Design and experiment of an inverse magnetron injection gun for a Ka-band gyro-TWT with non-superconducting magnet

Among high-power microwave sources, the gyrotron traveling wave tube (gyro-TWT) characterized by high power, high efficiency, and wideband has attracted much attention of researchers. It has promising applications in radar, electronic countermeasure and communication system.^1,2 Magnetron injection gun (MIG), which is the source of electron beam, plays a crucial role in the gyro-TWT. The electron beam is transported and compressed under the effect of electric and magnetic fields, then utilized for beam–wave interaction.³ 

A series of gyro-TWTs based on superconducting magnets (SCM) have been designed and experimented.^4–9 Nowadays, the millimeter waves are applied in industry and medicine fields.^10–12 If the millimeter waves are generated by gyro-TWTs, the gyro-TWT systems are expected to be moveable and quick-startup. However, these applications are limited by SCM. Though a continuous and strong magnetic field can be generated stably by SCM, the turn-on and -off time is extremely long (50 h), and a large power consumption is required to keep superconducting temperatures.¹³ As a result, non-superconducting magnet and system miniaturization are necessary. Room-temperature magnet gyrotron devices have been researched in.^13–19 Room-temperature magnet is easy to achieve at low frequencies, while it is difficult at high frequencies due to high power consumption. Researchers have tried pulsed magnets and higher harmonic gyrotron devices.^20,21

For a Ka-band gyro-TWT, to achieve the required magnetic field for the TE₀₁-mode fundamental wave using a coil magnet, the bore radius is intended to decrease from 40 mm to 20 mm, and the power consumption is only 1/4 of the 40 mm coil magnet in theory. To accommodate the magnet, a compact triode inverse magnetron injection gun (IMIG) for the Ka-band gyro-TWT is designed in this paper.

Reference 14 has proposed a compact X-band MIG with a coil magnet. It directly reduces the radial size of conventional MIG. However, it is not feasible to reduce the radial size too much due to the limitations of maximum electric field at cathode region and the space charge effect. Another paper has proposed a novel electron gun with an externally assembled anode theoretically.²² Its anode can be assembled after the tube is installed in the magnet bore. In contrast to these two solutions, this paper adopts another form of electron gun called IMIG.

Nowadays, the IMIG is used in gyrotrons and coaxial-cavity gyrotrons of high power (1–2 MW) and high frequency (>140 GHz). Researchers at the Karlsruhe Institute of Technology (KIT) and the Institute of Applied Physics (IAP) have proposed an IMIG for a coaxial 1.5 MW, 140 GHz gyrotron, and the gyrotron experimental operation results are first performed at KIT.^23,24 Furthermore, a triode IMIG for KIT 2 MW, 170 GHz coaxial gyrotron has been proposed, and its tolerance has been studied by Ruess et al.^25,26 There is direct oil cooling between SCM and MIG, and the IMIG temperature distribution is low and the thermal deformation is small.²⁷ For this paper, the structures of a Ka-band conventional diode-type MIG and a triode-type IMIG are shown in Fig. 1. The IMIG consists of a cathode, anode, modulating anode, and isolation ceramic. Unlike the conventional MIG, the IMIG's modulating anode is surrounded by the cathode, which effectively reduces the gun maximum radius.

Another important concept is about stray electrons. The electrons emit from other electrodes except emitter are called stray electrons. These undesired electrons may hit the ceramic, and in some cases, they may even move backward and hit the cathode. This may lead to harmful oscillations and outgassing in gyro-TWT. Therefore, the stray electrons should be suppressed as much as possible. Cathode isolation gaps and a cooling structure for IMIG are adopted to suppress stray electrons in this paper.

To design an electron gun with a maximum radius of less than 20 mm, a triode IMIG configuration is adopted. Since the electric field adjusting ability of the IMIG is weaker, a curved emitter and cathode steps are adopted to improve beam quality. A ceramic is installed between the cathode and anode, increasing the probability of ceramic being hit by stray electrons. Therefore, cathode isolation gaps and a cooling structure are designed to ensure low temperatures of the inner and outer electrodes, and suppress the emission of stray electrons. Furthermore, the tolerance of modulating anode and isolation gap width is studied, and the cathode surface morphology is also investigated, offering a valuable reference for subsequent application. Finally, the cathode is assembled into a gun shell, and its temperatures are measured under both cooling and no cooling conditions. It indicates that the temperatures of the inner and outer electrodes are low enough.

Based on the parameters of the Ka-band gyro-TWT in Ref. 28, the IMIG is designed. As shown in Fig. 2, the IMIG cathode consists of emitter strip, inner electrode and outer electrode. Cathode steps are utilized for a better control of the electric field distribution. A curved emitter is adopted for a better match of the electric and magnetic fields along the emitter strip. Furthermore, thermal isolation gaps are adopted to prevent direct heat conduction from the emitter to the inner and outer electrodes.²⁹ 

It is important to keep proper distances between the cathode, modulating anode, and anode to maintain an acceptable electric field distribution and prevent breakdown. In this design, the distance between the cathode and anode should not be less than 30 mm. According to adiabatic compression and trade-off principle,³⁰ the initial parameters can be determined, then the IMIG is further optimized.

Cathode steps and a curved emitter are adopted to enhance beam quality. The outer electrode step has little impact on beam quality, so the step is set as 0 mm. The influence of the inner cathode step and curved emitter are presented in Figs. 3 and 4. With the step height increasing from 0 to 0.8 mm, the velocity spread deceases first and then increases. A 0.6 mm step is adopted, and the electric field matches better with magnetic field. Based on the flat emitter strip, there is a depression in the electric field at the middle position. Then a convex emitter strip with a radius of 90 mm is adopted. With the cathode step and curved emitter, the electric and magnetic fields along the emitter match well with each other. The velocity spread decreases from 3.7% to 2.87%, and finally 2.43%.

Some important IMIG parameters are listed in Table I. The beam quality is relatively good, with a velocity ratio of 1.14, and a transverse velocity spread of 2.43%. The guiding center radius is about 2.93 mm, and the thickness of the electron beam is about 1.53 mm. After the emission of the electron beam, the minimum distance between the electrons and the modulating anode is 0.96 mm. The electron trajectories are perfectly laminar, and the electrons transport without any interception, as shown in Fig. 5. The maximum electric field is located at the left tip of the modulating anode, measuring 14.1 kV/mm. This value satisfies the design requirement. At the gun exit, the transverse velocity spread distributes between 9 × 10⁷ and 10.8 × 10⁷ m/s, but most electrons locate around 10.1 × 10⁷ m/s.

In this design, the maximum radius of the IMIG and the anode radius are 19.5 and 15.4 mm, respectively. Compared with our previously designed diode MIG,²⁸ whose maximum radius is 35 mm, a 44% reduction in gun maximum radial size is realized. The key performance parameters are listed in Table II.

Assuming that the number of electrons is proportional to the emitting surface area, the stray electron simulation results are illustrated in Fig. 6. The stray electrons from the inner electrode are shown in (a), about 37% of them are intercepted by the left side of the modulating anode. The outer electrode stray electrons, as shown in (b), enlarge the beam radius, and the minimum distance between the ceramic and the electrons is 1.36 mm at gun region. A capture structure is utilized to ensure the gap of the waveguide wall and electrons at interaction region. A portion of the outer electrode stray electrons is intercepted by the capture structure. The cross section of electrons at the gun exit is shown in (c), the average radius of the inner electrode stray electrons is smaller than the main electrons, and the minimum beam radius is 1.71 mm. The outer electrode stray electrons enlarge the beam radius to 4.13 mm, and the gap between the beam and waveguide wall is 2.57 mm.

Considering the actual situation of electron emission, the emission ability relies on factors such as surface area, electric field, temperature, and electrode material. Compared with the conventional Ka-band MIG, the electric field near the inner and outer electrodes is reduced, and the temperatures of the inner and outer electrodes are lower, which will be further discussed in the following. Consequently, the proportion of stray electrons is decreased, and the influence on modulating anode is weak.

In practical applications, the modulating anode inevitably involves some deviations during the installation process. The beam quality is especially sensitive to the parameters of the modulating anode.

The velocity ratio can be conveniently adjusted by changing the modulating voltage. The relationship between modulating voltage and beam quality is depicted in Fig. 7(a). As the negative high voltage increases, the electric field strength near the emitter intensifies. Consequently, the velocity ratio gradually rises, while the velocity spread decreases first and then increases. The minimum velocity spread is observed at around −30 kV.

As illustrated in Figs. 7(b) and 7(c), when the modulating anode deviates axially forward, the velocity ratio decreases, while the velocity spread increases. On the contrary, if modulating anode deviates in the opposite direction, both the velocity ratio and velocity spread increase. To ensure the beam quality, the forward displacement should be less than 400 μm, and the backward displacement should be less than 200 μm. The reason for this phenomenon is the changes of electric field near the emitter as the anode moves closer to or farther away from the cathode. When the modulating anode deviates radially, the velocity ratio increases slowly, while the velocity spread rises sharply. To maintain an acceptable beam quality, the maximum radial displacement should be below 100 μm. As depicted in (d), as the radial displacement increases, the non-uniformity of the electric field distribution along the emitter circle becomes greater, leading to a larger velocity spread.

The isolation gaps are positioned on both sides of the emitter strip, preventing direct heat conduction from the emitter to the inner and outer electrodes. The influence of gap width on beam quality is demonstrated in Fig. 8. The isolation gaps induce a groove-shaped decrease in the electric field near the emitter's edges, and the width of these grooves expands as the gap width increases. When the gap width is less than or equal to 0.1 mm, it has little impact on velocity spread. However, the velocity spread gradually increases as the gap width exceeds 0.1 mm. A gap width of 0.1 mm is adopted, and the velocity spread is 2.4%.

The IMIG cathode has been fabricated, and its surface morphology has been also tested. The test machine and environment are depicted in Fig. 9(a). Coordinate measuring results are presented in Fig. 9(b). The inner electrode deviates 0.06 mm, and the outer electrode deviates 0.1 mm axially. Taking into account the 0.1 mm gaps, the deviations of the inner and outer electrodes lead to a velocity spread increase from 2.4% to 2.68%.

If further considering the emissions from the upper and lower sides of the emitter strip, as shown in Fig. 9(c), the velocity spread of the main electron beam will increase from 2.68% to 3.37%. The worst condition that the emitting ability of the two sides of emitter is the same as emitter surface is under consideration.

In summary, the presence of 0.1 mm isolated gaps does not affect beam quality. The minor deviations in cathode processing have a minor impact on the main beam. The displacement of the modulating anode is supposed to be controlled within specified tolerances. These investigations provide a reference for future use.

After cathode morphology test, the cathode is installed into an electron gun shell rather than a transparent shell. The reason is that the IMIG cathode has direct contact with the gun shell in its actual structure, and the shell may conduct partial heat of the cathode. The interior of the gun shell is vacuum environment to ensure the stable operation of the filament. There is a window at the exit of the gun shell, and the temperature of the cathode can be measured by an infrared thermometer or optical thermometer from the window.

The actual structure of the IMIG cathode and its thermal resistance equivalent circuit are depicted in Fig. 10, a cooling structure is designed to lower the temperature of the outer electrode through the flow of a cooling liquid. R_c, R_t, and R_rad represent the contact thermal resistance, conduction thermal resistance and radiation thermal resistance, respectively. R represents convection thermal resistance, and T is the temperature of the cooling liquid. The presence of 0.1 mm isolated gaps results in only two heat transmission paths. One path is from the heat source to the emitter, while the other path is from the heat source to the inner electrode. If a cooling liquid, such as water or oil, is introduced into the cooling structure, it will take away some of the heat from the cathode.

The cathode thermal test method is presented in Fig. 11. The test sample that contains the cathode and its gun shell is placed on an insulation plane. A filament power supply is utilized to heat the filament, and an optical thermometer is used to measure the temperatures of the cathode from the sapphire window. The cathode profile can be seen from the window. When the emitter is heated and turns red, its temperature can be measured by the optical thermometer (>700 °C). However, the interior of the gun shell is extremely bright due to the light of the emitter and filament. So, the temperatures of the inner and outer electrodes cannot be accurately measured by the infrared thermometer or the optical thermometer. The thermocouple is also difficult to achieve due to the vacuum environment. Only the brightest part—emitter—can be measured accurately. Therefore, as labeled in the figure, the temperatures of the center of the ceramic (point 1) and the copper near the window (point 2) are measured to further ensure the accuracy of the simulation.

When there is no cooling, at a heat power of 84 W, the emitter reaches 1050 °C, and the temperatures of point 1 and point 2 are about 150 and 90 °C, respectively. When the cooling water is flowing with a velocity of 0.35 L/min, at a heat power of 115 W, the emitter reaches 1050 °C, point 1 is 42 °C, and point 2 is about 41 °C. Based on the experiment result, a thermal simulation is carried out. The heat powers are set the same as experiment, and the other parameter settings are the same in the above two cases. The emissivity of the cathode's inner surface is set as 0.3, the emissivity of the other surfaces is set as 0.1, the contact thermal conductance is set as 5500 W/m²·°C, and the convection coefficient of the gun shell is set as 20 W/m² · °C. The water convection coefficient is chosen 2000 W/m²·°C. The simulated temperatures are shown in Fig. 12 and Table III. The temperatures of the outer electrode are extremely low in both two cases. The gun shell conducts partial heat from the outer electrode and its temperature is low. In practical applications, there is no need to use liquid cooling. However, in some special cases, for example, a gun shell with a lower temperature is needed, the cooling structure is useful. The simulation and experimental results comparison is shown in Fig. 13. The deviations are all less than 10%.

Compared with the previous conventional MIG,²⁹ the IMIG has lower temperatures of the inner and outer electrodes. The emission of stray electrons will be effectively suppressed, which ensures the stable operation of the gyro-TWT.

A Ka-band triode IMIG with a maximum radius of 19.5 mm is proposed in this paper, which will be applied in a gyro-TWT with a non-superconducting magnet for mobile and quick-startup systems. Compared with the conventional MIG, the maximum radius of the gun is reduced by 44%. On one hand, a curved emitter and a 0.6 mm cathode step are designed to improve beam quality. On the other hand, 0.1 mm isolated gaps and a cooling structure are designed to suppress stray electrons. A velocity ratio of 1.14, and a transverse velocity spread of 2.43% are obtained. Then the tolerance is studied, and the morphology of the fabricated cathode is also tested. Ultimately, the cathode is assembled into a gun shell, and its temperatures are measured under both no cooling and cooling conditions. When the emitter temperature reaches 1050 °C, the heat powers are 84 and 115 W, respectively. According to simulation and experimental results, the temperatures of the inner and outer electrodes are low enough to suppress the emission of stray electrons.

The other parts of the Ka-band gyro-TWT with non-superconducting magnet are under development currently.

The authors have no conflicts to disclose.

Boxin Dai: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Writing – original draft (equal); Writing – review & editing (equal). Yelei Yao: Conceptualization (equal). Wei Jiang: Resources (equal). Binyang Han: Methodology (equal). Chaoxuan Lu: Writing – review & editing (equal). Guo Liu: Resources (equal). Zewei Wu: Resources (equal). Xinge Dai: Methodology (equal). Jinhuan Chen: Resources (equal). Jianxun Wang: Resources (equal). Yong Luo: Resources (equal).

The data that support the findings of this study are available within the article.