5 MeV Microtron

B. Z. KANTER, V. V. LERMONTOV, D. A. NOSKOV, Yu. G. YUSHKOV

The principle of operation of the microtron was proposed in 1944 by the Soviet scientist V. I. Veksler. The microtron is a cyclic resonant accelerator with a constant controlling magnetic field and a constant frequency of the accelerating electric field. Electrons, receiving energy from the electric field while passing through the gap of the resonator, move freely in the constant magnetic field along circular orbits and again enter the resonator gap. In this accelerator, the increase in the period of revolution of the electrons with increasing energy is equal to, or a multiple of, the period of the accelerating high-frequency electric field of the resonator.

The stability of the motion of the electrons is ensured, as in other resonant accelerators, by the action of the principle of autophasing. In contrast to linear accelerators and synchrotrons, the magnitude of the phase-stable region of the microtron is \(20 \div 30^\circ\).

The microtron has a number of advantages that distinguish it favorably from other accelerators. These are, first, the simplicity of the design of the accelerator and of the required equipment; second, the high monoenergeticity of the spectrum of accelerated electrons. The large distance between the orbits considerably simplifies the problem of extracting electrons from the accelerating chamber of the microtron.

The main components of the microtron are: an electromagnet with its power-supply circuit, a vacuum chamber, a resonator with a high-frequency line, an ultrahigh-frequency generator, and a control system. The block diagram of the microtron is shown in Fig. 1. A vacuum chamber (Fig. 2), made of stainless steel, is placed in the interpolar space. The vacuum required for normal acceleration of the electrons is produced in the chamber by means of an RVN-20 fore-vacuum pump and an N-5 diffusion pump.

Fig. 1. Block diagram of the microtron. M. G.—magnetron generator; F. R.—phase shifter; S. L.—matched load; V. C.—vacuum chamber.

A toroidal resonator is used as the accelerating element in the microtron; the accelerating voltage is developed between its hollow cones.

For normal operation of the microtron, it is necessary to obtain a voltage of about 600 kV in the resonator gap.

The resonator used in the microtron is made of brass (Fig. 2). After silver plating and chromium plating it has a quality factor \(Q \sim 2000\), with a coupling-aperture diameter of 2.2 cm. In this case the power reflection coefficient is \(k = 19.5\%\) (VSWR = 2.6). The resonator is excited through a round coupling aperture from the waveguide high-frequency line connected to the magnetron generator (Fig. 1).

To ensure stable operation of the magnetron, the power coming from the generator is divided between the resonator and a matched load (ML). In order to vary the electrical length of the section of waveguide between the tee and the resonator, a phase shifter (PS) is used.

Fig. 2. Vacuum chamber of the microtron with the upper cover and resonator removed.

A phase shifter is also placed ahead of the tee in the high-frequency line; it changes the electrical length of the tee–magnetron section, which makes it possible to provide a more favorable operating regime for the magnetron.

The gap between the poles was taken to be 12.5 cm, allowing for the dimensions of the resonator, the thickness of the chamber walls, and the gaps between the chamber walls and the poles.

The diameter of the poles is determined by the radius of the last orbit (\(r = 17.43\) cm), allowing for the field falloff at the edges of the poles, and was taken equal to 55 cm.

The cross section of the microtron magnetic circuit is determined by the magnitude of the magnetic flux, allowing for leakage, and is equal to \(30 \times 12\ \text{cm}^2\).

The leakage coefficient and the field distribution in the interpolar space were checked on a model electromagnet made to a scale of 1:2.5.

The main parts of the microtron were manufactured by 1959, and immediately afterward work began on adjusting the individual units and testing them.

Measurement of the parameters of the resonator and matching of the RF line were carried out in continuous mode at low power. After this the line was tested at high power in pulsed mode.

In parallel, work was carried out on adjusting the vacuum system and eliminating a leak in the microtron vacuum chamber. At the same time, measurements were made of the energy of electrons accelerated in the resonator gap. For this purpose an auxiliary vacuum chamber and an electrostatic analyzer of electron velocities were fabricated. The measurements showed that, with the existing design of the resonator and RF system, a voltage of \(250 \div 300\ \text{kV}\) can be obtained in the accelerating gap.

On the basis of measurements and an analysis of the shortcomings of the resonator design, a new resonator was developed and fabricated, making it possible to obtain voltages up to \(400\ \text{kV}\).

With this resonator the microtron was started at an energy of \(2.5\ \text{MeV}\). A current was recorded on 9 orbits, and a glow was observed on a screen moving inside the chamber.

To investigate the possibility of increasing the power delivered by the generator to the resonator, a series of experiments was carried out, one of which was the study of the influence of the diameter of the coupling aperture between the resonator and the waveguide on the magnitude of the high-frequency power passing into the resonator.

The diameter of the coupling aperture was varied from 16 to 26 mm in steps of 2 mm, and in each case the VSWR near the resonator was measured.

The experiment carried out permits the following conclusions to be drawn.

1. As the diameter of the coupling aperture is increased, the power entering the resonator increases.
2. As the diameter of the coupling aperture is increased, the \(Q\) of the coupled resonator worsens.
3. As the diameter of the coupling aperture is increased, the matching of the resonator with the waveguide improves, i.e., the fraction of the power going to the resonator increases.

The combined action of these factors leads to the fact that, as the diameter of the coupling aperture is increased, the voltage in the resonator at first increases, reaches a maximum, and then decreases.

After these experiments the diameter of the coupling aperture was increased to \(2.2\ \text{cm}\), and the voltage across the resonator gap was raised to \(650\ \text{kV}\), which made it possible to start the microtron in the main mode at an energy of \(5\ \text{MeV}\). With an EPPV instrument, the current on 9 orbits was recorded (Fig. 3); the orbits were also observed visually on a screen through a window in the accelerating chamber.

Fig. 3. Distribution of current over the orbits of the microtron.

In the main mode, as in the “half” mode (\(2.5\ \text{MeV}\)), a rapid fall of the current is observed, and on the 9th orbit the current in the pulse amounts to several \(\mu\text{A}\).

During commissioning, work was carried out on the microtron aimed at increasing the current by using cold and heated cathodes in the resonator.

In most microtrons the introduction of electrons into acceleration— injection—is accomplished by autoelectronic emission from the surface of the electrodes of the accelerating resonator. Calculation shows that the duration of the injection-current pulses is about—

...about \(1/10\) of the period of the high-frequency voltage, and in phase these pulses coincide with the high-frequency voltage. The injection method described has substantial drawbacks: first, the emission properties of the cold surface of the resonator electrodes are almost not adjustable and change greatly with time; second, because of the large width of the energy spectrum, only a relatively small fraction of the electrons reaches the second and subsequent orbits, i.e., the useless electron loading of the resonator is large.

It is known that by using thermionic emission from a heated cathode one can obtain a broadening of the injection-current pulse, which should lead to a narrowing of the energy spectrum and a shift of its maximum toward higher energies.

Measurement results show that, in the case of injection from a cold cathode, the energy spectrum is characterized by a blunt maximum corresponding to an energy of about 70% of the maximum, and by a gradual decline toward the upper edge of the spectrum (Fig. 4).

A characteristic difference of the energy spectrum of accelerated electrons under injection from a heated cathode is a sharp maximum near the upper boundary of the spectrum. This maximum corresponds to an energy of about 90% of the maximum and changes into a sharp drop toward the upper edge of the spectrum. Its characteristic shape is preserved over a large temperature range, \(2500\text{–}3000^\circ\mathrm{K}\). Comparison of curves 1 and 2 in Fig. 4 shows that the electron current that can reach the second and subsequent orbits under injection from a heated cathode is several times greater than under injection from a cold cathode, with the electron loading unchanged.

Fig. 4. Energy spectra of electrons: 1—from a cold cathode, 2—cathode temperature \(2500^\circ\mathrm{K}\), 3—cathode temperature \(3000^\circ\mathrm{K}\).

Fig. 5. Change in the current on the orbits with increasing cathode temperature: 1—current on the sixth orbit, 2—current on the fifth orbit.

Operation of the heated cathode was checked in the acceleration regime up to \(2.5\ \mathrm{MeV}\).

The dependence of the current on the fifth and sixth orbits on the cathode temperature is shown in Fig. 5. It is clear from the graph that the use of a heated cathode of the first design made it possible to regulate the current of accelerated electrons over wide limits. Even with mismatched ...

In the resonator, average currents on the sixth orbit of about \(2 \cdot 10^{-9}\ \mathrm{A}\) were obtained.

Since the power fed into the resonator during the experiment was small, an increase in the electron load led to a decrease in the voltage on the resonator and to disruption of the acceleration regime, as is evident from the maximum in the graphs.

It is proposed to extract from the chamber of this microtron a beam of accelerated electrons. Since the distance between two neighboring electron orbits in the microtron is about \(3\ \mathrm{cm}\), extraction of the electron beam can be accomplished by means of a magnetic shunt.

For extracting the beam from various orbits, a device has been developed that makes it possible to introduce an iron tube, serving as the shunt, into the chamber and to rotate it in azimuth within the required limits.

At present, work is being carried out to determine the causes of the decrease in current on the orbits and to increase the current of the accelerated electrons.

1. Izv. TPI, vol. 122.