RESULTS OF THE START-UP OF A PULSED TWO-CHAMBER 25 MeV STEREOBETATRON

V. A. Moskalev, B. V. Okulov, Yu. A. Otrubyannikov,
Yu. M. Skvortsov, A. G. Skorikov, V. G. Shestakov

The problem of increasing the intensity of the radiation generated by charged-particle accelerators has recently been under development in many laboratories. Interest in this problem is due to the fact that the radiation intensity of accelerators now in operation often proves insufficient for carrying out a number of interesting physical investigations of great theoretical and practical importance (for example, the study of the structure of elementary particles and atomic nuclei in nuclear physics, or the interaction of radiation with special materials in radiation chemistry). In addition, industrial enterprises (flaw detection of thick-walled products), medicine (treatment of cancerous tumors), biology (the effect of radiation on a living organism), and so forth are interested in obtaining an installation generating high-intensity radiation.

In 1960, at the Scientific Research Institute of the Tomsk Polytechnic Institute, a two-chamber pulsed 25 MeV stereobetatron with increased radiation intensity was started up and put into operation.

The design of the installation is described in [1], and the results of the preliminary start-up of the stereobetatron in the mode of power supply by industrial-frequency current are given in [2].

Principal operating modes of the stereobetatron circuits

The electromagnet of the installation is supplied from a pulsed circuit by current pulses of 2760 A at a voltage of 7.5 kV. The pulse repetition frequency is 0.2 Hz. A capacitor bank of capacitance 1200 μF is discharged through the magnetizing windings of the electromagnet and the valve system of five parallel-connected TR1-85/15 tyratrons. At the radius of the equilibrium orbit this produces, in the air gaps of the electromagnet, a magnetic field with a strength of up to 3520 Oe, which ensures obtaining the maximum calculated energy of the accelerated electrons, 25 MeV, in both accelerating chambers.

Electrons are injected into each accelerating chamber by means of an external injector designed for an operating voltage of up to 300–400 kV and operating together with an input electrostatic capacitor-inflector. The injector is supplied from a pulse generator with voltage pulses of up to 400 kV, neg—

of negative polarity with a duration of 4 μsec. A positive-polarity pulse with a flat top, an amplitude of up to 60 kV, and a duration of up to 15 μsec is applied to the inflector. The phase of the inflector pulse leads the phase of the voltage pulse on the injector by several microseconds. The relationship between the phases of these pulses can be smoothly varied by means of a special circuit, thereby ensuring adjustment of the installation for stable operation.

To select the optimum conditions for capturing electrons into acceleration, the design of the inflector plates and their mounting provides devices that make it possible, within the required limits, to vary the distance between the plates, the divergence angle of the plates, and also the angle at which the electrons enter the accelerating chamber. The design of the injector provides for the possibility of varying the interelectrode distance between the anode and the focusing electrode, as well as the depth to which the cathode is immersed in the focusing electrode. All these adjustments can be made without breaking the vacuum.

A tungsten filament coated with thorium oxide is used as the electron source. The filament consists of five separate sections connected in series with one another and arranged in a zigzag pattern (in the form of the letter W). The pulsed current of such a cathode reaches several amperes, with a service life of not less than 100 hours. The current injected from the inflector into the accelerating chamber reaches 1.6 A.

In order to obtain a radiation pulse of minimum duration, the displacement of the accelerated electrons from the equilibrium orbit is carried out in two stages. First, the winding for preliminary expansion of the orbit, placed on the central cores of the electromagnet, is switched on; in this case the electron beam begins to move along a uniformly expanding spiral to a radius close to the radius at which the target is located. Then the circuit for the final dumping of the electrons onto the target is switched on. This circuit produces current pulses with an amplitude of up to 2500 A and a duration of 1.5 μsec. To reduce stray inductances, the circuit is brought out directly to the accelerator; noninductive capacitors of the IM-100/0.1 type are used in it, and a tungsten spark gap developed in the institute laboratory is used as the switching device. The deflecting winding, containing two sector turns subtending about 120° in azimuth, is placed inside the accelerating chamber above and below the plane of the electron orbits.

The use of such a system for dumping electrons onto the target made it possible to obtain bremsstrahlung pulses not exceeding 0.2 μsec in duration. Modernization of the control circuits [3] made it possible to achieve stable operation of the installation.

Estimation of the Radiation Dose

A standard “Kaktus” roentgenometer with a DIG-1 aluminum ionization chamber of volume 1 l was used to register the radiation pulses. The instrument readings, in the form of deflections of the roentgenometer pointer, served for a qualitative assessment in tuning the installation to the optimum operating mode. Special measurements of the radiation pulses were carried out by the method of charging a definite capacitance by the ionization current of the chamber, in order to verify the readings of the “Kaktus” roentgenometer. It turned out that the deflection of the roentgenometer pointer, at any radiation intensity attainable on our accelerator, remains strictly proportional to the readings of the control instrument,

i.e., proportional to the ionization current of the chamber in a pulse. However, it did not seem possible to measure the radiation dose in accepted units (for example, roentgens), first because of the high radiation energy and, second, because all the instruments at our disposal were unsuited for measuring short radiation pulses. Therefore, for a rough quantitative estimate of the dose rate of the radiation generated by the stereobetatron in one pulse, we used the method of pulsed transillumination of a lead layer of the greatest possible thickness. A special lead wedge was made in the form of a block with 7 cylindrical recesses arranged symmetrically around the circumference, their depth varying in steps of 5 mm. As the $\gamma$-radiation indicator, Ilford x-ray film was used, with front and rear intensifying screens. The optimum intensity of the $\gamma$-radiation made it possible to transilluminate, in one pulse from one accelerating chamber at a distance of 1 m from the accelerator target, a lead block 14 cm thick.

In betatron flaw detection of thick-walled objects [4], transillumination of a steel block of equivalent thickness is achieved with a dose of about 50 r at a distance of 1 m from the target. If it is assumed that the dose in the stereobetatron beam is even 10 times smaller than this value, i.e., is only 5 r per acceleration cycle, this gives an increase in radiation output by a factor of 250–300 in comparison with existing betatrons of the same energy.

Estimation of the dimensions of the focal spots of the stereobetatron

Photographs of the focal spot from both the right and left accelerating systems were taken on the stereobetatron. For the photography, a lead photographic camera was made. With the camera lens set at a distance of 600 mm from the target, an image of the focal spot was obtained on the photographic film at a scale of 1:2.

The photographs showed that the dimensions of the focal spot in the right accelerating chamber do not exceed $4 \times 2\ \mathrm{mm}^2$, and in the left accelerating chamber do not exceed $10 \times 1\ \mathrm{mm}^2$, the focal spots being elongated in the vertical direction.

Fig. 1.

Estimation of the charge of the accelerated electrons

To estimate the magnitude of the accelerated charge, measurements were made of the electron current produced by the beam “falling” on the target.

electrons at an energy of about \(1.5\ \mathrm{MeV}\). The measurements gave a value on the order of \((5 \div 7)\cdot 10^{11}\) electrons, which corresponds to a charge of about \(10^{-7}\) coulomb. This value exceeds the charge accelerated in existing betatrons by approximately two orders of magnitude. The general appearance of the stereobetatron is shown in Fig. 1. On the basis of the pulsed stereobetatron that has been constructed, development has begun on a \(30\ \mathrm{MeV}\) betatron, which will operate at industrial frequency.

References

1. V. A. Moskalev, M. F. Filippov, A. G. Skorikov, Yu. M. Skvortsov. A high-current pulsed stereobetatron. Izvestiya vuzov MVO SSSR, Fizika, 5, 1959.

2. V. A. Moskalev, Yu. M. Skvortsov et al. Start-up of a pulsed stereobetatron at \(25\ \mathrm{MeV}\). Report at the 3rd Interuniversity Conference on Accelerators. Tomsk, 1959.

3. V. A. Moskalev, B. V. Okulov, Yu. A. Otrubyannikov, Yu. M. Skvortsov, A. G. Skorikov, V. G. Shestakov. Results of the start-up of a pulsed two-chamber stereobetatron at \(25\ \mathrm{MeV}\). Report of the Research Institute at TPI, Tomsk, 1960.

4. V. I. Gorbunov. Dissertation. Tomsk, 1958.