SIMULTANEITY IN THE COMPTON EFFECT

The question of the experimental proof of the simultaneity of the appearance of a scattered $\gamma$-photon and a recoil electron in the Compton effect played an important role in establishing the applicability of the conservation laws to an elementary act of scattering. As is known[^1], the existence of this simultaneity was first shown by Bothe and Geiger, who established that two point counters of Geiger, one of which, by virtue of its construction, counted only $\gamma$-photons and did not respond to recoil electrons, while the other counted only the latter, in the Compton effect operated simultaneously. However, the assertion of simultaneity here must be understood in the sense: “simultaneously within the limits of the time-resolving power of the scheme employed.” This resolving power in the experiment of Bothe and Geiger was $10^{-3}$ sec. Thus, the exact formulation of the result of the Bothe and Geiger experiment consists in the fact that “the difference in time between the emission of the scattered $\gamma$-photon and the appearance of the recoil electron does not exceed $10^{-3}$ sec.”

Subsequently, in connection with the experiments of Shankland[^2], which allegedly proved the inapplicability of the conservation laws to elementary acts of scattering, the Bothe and Geiger experiment was repeated with improved technique, and it was shown that simultaneity takes place within $10^{-4}$ sec, i.e. an order of magnitude more accurately than in the first experiments. However, even this interval of time many times exceeds the “theoretical simultaneity” which should occur with strict applicability of the conservation laws to an elementary act of scattering. The admissible non-simultaneity of both processes in the ideal case is of the order of the Compton wavelength divided by the velocity of light, i.e.

\[ \frac{0.024 \cdot 10^{-8}}{3 \cdot 10^{10}} \simeq 10^{-20}\ \text{sec}. \]

Therefore, although a positive answer to the basic question that gave rise to these experiments—the applicability of the conservation laws—at present no longer gives rise to doubt, repeating them with schemes that make it possible to record simultaneity within the limits of an ever higher resolving power is of interest.

In a recently published paper[^3], owing to the use of a new, considerably improved counting technique connected with the application of scintillation counters, Bothe’s experiment was repeated with a considerably improved resolving power, and it was shown that simultaneity in the Compton effect takes place within $1.5 \cdot 10^{-8}$ sec, i.e. with an accuracy four orders of magnitude greater than that obtained earlier. The experiment was set up as follows. As the source of $\gamma$-rays, a Co$^{60}$ preparation was used, giving $\gamma$-photons with energies of 1.17 and 1.33 MeV. A narrow parallel beam of $\gamma$-rays was separated by a thin channel in two lead blocks of total length 27.5 cm. This beam fell on an organic stilbene crystal, which the authors call the “scatterer.” This crystal played a double role. First, Compton scattering occurred in it, and secondly, the recoil electrons arising in this scattering produced luminescence in the crystal itself, which was recorded by a pair of exactly identical photomultipliers. For counting the scattered $\gamma$-photons, a second stilbene crystal was used, placed at some distance from the first (in most experiments, at a distance of 7.3 cm); this second crystal—the “detector”—could, in addition, operate at various angles relative to the primary beam of $\gamma$-photons. Flashes from the scattered photons arising in this crystal were also recorded by a pair of connected exactly identical photomultipliers. Both pairs of counters[^4]...

were connected to amplifying radio-engineering circuits, a substantial feature of which was the presence in each of them of a long line (up to 100 meters), thanks to which it was possible, with simultaneous signals at the inputs, to create at the output a definite delay of one signal relative to the other. The amplifying circuit of the detector photomultipliers fed, through the long line to the output, a voltage to one vertical plate of the cathode-ray oscilloscope, while the corresponding signal from the scatterer-crystal photomultipliers was fed to the opposite plate. As a result (of course, with a corresponding sweep), the detector photomultipliers and the scatterer photomultipliers produced on the screen of the cathode-ray oscilloscope peaks directed in opposite directions; if one was directed upward, the other was directed downward. The distance between these peaks corresponded to the delay of one signal relative to the other. For calibration, knowingly simultaneous signals were fed to the beginnings of both long lines, namely either simply one and the same signal, or signals from both pairs of counters, which, however, in this case were placed near one crystal scatterer and thus registered one event—the flash from the recoil electron. The time calibration was carried out with the aid of the curve of a standard generator with a frequency of 10 MHz on the same oscilloscope screen. Comparison of such coincidences, caused simultaneously by the curve, by the gamma-Compton effect, showed that in 89 cases, where both peaks of the curve were due to the registration of a single elementary Compton process, the registration of the recoil electron and of the scattered photon coincide in time within \(1.5 \cdot 10^{-8}\) sec.

E. Sh.

References Cited

1. E. V. Shpolskii, Atomic Physics, Vol. I, p. 265 (1950).
2. E. V. Shpolskii, Uspekhi Fizicheskikh Nauk 16, 458 (1936); K. S. Vul'fson, Uspekhi Fizicheskikh Nauk 17, 83 (1937).
3. R. Hofstadter and J. A. McIntyre, Phys. Rev. 78, 24 (1950).