.79\). \(C_2\) is a Cherenkov counter of special design, which registers
only particles whose velocities lie in the narrow interval \(0.75<\beta<0.78\). These counters will be described in a special paper. In principle they are analogous to some of the counters described by Marshall\(^{5}\). Registration of a particle by such a counter thus also constitutes one method of determining its velocity.
The velocity of the particles was also determined by one more method, namely from the time of flight over the distance between counters \(S_1\) and \(S_2\), equal to 120 m. Time-of-flight measurements make it possible to separate \(\pi\)-mesons very effectively from particles with the proton mass. Mesons with momentum \(1.19\ \mathrm{Bev}/c\) have \(\beta=0.99\), whereas for particles with the same momentum and the proton mass \(\beta=0.78\). The corresponding times of flight for a base of 120 m are 40 and 51 m\(\mu\)sec.
The beam passing through the system consists almost exclusively of \(\pi\)-mesons. One of the main difficulties of the experiment was the selection of a small number of antiprotons against the overwhelming background of \(\pi\)-mesons. This was achieved by selecting coincidences among counters \(S_1\), \(S_2\), \(C_2\), and \(S_3\). Coincidence of counts in \(S_1\) and \(S_2\) indicated that particles with momentum \(1.19\ \mathrm{Bev}/c\) traversed the system in a time of flight approximately equal to 51 m\(\mu\)sec. The next requirement, coincidence with \(C_2\), meant that the particles had velocities in the interval \(0.75<\beta<0.78\). Further, the coincidence condition with \(C_2\) corresponds to a measurement of the particle velocity entirely independent of the rough determination by time of flight, performed with the aid of the electronics. Finally, coincidence with counter \(S_3\) is necessary in order to be sure that the particle passed along the axis of the quartz radiator in \(C_2\) and did not undergo scattering through a large angle.
From the foregoing it is clear that, when measuring the particle velocity with our apparatus, certain errors are possible. First of all, some mesons may be registered owing to accidental coincidences of \(S_1\) and \(S_2\), even if an individual meson is completely excluded because its time of flight is too short. Secondly, the Cherenkov counter \(C_2\) may be excited by a meson (for which \(\beta=0.99\)) if it has undergone nuclear scattering in the radiator of the counter. About 3% of the mesons which, in the ideal case, should not be registered in \(C_2\), are registered in this way. Both these errors are eliminated by the presence of the guard counter \(C_1\), which registers particles with \(\beta>0.79\). A pulse from \(C_1\) indicates that the particle (meson) is moving too fast to be an antiproton with the selected momentum and, consequently, such an event must be rejected.
Pulses from counters \(S_1\), \(S_2\), and \(C_1\) are displayed oscillographically and photographed. The time of flight of the particle can be measured from the separation between the pulses from counters \(S_1\) and \(S_2\) with an accuracy up to 1 msec. The oscillogram also makes it possible to measure the magnitude of the pulse in the guard counter \(C_1\). In Fig. 2 three oscillographic photographs with pulses from \(S_1\), \(S_2\), and \(C_1\) are shown.
In Fig. 2a are shown pulses caused by a meson passing through the system. This case was recorded when the system was adjusted (for calibration purposes) to the meson time of flight. In Fig. 2b pulses from an antiproton are recorded. The spacing between the pulses from \(S_1\) and \(S_2\) corresponds exactly to the time of flight of the antiproton, and the absence of a pulse from \(C_1\) indicates that no meson passed through counter \(C_1\). In Fig. 2c a random coincidence between two mesons is recorded, occurring with such a difference in time that it was detected by the electronic circuit. The presence of a pulse from \(C_1\) or of multiple pulses from \(S_1\) and \(S_2\) is sufficient to conclude that this trace was caused by one or two mesons.
Fig. 2. Oscillograms on which, from left to right, pulses from counters \(S_1\), \(S_2\), and \(C_1\) are visible; (a)—meson, (b)—antiproton, (c)—random coincidence.
A complete test of the apparatus was carried out by changing the position of the target in the Bevatron and reversing the sign of the magnetic field in \(M_1\), \(M_2\), \(Q_1\), and \(Q_2\). In this case positively charged protons were recorded.
Each oscillographic trace of the type shown in Fig. 2 can be used for an approximate determination of the particle mass, since the magnetic fields determine the particle momentum, and the separation between the \(S_1\) and \(S_2\) pulses determines the time of flight. For protons with the momentum selected by us, when measurements are made by this method alone, the mass is determined with an accuracy of about 10%.
Decisive significance can be attached to the observed antiproton times of flight because the electronic scanning time is considerably greater than the observed spread in the antiproton times of flight. The electronic devices record events lying within \(\pm 6\ \mu\mu\text{sec}\) of the true antiproton time of flight, whereas traces obtained from the passage of antiprotons show that the spread of times of flight does not exceed \(\pm 1\) or \(2\ \mu\mu\text{sec}\). In Fig. 3a a histogram of meson times of flight is shown; in Fig. 3b an analogous histogram for antiprotons is given. A large number of the photographs (about \(2/3\)) taken during antiproton recording are caused by random coincidences. A histogram of the apparent times of flight for random coincidences is shown in Fig. 3c. It should be noted that the random coincidences do not reveal a grouping of times of flight about a single value, characteristic of the times of flight of antiprotons or mesons.
Mass measurements. Further checking of the apparatus was carried out by tuning it to particles of different masses in the mass region
proton. A check on the reality of the existence of the newly discovered negative particles is that the maximum of their intensity falls at the proton mass, with a small background in the adjacent mass regions. By varying only the values of the magnetic field in \(M_1\), \(M_2\), \(Q_1\), and \(Q_2\), particles of different momenta can be selected. If, with
Fig. 3. (a) Histogram of the flight times of mesons used for calibration. (b) Histogram of the flight times of antiprotons. (c) Apparent flight times for a group of random coincidences. The flight time is in \(10^{-9}\) sec; the ordinates give the number of events in a time interval of \(10^{-10}\) sec.
this velocity selection left unchanged, the apparatus is thereby tuned to particles of different mass. Such tests were carried out for positive and negative particles with masses near the proton mass. Figure 4 shows the curve obtained for positive protons, which represents the mass-resolution curve in our apparatus. Figure 4 also gives the experimental points obtained with antiprotons. These observations indicate the existence of an intensity maximum at the proton mass and the absence of background when the apparatus is tuned to masses appreciably larger or smaller than the proton mass. Such a check is one of the most important for establishing the reality of our observations, since the background, if it exists, should appear in various mass regions registered by the apparatus. On the basis of the proton maximum obtained
it may be asserted that the mass of the new particles agrees, to within 5%, with the mass of the proton. Mainly on this basis the new particles must be identified with antiprotons.
Fig. 4. The solid curve gives the form of the proton line in the apparatus. Also shown are experimental points obtained with antiprotons.
Fig. 5. Excitation curve giving the ratio of the cross sections for production of antiprotons to the cross section for production of mesons, as a function of the energy of the Bevatron beam. Along the ordinate axis is the number of antiprotons per \(10^5\) \(\pi\)-mesons.
Excitation function. A very crude determination was made of the dependence of the cross section for production of antiprotons on the energy of the Bevatron proton beam. A more exact study of this dependence is a matter for the future, since up to now there has been no possibility of reliably determining the intensity of the beam incident on the target. Furthermore, the solid angle of the detecting system depends on the energy of the Bevatron, since the form of the orbit along which the antiprotons are emitted depends somewhat on the magnitude of the magnetic field in the Bevatron magnet. It is possible, however, to measure the ratio of the number of antiprotons to the number of mesons (both the former and the latter with momentum \(1.19\ \mathrm{BeV}/c\)) emitted in the forward direction from the target, as a function of the Bevatron energy. The approximate excitation curve obtained in this way is shown in Fig. 5 by three experimental points. Even at \(6.2\ \mathrm{BeV}\), one antiproton corresponds to 44,000 \(\pi\)-mesons. If the decay of the \(\pi\)-mesons in the detecting system is taken into account, this number will correspond to 62,000 \(\pi\)-mesons generated in the target. From Fig. 5 it follows that at low energies no appreciable production of antiprotons occurs. Despite the fact that the production of antiprotons does not show as rapid an increase with energy as might have been expected, the data obtained indicate the existence of a reasonable threshold for antiproton production. It should be noted once again that Fig. 5 gives the ratio of the excitation function of antiprotons to the excitation function of mesons.
...and that the true excitation function is at present unknown. As soon as the excitation function for meson production becomes known, data similar to those presented in Fig. 5 will make it possible to obtain the true excitation function for antiprotons. It should also be noted that the emission angle from the target changes somewhat with the change in the energy of the bevatron. At 6.2 Bev it is equal to 3°, at 5.1 Bev—6°, and at 4.2 Bev it is 8° with the forward direction from the bevatron target.
Possible sources of error. The possibility that negative hydrogen ions might be mistaken for antiprotons can be rejected on the basis of the following arguments. It is highly improbable that such an ion would pass through all the counters without losing its electron. It should be added that, with the exception of a few feet near the target, the entire trajectory of the particle in the apparatus passes through gas at atmospheric pressure or through air, or else (near the magnetic lenses) through helium introduced to reduce multiple scattering.
None of the known heavy mesons or hyperons have masses by which the results obtained could be explained. Moreover, no such particles are known with a lifetime sufficiently long for them to pass through the apparatus without decaying. Indeed, for a particle of proton mass the flight time through the apparatus is equal to \(10.2\cdot 10^{-8}\) sec. However, such a possibility cannot be completely discarded. In identifying the new particles with antiprotons, one should keep in mind the possibility that there may exist, at present unknown, negative particles with a mass very close to 1840 electron masses.
Observation of the magnitudes of the pulses in counters \(S_1\) and \(S_2\) shows that the new particles are singly charged. Multiply charged particles cannot explain the results obtained.
At present, in our laboratory and in Rome, Italy, searches are being carried out, using emulsions irradiated at the bevatron, for phenomena occurring at the end of the antiproton range. So far, however, no positive results have been obtained. Together with other physicists, an experiment is being prepared to record the energy released when an antiproton is stopped in a large Cherenkov counter made of lead glass. Observation of annihilation processes involving an antiproton in a Wilson chamber is also planned; the apparatus described here is to be used to control the chamber.
FROM THE TRANSLATOR
The authors of the article write that “at present, using emulsions irradiated at the bevatron, searches are being carried out for phenomena occurring at the end of the antiproton range.” In Science News Letter of December 24, 1955, a photograph is given of the first star produced
as a result of the impact of an antiproton at rest into a heavy nucleus of the emulsion. This star (Fig. 6) was found by the Rome group of physicists examining emulsions irradiated in the Bevatron antiproton beam. The antiproton track is denoted by \(L\). Tracks \(a\) and \(b\) apparently belong to mesons; the remaining “black” tracks belong to protons and \(\alpha\)-particles that emerged from the nucleus as a result of the “evaporation” process.
Fig. 6. Photograph of a star produced as a result of the impact of an antiproton at rest into an emulsion nucleus.
From the photograph presented it is difficult to estimate the energy released in the annihilation of the antiproton; however, the presence among the particles of the star of two relativistic mesons and 7 “evaporation” particles indicates that this energy is of the order of Bev.
CITED LITERATURE
- G. Feldman, Phys. Rev. 95, 1967 (1954).
- E. Hayward, Phys. Rev. 72, 937 (1947).
- Amaldi, Castagnoli, Cortini, Franzinetti and Manfredini, Nuovo Cimento 1, 492 (1955).
- Bridge, Courant, DeStaebler and Rossi, Phys. Rev. 95, 1101 (1954).
- J. Marshall, Ann. Rev. Nuc. Sci. 4, 141 (1954).