Session BO03: Beams: Plasma Wakefield Acceleration (PWFA) and Other Radiation Sources

Live

The AWAKE Run2 experiment at CERN aims~to achieve high-charge bunches of electrons accelerated to high energy (about 10 GeV) while maintaining beam quality and showing that the process is scalable.~ By the end of Run 2 AWAKE should be in the position to use that scheme for first particle physics applications. AWAKE Run 2 consists of four phases: Run 2a starting in 2021 will use the existing AWAKE$^{\mathrm{1}}$ facility to investigate the seeding of the self-modulation with the current electron beam. In Run 2b, a new plasma source with a density step will be implemented in order to maintain strong and constant acceleration fields. In Run 2c a new high energy (150MeV) electron beam system and a second plasma source will be installed to demonstrate the electron acceleration to high energies and keeping good emittance. In Run 2d the second plasma source will be replaced with a scalable one, which could be used for long-distance acceleration for first applications once demonstrated.~An overview of the four phases of AWAKE Run 2 will be given. The technical challenges as well as the proposed schedule will be shown. $^{\mathrm{1}}$E. Gschwendtner, M. Turner et al. (The AWAKE Collaboration), Phil. Trans. R. Soc. A 377: 20180418 (2019).   

Live

The AWAKE experiment accelerates externally injected electrons in plasma wakefields driven by a proton bunch from the CERN SPS.\footnote{AWAKE Collaboration, Nature 561, 363 (2018)} Improvements to the 18~MeV electron beam\footnote{C. Bracco et al., Proceedings of IPAC, 2019}, aimed at achieving parameters required for seeding the self-modulation of a long proton bunch in plasma, are in progress. First, we use standard techniques to improve beam parameters, including control and prediction of position and transverse properties at the plasma entrance\footnote{F. Pe\~{n}a et al., Proceedings of EAAC, 2019}, and to refine models used in predicting wakefields generated by different bunches. Second, we explore model-independent machine learning techniques to automatize and speed up the initial setup process, and to continuously react to external changes.\footnote{F. Velotti et al., paper in preparation}$^{,}$ \footnote{V. Kain et al., paper in preparation}$^{,}$ \footnote{A. Scheinker et al., AIP Advances 10, 055320 (2020)} We will present an overview of the 18~MeV electron beamline as well as our latest beam optimization and automation results.   

Live

The generation of high quality multi-GeV beams using plasma wakefield acceleration (PWFA) has attracted significant interest in applications involving compact particle accelerators and next generation x-ray light sources. Recently, we proposed a new method of injection that relies on reducing the phase velocity of the plasma wake by focusing an electron drive bunch. Two regimes were examined in which the driver was focused by either conventional optics or by the plasma wake. In both regimes, we were able to generate beams with peak normalized brightness as high as $\sim 10^{20}$ A/m$^2$/rad$^2$, projected energy spreads of $<1\%$, and energies up to $\sim 1.86$ GeV for plasma densities of $10^{19}$ cm$^{-3}$. In this talk, we will examine how driver parameters, such as emittance, energy and duration, affect the final energy and current of the injected beam in the regime where plasma self-focusing effects are dominant. Particle-in-cell simulation results using OSIRIS indicate that it may be possible to generate beams with energies of up to $\sim 18.3$ GeV, projected energy spreads of $\sim 0.5\%$, and normalized brightness as high as $\sim 10^{20}$ A/m$^2$/rad$^2$ for plasma densities of $10^{19}$ cm$^{-3}$.   

Live

A high density plasma ($n_{e} \approx 10^{21}m^{-3})$ is needed to achieve Wakefield acceleration of electrons from an axial electric field in the GV/m range. Helicon plasma test cells have been constructed at IPP-Greifswald and UW-Madison to test their viability in this application. It has been shown at the IPP test cell, which has been moved to CERN for further studies, that sufficient densities are transiently achievable during 5 ms pulses. It is still necessary to determine the axial density homogeneity, which is required to stay within 0.25{\%} for Wakefield applications. A laser induced florescence technique which pumps the 668.614 nm argon ion line is being developed to measure the axial density homogeneity on sub-millisecond timescales. The technique involves splitting a single-mode laser into four wavelength shifted beams using acousto-optic modulators. This allows the velocity distribution function to be sampled at four points simultaneously at high temporal resolution. This is a significant advantage over traditional wavelength scanning techniques that can require \textgreater 100s to acquire the velocity distribution function.   

Live

The performance of an electron beam-driven plasma wakefield accelerator (PWFA) depends critically on the density profile of the plasma source. Often, it is advantageous to preform the plasma source ahead of the arrival of the beam (e.g. by laser ionization of a gas) in order to better control the plasma density profile. Due to the geometry and density range of a PWFA plasma source, which comprises a filament less than 1 mm in diameter and up to 1 m in length with a typical core density of 10\textsuperscript{16-17} cm\textsuperscript{-3}, it is challenging to accurately diagnose. Shadowgraphy techniques deployed with laser-driven plasma WFA that have much shorter plasma sources with much higher density cannot easily provide a complete picture of the PWFA plasma source. The most robust diagnostics for the PWFA plasma source rely on the optical properties of the plasma recombination light. However, in order to correctly interpret such signals, the temporal evolution of the plasma filament decay process must be well understood. We present experimental results of a method that combines the signals from shadowgraphy and plasma recombination light to diagnose the density profile of a PWFA plasma source supported by models and simulations describing the PWFA plasma source decay process.   

Live

We study experimentally the effect of a linear plasma density gradient on the self-modulation of a long proton bunch in a dense plasma. The density gradient modifies the phase velocity of the wakefields. It can thus modify the development of the self-modulation process through the time or distance along the plasma that protons spend in the focusing or defocusing phase of the wakefields. This could change the number and the charge of micro-bunches that result from the periodic action of the wakefields. This effect varies along the bunch and along the plasma and is observed after the saturation of the process. Plasma density changing along the proton bunch path can also change the modulation period of the bunch. Experimental results obtained in the AWAKE experiment\footnote{P. Muggli et al. (AWAKE Collaboration), Plasma Physics and Controlled Fusion, 60(1) 014046 (2017)} will be presented\footnote{F. Braunmueller, T. Nechaeva et al. (AWAKE Collaboration), to be submitted}.   

Live

Several techniques for cooling light ion beams at high energies rely on amplification or conversion of the density or momentum signature that individual ions imprint on a co-propagating electron beam. The magnitude of such ion-induced density modulation is several orders of magnitude smaller than the local number density in the background electron beam, the energy modulation being similarly subtle, and can be prohibitively expensive to resolve in regular particle-in-cell (PIC) simulations against the backdrop of the discreteness noise in the macroparticle distribution. We implemented, in the open-source plasma and beam physics code Warp, a well known $\delta f$ PIC technique that offers a solution to this problem by treating the background distribution as a continuous function and representing only the perturbation by variable-weight macroparticles. We use the new capability to simulate the electron momentum signature associated with Debye shielding of co-propagating ions and its subsequent amplification by external magnetic fields.   

Live

A method is presented that accurately simulates steady-state field emission and space-charge-limiting effects in nanoscale vacuum channel transistors (NVCTs). NVCTs consist of a tip (radius ~10nm) that field-emits electrons into a large voltage gap (~1mm). A significant challenge of simulating NVCTs is the need to resolve vastly different length scales, from the 10nm tip to the 1mm gap. One approach to this problem would be use of a variable grid. However, this approach is complicated and the speed of particle pushing is severely limited by the crossing time of the smallest grid scale. A regular grid is preferable for simplicity and speed so is used for this method. Instead of simulating the entire NVCT, this method breaks the simulation into two linked finite difference electrostatic simulations with different grid scales. Simulation A focuses on the dynamics of field emission near the tip, while Simulation B resolves the space-charge-limiting effects. The simulations are linked by injecting emitted electrons from A into B and imposing a Dirichlet boundary condition on A determined by B. Successive iteration between the two simulations reaches a solution. This talk will discuss the advantages and disadvantages of this method as well as present an analysis of the NVCT.   

Live

A fundamental theoretical study of Brillouin flow has been applied to the design of a magnetically insulated line oscillator (MILO) for operation on the Michigan Electron Long Beam Accelerator (MELBA). MELBA applies -300 to -500 kV and up to 10's of kA for 0.3-1.0 $\mu $s. Simulations in CST-Particle Studio have been used to corroborate the theoretical predictions, and preliminary experiments on MELBA will be discussed. CST-PS has also been applied to gain understanding of a GW-class MILO for which experiments are planned to take place at UM [1]. [1] Packard et al, ``HFSS and CST Simulations of a GW-Class MILO'', IEEE T-PS, vol. 48, 1894, (2020).   

Live

Reducing the size, power needs and cost of accelerators opens new opportunities in mass spectrometry, ion implantation and ultimately plasma heating for fusion. Our technology is based on wafer-based components (silicon or circuit boards) where beam transport is in the direction of the surface normal to the wafer. This allows stacking of wafers to increase beam energy while limiting the peak voltage to several kilovolts. The wafer-based implementation allows us to operate multiple ions beams on a single wafer in parallel for much increased current densities per wafer in a multi-beamlet arrangement compared to a single beam with one large aperture. We will report the experimental results of scaling up to mA of beam current using an array of 112 beamlets, and an average energy gain of 8 keV per acceleration gap. We will also discuss the effort of building a compact accelerator to achieve beyond 100's keV beam energy.   

Live

Multipactor is a discharge phenomenon that occurs in RF vacuum electronics and transmission lines. It is of particular concern in modern satellite communication. In coaxial systems, theoretical studies are far more limited, in part because of the complexity in the electron orbits. Several previous experiments [1, 2], have been performed in coaxial transmission lines at low frequencies (10s-100s MHz), but there exist very little experimental data at higher frequencies. We will present experimental results based on our previous simulations [3] on multipactor in a coaxial geometry at multi-GHz frequencies. Diagnostics indicate an increase in RF attenuation due to the multipactor discharge as well as a coincident increase in electron multiplier signals. We also examine the timing characteristics of the onset of multipactor relative to the start of the microwave pulse. [1] Woo R., J. Appl. Phys. 39 1528, (1968). [2] Graves T., doctoral dissertation, MIT, 2006. [3] S. V. Langellotti, et al, IEEE Trans. Plasma Sci. 48, 1942 (2020).   

Live

Numerical simulation results\footnote{A. Caldwell and K.V. Lotov, Phys. Plasmas 18, 103101 (2011)} suggest that placing a positive plasma density step along the self-modulation process of a long proton bunch leads to wakefields that remain at near saturation values over long plasma distances, instead of rapidly decaying after saturation. % We describe the development of a plasma source based on a laser-ionized rubidium vapor\footnote{E. Oz, P. Muggli, NIMA 740(11), 197 (2014)} that allows for imposing a density step in the 0 to 10\% range at different locations, every 50\,cm over the first 4\,m. % This source will be used to optimize the effect of the step on wakefields' amplitude at the end of the 10\,m-long plasma. % The source thus includes ports for a THz plasma density diagnostic\footnote{A. Gopal, private communication} to directly image the plasma electron density perturbation that sustains wakefields. % The source design and parameters for the AWAKE experiment\footnote{P. Muggli (AWAKE Collaboration), accepted for publication in J. of Phys.: Conference Series (JPCS) arXiv:1911.07534 [physics.acc-ph]}   

On Demand

Originally derived exactly for planar diodes [1], more recent analysis of space-charge-limited current (SCLC) derived exact, closed form solutions for general rectilinear geometries using variational calculus (VC) [2]. However, deriving rigorous solutions for SCLC for curvilinear diodes remains challenging. This talk applies conformal mapping (CM) to derive exact, closed solutions for multiple curvilinear geometries [3] from first principles. Starting from the known classical CL law for the planar diodes, we applied CM to map several geometries onto the planar geometry to derive SCLC in such geometries. This provides a means to rapidly screen SCLC for more realistic geometries to provide targeted parameters prior to more detailed simulations. Since this method is based on the geometry of the diodes, any modification to the classical CL law, such as relativistic or quantum modifications, may ultimately be incorporated. [1] P. Zhang, A. Valfells, L. K. Ang, J. W. Luginsland, and Y. Y. Lau, Appl. Phys. Rev. 4, 011304 (2017). [2] A. M. Darr, A. M. Loveless, and A. L. Garner, Appl. Phys. Lett. 114, 014103 (2019). [3] H. F. Ivey, J. Appl. Phys. 24, 1466 (1953).   

Beam Loading Scenarios with Electrons for PWFA LC in the Blowout Regime

We examine realistic beam loading scenarios for PWFA LC based on stages or an afterburner. We consider loading an electron beam into the wake created by a an electron beam driver. For collider applications the witness beam will have .1 to 1 nC of charge and normalized emittances of $\sim 100$ nm. This leads to matched spot sizes $\sim 100$ nm and witness beam densities $10^5$ times the background density. For such parameters it may be important to consider ion motion. We use QuickPIC and QPAD simulations and theory to investigate the effects of asymmetries in the wakes and the effects of ion motion both in isolation and together. Preliminary results will be presented on the energy spread and emittance preservation, and efficiency of from the drive beam to the witness beam for energies of interest to a collider.   

A Study of Miram Curves in Thermionic Cathodes

The anode current vs. cathode temperature plot of a diode using a thermionic cathode is commonly known as the Miram [1] curve, and the transition from temperature to space charge limited flow is referred to as the `knee' in the curve. The physical reasons behind the shape of the knee are significant because a thermionic cathode is almost always operated in the vicinity of the knee to improve cathode life. This paper presents a novel analytic model, which solves the Poisson equation in 3D assuming an infinite axial magnetic field, including an arbitrary work function distribution on the cathode surface [2]. An earlier version that solves the Poisson equation in 2D yields excellent agreement with the corresponding MICHELLE code results [3]. Our model points to the deficiency of the customary Practical Work Function Distribution model [1], and demonstrates the necessity of a large fraction of non-emitting area to yield a Miram curve with a smooth knee, as often observed in experiments [1]. [1] M. Cattelino, G. Miram. Appl. Surf. Sci., vol. 111, pp. 90-95, 1997 [2] A. Jassem, D. Chernin, S. Ovtchinnikov, J. J. Petillo, Y. Y. Lau. Proc. IEEE Int. Vac. Electron Conf. (IVEC), Monterey, CA, USA, 2020. [3] D. Chernin et al. IEEE Trans. Plasma Sci., vol. 48, no. 1, pp. 146-155, 2020.