Search

Foilless diode are widely used in high-power microwave devices, but the traditional foilless diodes have large volume, heavy weight, and high power consumption, which are not conducive to the application of high-power microwave system on mobile platform. In order to reduce the size of the foilless diode, improve the transmission efficiency of electron beams, and reduce the weight and power consumption of the guiding magnetic field system, an axial foilless diode with a composite guiding magnetic field system is developed in this paper. By adjusting the structure size and magnetic field parameters of solenoid coil, permanent magnet, and soft magnet, the configuration of the composite magnetic field is optimized. The diameter of the anode tube is about 40% smaller than that of the original structure, and the weight and power consumption of the guiding magnetic system are about 40% lower than that of the original system when the same axial magnetic field intensity in the uniform region is generated. When the magnetic field strength of the permanent magnet is set as 1.4 T and that of the solenoid coil is in the range of 0.5 T∼1 T, the electron beam transmission efficiency is 100%, and the diode impedance is adjustable in the range of 100 Ω∼240 Ω. The experimental results verify the correctness of the simulation analysis. The experimental results show that when the magnetic field strength of the solenoid coil is 0.98 T (0.5 T) and that of the permanent magnet is 1.4 T, the transmission efficiency of the high-current annular electron beam with a peak voltage of 636 kV (590 kV) and a peak current of 3.3 kA (2.6 kA) is 100%, and the diode impedance is about 194 Ω (220 Ω).

The electrode erosion and lifetime performance of a compact and repetitively triggered field distortion spark gap switch were studied at a repetitive frequency rate of 30 Hz, a peak current of 8.5 kA, and a working voltage of ±35 kV when the switch was filled with a gas mixture of 30% SF6 and 70% N2 at a pressure of 0.3 MPa. The variations of the time-delay jitter and the self-breakdown voltage were both studied for the whole service lifetime of the spark gap switch. The morphology of both the electrodes and the plate insulator, before and after the service lifetime tests, is also analyzed. The results show that during these tests, the time-delay jitter is basically synchronized with the self-breakdown voltage jitter, and both undergo firstly a process of rapidly decreasing their values, then remaining stable, and finally and gradually increasing after 70 000 pulses. The change in the electrode surface roughness (i.e., surface profile) is caused by erosion and chemical deposits in the switch cavity, which are mainly the two factors that affect the time-delay jitter of the switch. Tip protrusions on the electrode surface, due to electrode erosion, contribute to reducing the time-delay jitter. However, due to chemical reactions, fluorides and sulfides are deposited on the switch components, as well as metal particles caused by electrode erosion sputtering. Slowly, after a large number of shots, all these phenomena affect the self-breakdown performance resulting in an increased self-breakdown voltage jitter, which also causes the time-delay jitter to increase. Although there are a number of reasons that contribute to the deterioration of the performance of the switch, it is fortunate that if a switch suffering a degraded performance is reassembled, with the electrodes mechanically polished and all the components cleaned, the optimal performance of the switch can be restored. If maintenance work is carried out regularly to preserve the condition of the switch’s inner components, the service lifetime of the switch can be prolonged.

A three-section pulse forming network (PFN) based on Guillemin type-C circuit was developed to meet the challenge of a compact design, high withstand voltage, and high-quality output waveform with fast rise time, flat-top duration, and 100-ns pulse width. A simplified pulse forming circuit was proposed and studied that includes only three LC-sections connected in parallel, with each section containing an inductor and a capacitor connected in series. The effect of the capacitance deviation on the output waveform was investigated. The simulation results show that when the capacitance deviation exceeds +3%, both the flat top and fall time of the output waveform of single PFN module deteriorate greatly. Fortunately, in a multi-stage PFN-Marx circuit, even if the capacitance deviation exceeds +10%, when the average capacitance of the same LC sections is close to the theoretical value, the output waveform maintains a good quality and is in good agreement with the theoretical prediction. The compact three-section PFN developed during this project has a size of only 360 mm × 342 mm × 65 mm, and a maximum withstand voltage of 120 kV. Sixteen PFN stages were assembled to form a Marx generator with design parameters to provide of an output peak power of 12 GW and a maximum peak current of 15 kA. The tested output waveform agrees well with the theoretical results, having a rise time of 31 ns, a flat-top of 104 ns, and a pulse with of 164 ns.

This work was stimulated by the assumption that for a gas-filled spark gap closing switch operating at a high repetition frequency, there is an optimal frequency range in which the time-delay jitter reaches a minimum value. The experiments to test this assumption use an electrically triggered, field-distortion spark gap filled with the SF6/N2 gas mixture. The results show that indeed, the time-delay jitter decreases for a range of frequencies for which the filling gas can substantially restore the interelectrode insulation before increasing at a higher operation frequency. The experimental results demonstrate the correctness of the abovepresented assumption: the time-delay jitter of the field-distortion spark gap has its minimum when the unit operates in the repetition frequency range between 20 and 30 Hz. Since the recovery time depends on the gas species and the gap distance, the optimum operation frequency range should also vary depending on the spark-gap distance and the filling gas properties.

A high resolution Thomson parabola spectrometer (TPS) based on cylindrical Halbach dipole magnets is proposed. With our design, the magnetic field can reach 1 T and a compact TPS shorter than 30 cm was fabricated. It can distinguish protons up to 100 MeV from other ions and the energy resolution for 100 MeV proton reaches Δ E/E

This paper describes the development of a small field-distortion spark gap switch. Comprehensive experimental studies made with the switch revealed its self-breakdown characteristics, voltage operating range, time delay and jitter, repetition frequency, and finally the electrode erosion processes and duration of its service lifetime. A series of design improvements were required to be implemented in the switch design. For example, to reduce the structural inductance of the switch and to improve its working characteristics, the main electrode was designed as a circular track structure. Similarly, to increase the operation stability and reduce the self-breakdown probability, the trigger electrode was designed as a disk-like structure. Moreover, to balance the gas pressure in the discharge region and increase the stability during closure, a circular hole was added at the center of the trigger electrode. The compact switch has a small size of only 150 mm $\times42$ mm, a weight of 1.5 kg, and can be successfully operated at a voltage over 110 kV, at a repetition rate between 1 and 50 Hz and having a jitter of less than 4 ns. It was experimentally demonstrated that during normal operation conditions, the switch lifetime exceeds 100 000 shots.

A compact low jitter high power repetitive long-pulse relativistic electron beam source with novel circuit structure and mechanical structure is presented. Based on Marx type circuit and direct-drive idea, the key technologies such as high energy density pulse forming network, repetitive high voltage pulse generator and low jitter electrical trigger technology are developed. The volume and weight of the pulsed electron beam source are reduced remarkably by using two-capacitor pulse forming modules, alternating superposition structure on both sides and high electric field shielding technology. The key technical parameters of the electron beam source design include a peak output voltage 1 MV with peak current 20 kA. The impulse is 180 ns and has a repetition frequency 1 ∼ 50 Hz, however, the output power and repetition frequency are able to be adjusted within a certain operating range. The machine has a total volume of 2.5 m3 and weighs 2.2 tons, making it much more compact than conventional electron beam sources. The experimental results show that when the e-beam source is operated under a single pulse condition, the output voltage, current, and power are 0.98 MV, 19.6 kA, and 19.2 GW, respectively. When the e-beam source has a continuous operating time of 10 s at a repetition frequency of 30 Hz and an output power of 16.7 GW, the voltage fluctuation range is limited to within ± 1.5%, and the current fluctuation range is limited to within ± 2%. The delay time jitter between the trigger pulse and the output pulse is limited less than 6.5 ns standard mean deviation. Statistical distributions of the delay time jitter and current amplitude variation of the 300 pulses are approximately Gaussian. When the cooling system is working, the electron beam source can work continuously for 100 s under the conditions of an output power of 16.7 GW and repetition frequency of 30 Hz, and the system operation is stable and reliable.

This paper introduces recent activities on Marx-based compact repetitive pulsed power generators at the Institute of Applied Electronics (IAE), China Academy of Engineering Physics (CAEP), over the period 2010–2018. A characteristic feature of the generators described is the use of a simplified bipolar charged Marx circuit, in which the normal isolation resistors or inductors to ground are removed to make the circuit simpler. Several pulse-forming modules developed to generate a 100 ns square wave output are introduced, including thin-film dielectric lines of different structures, a pulse-forming line based on a Printed Circuit Board, and non-uniform pulse-forming networks. A compact repetitive three-electrode spark gap switch with low-jitter, high-voltage, and high-current was developed and is used in the generators. A positive and negative series resonant constant current power supply with high precision and high power is introduced. As an important part of the repetitive pulse power generator, a lower jitter pulse trigger source is introduced. Several typical high-power repetitive pulsed power generators developed at IAE are introduced including a 30 GW low-impedance Marx generator, a compact square-wave pulse generator based on Kapton-film dielectric Blumlein line, a 20 GW high pulse-energy repetitive PFN-Marx generator, and a coaxial Marx generator based on ceramic capacitors. The research of key technologies and their development status are discussed, which can provide a reference for the future development and application of miniaturization of compact and repetitive Marx generators.

Resonant acceleration of electrons by a laser in the background of an extra longitudinal magnetic field is investigated analytically and numerically. The resonant condition is independent of laser intensity, and when satisfied, the energy gain is proportional to $a_0^2 $ and the square of phase difference. This process is mainly limited by the magnitude and spatial size of the extra magnetic field. Under the laboratory conditions, simulation results show that a monoenergetic and collimated electron bunch can still be obtained in ~ GV/cm scale, which sheds a light on the vacuum table-top laser-driven electron accelerators.

Relativistic collisionless shock charged particle acceleration is considered as a possible origin of high-energy cosmic rays. However, it is hard to explore the nature of relativistic collisionless shock due to its low occurring frequency and remote detecting distance. Recently, there are some works attempt to solve this problem by generating relativistic collisionless shock in laboratory conditions. In laboratory, the scheme of generation of relativistic collisionless shock is that two electron–positron pair plasmas knock each other. However, in laboratory, the appropriate pair plasmas have been not generated. The 10 PW laser pulse maybe generates the pair plasmas that satisfy the formation condition of relativistic collisionless shock due to its ultrahigh intensity and energy. In this paper, we study the positron production by ultraintense laser high Z target interaction using numerical simulations, which consider quantum electrodynamics effect. The simulation results show that the forward positron beam up to 1013/kJ can be generated by 10 PW laser pulse interacting with lead target. The estimation of relativistic collisionless shock formation shows that the positron yield satisfies formation condition and the positron divergence needs to be controlled. Our results indicate that the generation of relativistic collisionless shock by 10 PW laser facilities in laboratory is possible.

Laser interaction with an ultra-thin pre-structured target is investigated with the help of both two-dimensional and three-dimensional particle-in-cell simulations. With the existence of a periodic structure on the target surface, the laser seems to penetrate through the target at its fundamental frequency even if the plasma density of the target is much higher than the laser’s relativistically critical density. The particle-in-cell simulations show that the transmitted laser energy behind the pre-structured target is increased by about two orders of magnitude compared to that behind the flat target. Theoretical analyses show that the transmitted energy behind the pre-structured target is actually re-emitted by electron ‘islands’ formed by the surface plasma waves on the target surfaces. In other words, the radiation with the fundamental frequency is actually ‘surface emission’ on the target rear surface. Besides the intensity of the component with the fundamental frequency, the intensity of the high-order harmonics behind the pre-structured target is also much enhanced compared to that behind the flat target. The enhancement of the high-order harmonics is also related to the surface plasma waves generated on the target surfaces.

The coupling between a laser and a prestructured target with an arbitrary structure period is investigated with the help of two-dimensional (2D) particle-in-cell (PIC) simulations. It is shown that new electromagnetic (e.m.) waves will be generated after the coupling. Since the coupling is a resonant process, strong surface currents will be generated, which result in the generation of strong quasistatic magnetic fields. The frequencies which the newly excited e.m. waves contain are harmonics of the laser frequency and the frequencies can be controlled by the structure period. Also, the propagation directions of the newly excited e.m. waves are well controlled by the ratio between the structure period and the laser wavelength, which means e.m. waves excited by lasers with different frequencies have different propagation directions. As a result, the prestructured target can act as a new kind of optical gratings which can be used to split superintense laser pulses. The controlling of both the harmonic frequencies and the propagation directions can be explained by matching condition of the coupling, which is a formula resembling but physically different from the diffraction grating equation. The quasistatic magnetic fields are on the target front surface and as strong as hundreds of teslas, but the amplitude will be decreased by the newly excited e.m. waves when they are propagating along the target surface. Since the propagation directions are controlled by the structure period, with an optimal structure period, the prestructured target can also act as a source of strong quasistatic magnetic fields.

Imaging of DD-produced protons from an implosion hot spot region by miniature permanent magnetic quadrupole (PMQ) lens is proposed. Corresponding object-image relation is deduced and an adjust method for this imaging system is discussed. Ideal point-to-point imaging demands a monoenergetic proton source; nevertheless, we proved that the blur of image induced by proton energy spread is a second order effect therefore controllable. A proton imaging system based on miniature PMQ lens is designed for 2.8 MeV DD-protons and the adjust method in case of proton energy shift is proposed. The spatial resolution of this system is better than 10 μm when proton yield is above 109 and the spectra width is within 10%.

Electrons can be efficiently injected into a plasma wave by colliding two counterpropagating laser pulses in a laser wakefield acceleration. However, the generation of a high-quality electron beam with a large charge is difficult in the traditional on-axis colliding scheme due to the growth of the electron beam duration coming from the increase of the beam charge. To solve this problem, we propose an off-axis colliding scheme, in which the collision point is away from the axis of the driver pulse. We show that the electrons injected from the off-axis region are highly concentered on the tail of the bubble even for a large trapped charge, thus feeling almost the same accelerating field. As a result, quasimonoenergetic electron beams with a large charge can be produced. The validity of this scheme is confirmed by both the particle-in-cell simulations and the Hamiltonian model. Furthermore, it is shown that a Laguerre-Gauss (LG) laser can be adopted as the injection pulse to realize the off-axis colliding injection in three dimensions symmetrically, which may be useful in simplifying the technical layout of the real experiment setup.

The stimulated Raman scattering (SRS) instability of a left-handed circularly polarized (LH-CP) laser in strongly axially magnetized plasmas is investigated in detail with the help of one-dimensional (1D) and two-dimensional (2D) particle-in-cell (PIC) simulations. Since the LH-CP laser has a larger critical density in the axially magnetized plasmas, the SRS instability could be excited in over quarter-critical density plasmas, which is verified by the PIC simulations. This phenomenon could be used to amplify a seed with a frequency smaller than half of the laser frequency, which is impossible for traditional simulated Raman amplification. The simulation results also show that the scattered laser becomes right-handed circularly polarized. With this conclusion, we re-derive the temporal linear growth rate of the SRS instability of the LH-CP laser in the axially magnetized plasmas. The results show that the larger the external magnetic field is the smaller the temporal growth rate is. The suppression of the SRS by the external axial magnetic field in the linear region is verified by both 1D and 2D PIC simulations. The simulation results also show that the phase velocity of the electron plasma wave (EPW) will be decreased by the external magnetic field as expected by the theory, which makes it easier for the EPW to trap electrons and for the nonlinear frequency shift of the EPW to happen. As a result, not only the linear growth rate of SRS but also the saturation level of SRS is decreased by the external axial magnetic field. When the external magnetic field is strong enough, the saturation level of SRS can be suppressed by several times. So, this work also provides an efficient way of suppressing the SRS instability. Besides, the 2D simulation results show that some transverse instabilities of the electron plasma wave are also suppressed by the external magnetic field and this suppression will provide us with an electron plasma wave with a better structure, which may benefit the simulated Raman amplification.

The two-plasmon decay (TPD) instability of the backscattered light of the simulated Raman scattering (SRS) is first investigated with the help of one-dimensional (1D) and two-dimensional (2D) particle-in-cell (PIC) simulations. The 2D PIC simulation results show that the backscattered light of the SRS does excite TPD instability. Further, the comparison between 1D and 2D PIC simulation results shows that the backward SRS is suppressed by the excitation of TPD instability, which means the TPD instability of the scattered light could be a saturation scheme of SRS. It is also shown that, after the excitation of TPD, the electrons could experience a two-stage or even three-stage acceleration and their energy may reach to or even exceed 100 KeV, which means the TPD instability of the backscattered light could be a source of super-hot electrons.

Strong coupling stimulated Brillouin backscattering (sc-SBS) in plasma is potentially an efficient method of amplifying laser pulses to reach exawatt powers. Here, we report on a new regime of brillouin-based plasma amplification, producing an amplified pulse with a duration of 5 fs and unfocused intensity of 6 × 1017 W/cm2. The results are obtained from 2D particle-in-cell simulations, using two circularly polarized pump and seed pulse with Gaussian transverse profile, both at an intensity of 2.74 × 1016 W/cm2, counter-propagating in a 0.3nc plasma. The significant compression of amplified seed is achieved as a result of sc-SBS amplification as well as additional compression by the interplay between self-phase modulation and negative group delay dispersion. We show that the amplified seed retains high beam qualities since the filamentation can be prevented due to the fast compression. This scheme may pave the way for few-cycle laser pulses to reach exawatt or even zetawatt regime.

The time response functions of an ultrafast quenched plastic scintillation detector used to measure neutron time of flight spectra were calibrated by utilizing cosmic rays and implosion neutrons from DT-filled capsules at the Shenguang II laser facility. These sources could be regarded as δ function pulses due to their much narrower time widths than those of the time response functions of the detection system. The results showed that the detector responses to DT neutrons and to cosmic rays were 1.18 and 0.96 ns FWHM, respectively.

Backward Raman amplification (BRA) in plasmas has been promoted as a means for generating ultrapowerful laser pulses. For the purpose of achieving the maximum intensities over the shortest distances, an envelope matching between the seed pulse and the amplification gain is required, i.e., the seed pulse propagates at the same velocity with the gain such that the peak of the seed pulse can always enjoy the maximum gain. However, such an envelope matching is absent in traditional BRA because in the latter the amplification gain propagates at superluminous velocity while the seed pulse propagates at the group velocity, which is less than the speed of light. It is shown here that, by using self-ionizing plasmas, the speed of the amplification gain can be well reduced to reach the envelope matching regime. This results in a favorable BRA process, in which higher saturated intensity, shorter interaction length and higher energy-transfer efficiency are achieved.

The fusion fuel ion temperature is a very important implosion dynamics parameter in initial confinement fusion (ICF) research. It is often determined by the neutron time of flight (TOF) technique,1 , 2 which utilizes the neutron Doppler broadening due to the center-of-mass motion of the reacting ions in the fusion fuel. However, the relative yield method has an advantage when the neutron yield is low. On this condition the detector must be placed at a closer distance from the fuel-filled implosion capsule for enough statistical counts, which will cause a narrower neutron TOF width and thus a greater error as the detection system has a certain response time. But it is possible to employ a suitable fuel mixture in which different reactions occur simultaneously although the implosion neutron yield is sensitive not only to the ion temperature alone. Supposing that the final density ratio of the different fusion fuel components in the compressed microballoon core remains constant as compared with the initial one, the relative yield depends only on the ratio of averaged for the different reactions, where σ and v denote the reaction cross section and the relative velocity, respectively. Since the dependence of on temperature is different for different reactions, the relative yield ratio can reflect the ion temperature.3, 4

A guidance and collimation scheme for fast electron beam in a traverse periodic quasi-static electromagnetic field array is proposed with the semi-analytic method and the particle-in-cell simulation. The sheath electric fields on the surfaces of nanowires and the magnetic fields around the nanowires form a traverse periodic quasi-static electromagnetic field array. Therefore, most of the fast electrons are confined at the nanowire surfaces and transport forward. More importantly, due to the divergent property of the beams, the magnitudes of the generated fields decrease with the target depth. The lateral momenta of the electrons convert into the forward momenta through Lorenz force, and they cannot recover their initial values. Therefore, the fast electrons can be guided and collimated efficiently in the gaps between the nanowires. In our particle-in-cell simulations, the observed guiding efficiency exceeds 80% compared with the reference target.

The realization technique of modularization on the long pulse high power generator was expounded. The relation between the ferromagnetic dissipation of the linear transformer and the exciting magnetic current was analyzed. The main influential factors on the wave distortion from the primary pulse to the system output waveforms were discussed. Finally, through engineering experiments, a long pulse power electron beam of up to several gigawatt with repetition frequency of several dozen hertz and rise time of 40—50 ns was obtained. This paper gives a new idea for research on the modularization of the long pulse high power generator with higher repetition rate.

The mechanism of modulation and rf extraction of an S-band on an injection locked, long pulse, relativistic extended interaction cavity oscillator （REICO） is studied in this paper by experiment, theory, and simulation. The origin of pulse shortening and mode competition of the REICO driven by the long pulse intense relativistic electron beams （IREBs） is explored. The modified structure of REICO with coupler is proposed for the first time, which suppresses the pulse shortening and mode competition remarkably. The pulse width of the modulated IREB is increased from 60 ns to 140 ns, and the magnitude of the modulated IREB is increased from 2 kA to 5 kA. The microwave radiation power and efficiency are promoted obviously after adjusting the parameters of the IREB and the leading magnetic field. Using a 580 kV, 48 kA, 210 ns electron beam with 11 T leading magnetic field, more than 400MW radiated power was extracted in 160 ns FWHM pulses at 289 GHz. The power efficiency is 27%, and the energy efficiency is 23%. The radiated power is 400 MW and the FWHM pulse is 184 ns when pulse repetition frequency of IREBs is 20 Hz. The experimental results are well consistent with the simulations. The injection locked REICO generates high power microwave of power 400 MW with pulse width 100 ns.