We present the concept and design of an efficient (similar to 56% overall efficiency) megawatt (MW) coherent Cherenkov radiation source operating at 94GHz combined with a 4-stage depressed collector. The source has a highly oversized interaction region, enabling MW continuous wave radiation. Mode selection, achieved via high-order mode coupling, is mediated by a 2D periodic surface lattice (PSL) interaction cavity. The overall device efficiency is competitive with gyrotrons in the requirements for heating and current drive in fusion plasma, with further application in plasma turbulence diagnostics, biochemical spectroscopy, radar and remote sensing. These sources are scalable; both in transverse size and frequency, with strong potential to address the long-standing terahertz gap. The presented energy recovery technique has broad applicability to many electron-beam-driven devices.

The need to bridge the THz gap is stimulated by a growing number of important applications including biochemical spectroscopy, plasma turbulence diagnostics and drive sources for tokamaks. Cherenkov sources based on two-dimensional (2D) corrugated surface lattice interaction structures, in which the diameter is several times greater than the radiation wavelength, hold strong promise to bridge the THz gap. In this paper, we demonstrate the ability to drive these sources, typically intended for steady-state operation, into the highly non-linear superradiant regime. We demonstrate, for the first time, the ability to generate superradiant pulses, for which the peak power scales as the number of electrons in the bunch squared, by exploiting slippage of an electron beam through subluminal surface waves close to the metallic 2D corrugation. The surface waves are scattered into low order, forward propagating TM 0, N modes which form the emitted 'super pulse'. To drive superradiance, the nanosecond electron bunch must have a fast rising edge with a suitably high (kiloamperes) electron current. Superradiant pulses have been simulated for cases where the relative difference between the group speed of the electromagnetic wave and the drift speed of the electron beam is in the correct range. We show that, for this transient process, the diameter-to-wavelength ratio of the interaction cavities can be scaled from 6 to 9, with a corresponding uplift in peak power from 450 MW to 750 MW, demonstrating the potential for exceptionally powerful THz pulses. The presented results have been obtained for a Cherenkov maser operating in the 83-94 GHz range. However, numerical dispersion analysis, validated by full-wave simulations, shows the potential to radically modify the wave dispersion by varying the 2D lattice geometry for highly controllable, powerful signals at any frequency from 1 GHz to 1 THz. Based on these results, the capability to eventually generate gigawatt-level pulses in the THz range can be projected.

We present the theory, concept and design of an efficient, megawatt coherent Cherenkov radiation source based on a two-dimensional periodic surface lattice (2D-PSL) cavity combined with a novel energy recovery system for the generation of highly efficient (> 50%) single-frequency radiation. We demonstrate the scalability of the transverse dimension of the 2D-PSL cavity of the Cherenkov source and thus the potential for efficient, continuous-wave, high-power (> 1 MW) operation; fundamental to the eventual realization of clean, fusion energy. These new sources, with the capacity to operate in the 0.1-10THz range, hold strong promise to address the long-standing "Terahertz gap". By combining a Cherenkov oscillator driven by a non-gyrating beam with an innovative four-stage depressed collector energy recovery system, the overall device efficiency can be increased to be competitive with gyrotrons in the requirements for heating and current drive in fusion plasma. In these Cherenkov devices, the frequency independence of the magnetic guide field enables advantageous frequency scaling without deployment constraints, making them especially attractive for high-impact applications in fusion science, turbulence diagnostics, non-destructive testing and biochemical spectroscopy. The novel energy recovery techniques presented in this paper have broad applicability to many electron-beam driven devices, bringing revolutionary potential to future THz source technologies.

Parametric instabilities [1] play an important role in understanding how electromagnetic power is absorbed by a plasma and are encountered across a wide range of environments: magnetic confinement fusion [2], inertial confinement fusion [3] and in the ionosphere [4]. The traditional tokamak has enjoyed success with electron cyclotron current drive; spherical tokamaks present a challenge since they are over-dense at low harmonics of the cyclotron frequency, impeding efficiency or limiting potential fusion performance. This forces the heating system to access the plasma using higher harmonics, reducing the efficiency. An alternate heating scheme could involve beat-wave type couplings like those found in parametric scattering processes; these could be used to resonantly couple the energy from injected electromagnetic waves to a plasma oscillation.

The helically corrugated interaction region (HCIR) is able to produce an almost ideal dispersion for microwave oscillators/amplifiers. The HCIR is studied and presented in this paper with particular application in a W-band broadband, kW-level gyrotron traveling wave amplifier (gyro-TWA). The dispersion of the waveguide was calculated, and measurements of a manufactured HCIR showed good agreement with predictions. Numerical simulations of the beam-wave interaction in different circuits showed stable output can be achieved in two gain regions separated by a sever (with purposeful reflections at each end of the circuit). The demonstration of a gyro-TWA based on the HCIR is described with a 3-dB bandwidth of over 5.5 GHz with more than 3.4 kW and the gain was around 36–38 dB. The work presented in this paper will widen the capability of the gyro-TWA from single pulse to high PRF operation.

High-power microwave sources are typically relativistic in nature, employing multi-kilo-ampere electron beams that require significant magnetic confinement for efficient operation. As the desired output power increases, so does the complexity, and overall energy requirements, of the source. It can, therefore, be advantageous to consider the use of several, moderate-power, sources operating as a phased array; for an array of ${N}$ sources, the far-field peak intensity scales as ${N}^{{\,{2}}}$ , and the peak-of-field may be steered electronically by varying the relative phases of the different output signals. In this article, we present the numerical analysis of a short-pulse ( $\sim 1$ ns) ${X}$ -band backward-wave oscillator, driven by a 210-keV, 1.4-kA electron beam, suitable for use as the radiative element in such an array. The investigation of the required magnetic confinement showed two peaks in performance, with the highest efficiency, of 43%, predicted at the low magnetic confinement peak at 0.3 T, corresponding to 125-MW peak output power. The magnitude and timing of the peak in the output pulse were functions of the rise time of the electron beam energy, with longer rise times resulting in delayed peak-of-field and lower peak output power. When operating in an array, to maintain effective output in the region of ${N}^{{\,{2}}}$ , it was determined that the beam rise times, across all sources, should be ≤150 ps with the adjustment of the relative timing between output’s being ±30 ps.

Parametric instabilities [1] arise across many plasma physics environments including inertial [2] and magnetic [3] confinement fusion and in ionospheric plasmas [4] . A helicon [5] plasma source can provide a stable, controllable, low temperature and relatively tenuous plasma for periods of hours, making it suited to diagnosis by techniques such as probes and interferometry. Our apparatus operates in inductive and helicon modes driven by an RF antenna in the range of 3<f<30 MHz at powers <4 kW, which creates a noble gas plasma in a 1 m diameter, 3 m long stainless-steel vacuum chamber surrounded by electromagnets providing an axial B0<90 mT. Typically, a source like this would be expected to provide a plasma with 10 15 <n e <10 18 m -3 and T e <10 eV. For these plasma parameters, parametric instabilities such as Raman and Brillouin scattering can be triggered by readily available, powerful microwave sources such as magnetron oscillators and travelling-wave tube amplifiers and broadband pulses developed in dispersive pulse compressors. These sources will launch microwaves into the plasma via a pair of Satoh horns providing a Gaussian beam for microwave scattering experiments. The characterisation of the plasma using the RF-compensated Langmuir probes and complementary microwave interferometry diagnostic is presented.

The gyrotron-traveling wave amplifier (gyro-TWA), operating in W-band has previously been shown to provide an unsaturated output power of 3.4 kW over a 3-dB bandwidth of 91-96.5 GHz in single pulse (350 ns) operation. The output section of the gyro-TWA was critical to maintaining the amplifier stability due to the 37 dB gain from a single gain section. This paper presents the re-design of the gyro-TWA output section suitable for several percent duty cycle and wideband operation over 90-100 GHz. We describe the design of the: wideband smooth output feedhorn, which couples to a fundamental Gaussian mode; the electron beam collector (forming part of the output horn), and the vacuum window. The feedhorn simulations predict a calculated >98% Gaussian efficiency and better than -30 dB return loss. The vacuum window is a multilayer design comprising three window discs: two quartz and one alumina. The predicted return loss is lower than -30 dB over the operating range.

The numerical study of a low confining-field (~0.3T) X-band Backward Wave Oscillator (BWO), suitable for implementation as the microwave source(s) in a phase-synchronous array, is presented. Analytical theory and CST Particle Studio simulations have been used to design a ~9.6GHz BWO, generating 120MW, 1ns, pulses when driven by a 4ns, 210kV, 1.5kA Radan 303B modulator.

Many important plasma processes, such as heating and current drive in tokamaks and laser-plasma interactions, involve the interaction of an externally generated electromagnetic wave with the plasma. The plasma can couple the incident wave to a range of other waves. Important examples include Raman scattering, where the incident wave couples to an electrostatic Langmuir wave and a scattered electromagnetic wave, and Brillouin scattering, where the Langmuir wave is replaced by an ion sound wave 1 . Both are observed in laser plasma experiments; however, these instabilities can also be generated by intense, short pulse microwave signals in a cool, tenuous plasma. A magnetized plasma supports additional beat wave interactions with natural cyclotron motions of charged particles and with magnetized plasma waves such as upper/lower hybrid waves as well as kinetic modes such as ion/ electron Bernstein waves. Applications include heating plasma that is inaccessible at low harmonics of the cyclotron frequency in high density fusion plasmas. Enabling these experiments is a large helicon apparatus, 1 m diameter, 3m length. The stainless-steel vacuum vessel is surrounded by several electromagnets providing a B 0 up to 0.0875 T. The plasma will be ionized by m = 0 helicon waves with 3 < f < 30 MHz launched by a flat spiral antenna 2 . Plasmas of 10 15 < n e < 10 18 m -3 and T e < 10 eV will be produced in a low pressure (<100 mTorr) noble gas. R.F. compensated Langmuir probes and line integrated interferometry will be used to diagnose the plasma. These measurements will be carried out alongside numerical simulations and theoretical analysis to enhance the understanding of these interactions.

This paper presents a numerical investigation of alternate gyrotron Traveling Wave Amplifier (gyro-TWA) configurations for increased gain. Previous experimental results showed a maximum stable gain of 37 dB over the frequency range 90-96 GHz. These results were limited in bandwidth by the input power from the solid-state pre-amplifier. The simplest way to mitigate the limited input power available is to increase amplifier gain. Doing so without inducing instability requires the introduction of losses to the RF circuit. We present PIC simulations contrasting a range of new designs combining two gain stages and lossy elements with results for the single section design presented in our earlier experimental program.

Helically corrugated waveguide can be used as a dispersive medium to compress long duration chirping signals into higher power shorter duration microwave pulses 1 , 2 , 3 . It has the advantage over smooth bore circular waveguide in that its dispersion can be shaped to have a large change in group velocity over a relatively small frequency range, but still be far from ‘cut-off’ where there is a high risk of reflection.

Electromagnetic waves are vital to the introduction of energy in many plasma applications. As a non-linear medium, plasma can enable complex interactions between multiple electromagnetic waves. Examples in laser-plasma physics include Raman coupling via a Langmuir oscillation or Brillouin scattering mediated by ion-acoustic waves. Signals with normalised intensity approaching that of recent laser plasma interaction experiments can be generated using powerful and flexible microwave sources, interacting in relatively tenuous, cool and accessible plasma. Intense short pulse signals can be generated by dispersive pulse compressors driven by chirped microwave signals. Other multi-wave interactions are interesting for magnetic confinement fusion plasmas, e.g. beat-wave interactions coupled to cyclotron motion of the ions and electrons or the hybrid oscillations may be useful in heating plasma or for driving currents.

We present initial predictions for the performance of a novel High-Power Microwave (HPM), X-band, Cherenkov oscillator, termed the 'Self-Insulating Backward-Wave Oscillator (SIBWO)‘ due to its close resemblance, in appearance, to the conventional (magnetically insulated) BWO. Conversion efficiencies (i.e. energy extracted from the electron beam, to the electromagnetic wave) in excess of 30% have been predicted by simulations using CST Studio Suite. This corresponds to output powers in excess of 300MW, from a 500keV, 2kA beam, when operating in the TM 01 mode at ~9.4GHz.

We present initial predictions from the investigation of a novel, short-pulse (~200ns), space-charge-limited-emission electron accelerating diode, or electron-gun, that operates purely under the influence of its own self-fields - i.e. without the requirement for externally applied magnetic insulation of the propagating electron beam. The optimised model, developed using CST Studio Suite, predicted a ~500keV, ~1.86kA electron beam, with a mean envelope-radius of 16.8mm propagating into a 18mm radius drift-tube. The beam was very slightly convergent at the entry to the drift-region with a convergence angle of 1.5° (0.026rad).

As a non-linear medium, plasma can exhibit diverse dynamics when excited bymultiple EM waves. Electromagnetic waves are vital to the introduction of energyin laser plasma interactions and the heating of magnetically confined fusion reactors.In laser plasma applications Raman coupling via a Langmuir oscillation or Brillouinscattering mediated by ion-acoustic waves are of interest. Signals with normalisedintensities approaching those used in some recent laser plasma interactions can begenerated using powerful and flexible microwave amplifiers, interacting in relativelytenuous, cool and accessible plasma. Other multi-wave interactions are interesting formagnetic confinement fusion plasmas, for example beat-wave interactions betweentwo microwave signals coupling to cyclotron motion of the ions and electrons or thelower hybrid oscillations may be useful in heating the plasmas or for driving currents.A linear plasma experiment is being built to test such multifrequency microwaveinteraction in plasma, based on prior research on geophysical cyclotron wave emissionand propagation [1,2]. The main section of the plasma will be magnetised at up to0.05T, with the plasma created by an RF helicon source to generate a dense, large,cool plasma with a high ionisation fraction. A range of frequency-flexible sources willprovide microwave beams to enable multi-wave coupling experiments. The paper willpresent progress on this apparatus and experiments.The authors gratefully acknowledge support from the EPSRC, MBDA UK Ltd andTMD Technologies Ltd.[1] Ronald K., Speirs D.C., McConville S.L., Phelps A.D.R., Robertson C.W., WhyteC.G., He W., Gillespie K.M., Cross A.W., Bingham R., 2008, Phys. Plasmas, 15,art.056503[2] Speirs, D.C., Bingham, R., Cairns, R.A., Vorgul, I., Kellett, B.J., Phelps, A.D.R.,Ronald, K, 2014, Phys. Rev. Lett., 113, art 155002