On the Problem of Electrons Bypassing the Injector in a Betatron During Injection
Abstract Generated abstract
This paper examines whether electrons injected into a betatron are significantly intercepted by the injector, a problem that one-electron theory had treated as unavoidable. Using a direct-current model of a 15 MeV betatron electromagnet with pulsed injector supply, the author measures current on a shutter placed behind the injector and compares it with the current associated with the closed beam during injection. The experiments indicate that only a very small fraction of electrons can strike the injector, roughly two orders of magnitude less than the captured beam current, and that this fraction does not materially affect capture. The results support the view that the injection process itself creates conditions for electrons to bypass the injector, consistent with explanations based on electrostatic interaction within electron beams.
Full Text
ON THE PROBLEM OF ELECTRONS BYPASSING THE INJECTOR IN A BETATRON DURING INJECTION
N. M. GOLOSHCHAPOV
(Presented by the scientific seminar of the Physico-Technical Faculty)
Up to the present time there has been an opinion that, during injection, the greater part of the electrons strikes the injector. Moreover, from the point of view of the one-electron theory, collisions of electrons with the injector are altogether inevitable. But since the betatron does, after all, operate, various attempts have been made to explain theoretically how electrons bypass (or, as some put it, “miss”) the injector, as well as attempts to increase the radiation output by means of various devices supposedly improving the conditions under which the electrons bypass the injector. As is known, these attempts have not yielded substantial results. This circumstance alone confirms that the bypassing of the injector by electrons is evidently ensured by the injection process itself.
We carried out the following experiment. We used a model of the electromagnet of a 15 MeV betatron, but with the magnet supplied by direct current and with pulsed supply of the injector [1]. Behind the injector there was installed a shutter with a separate lead-out. In size, the shutter corresponded exactly to the dimensions of the injector anode.
Thus, if some part of the beam can in fact strike the injector, then, by measuring the current from the shutter and observing the shape of the current pulse from it on the screen of an oscillograph, we can estimate this “harmful” action of the injector, which “intercepts” electrons. It turned out that when the injector filament is switched on and there is no injection voltage or magnetic field, a current from the shutter of the order of 0.5–1.0 μA is observed. Switching on the magnetic field does not change the magnitude of this current. The appearance of current from the shutter is explained by the fact that part of the electrons from the injector cathode strikes the stem of the injector socket and from there flows off through the shutter lead.
When the magnetic field and an injection voltage of up to 30 kV are switched on, the current from the shutter increases only very slightly (to approximately 1.5–2.0 μA, with the current from the chamber layer up to 300 μA).
Further, when the injector was displaced from the radius where \(n = 0.95\) to the radius where \(n = 1.35\) (i.e., by 7 mm), the current from the shutter increased by tenths and hundredths of a microampere.
By feeding the signal of the current pulse from the shutter to an oscillograph, it was possible to observe that the signal appears on the screen at the moment when the injection voltage corresponds to the required voltage, i.e., to the magnitude of the magnetic field.
Then an obstacle was placed in the chamber behind the injector, so that the electrons could make somewhat less than one complete revolution. In this case the signal from the shutter is absent on the oscillograph screen, while the current
from the shutter remains at the level determined by the leakage of electrons from the injector foot to the shutter lead-out (0.5–1.0 µA). Moving the injector produces no changes.
In the course of the experiments in this and another work [1] it was established that the current in the closed beam during injection is approximately two orders of magnitude greater than that which strikes the shutter, i.e., than that which could strike the injector.
Consequently, the problem of electrons bypassing the injector during the process of capturing electrons for acceleration does not play as important a role as has been assigned to it. This conclusion is also confirmed by the experiments of Goward [2] and Wideröe [3], in which a plate was moved beyond the injector into the chamber by 3–5 mm, and the intensity of the radiation from the target changed hardly at all.
Thus, although during the process of capturing electrons for acceleration a small fraction of the electrons does strike the injector, this does not hinder the capture process and, still less, cannot improve the capture conditions, as V. S. Melikhov [4] asserts.
The experiments carried out by us and the experiments of Goward and Wideröe may serve as indirect confirmation of the theoretical propositions of B. N. Rodimov [5] that the principal and determining factor in the process of capturing electrons for acceleration is the electrostatic interaction of electrons and electron beams, and that this interaction leads to favorable conditions for electrons to bypass the injector during their capture for acceleration.
References
- Goloshchapov N. M. Investigation of the properties of betatron injectors. On the existence of a closed electron beam in the constant potential field of focusing forces. Izvestiya TPI, vol. 87, 1957.
- F. K. Goward et al.—Proc. Inst. E. E. 97. 108. 320. 1950.
- Wideröe—Journ. Appl. Phys. 22. 3. 362. 1951.
- Melikhov V. S. Candidate dissertation. Tomsk Polytechnic Institute, 1954.
- Rodimov B. N. On the mechanism of capturing electrons for acceleration in a betatron. Izvestiya TPI, vol. 87, 1957.