IRUS Total

Development of laser wakefield accelerators

File Description SizeFormat 
Gruse-J-N-2021-PhD-Thesis.pdfThesis27.17 MBAdobe PDFView/Open
Title: Development of laser wakefield accelerators
Authors: Gruse, Jan-Niclas
Item Type: Thesis or dissertation
Abstract: This thesis investigates theoretically and experimentally the wakefield generation, elec- tron acceleration and x-ray emission in laser driven plasma wakefield acceleration. In preparation for a multi-stage laser accelerator with multiple laser, one of the Astra-Gemini laser pulses at 100 TW was used to reflect off a plasma mirror, a 125 μm Kapton foil. Its reflectivity of up to 70% and beam quality was measured. The beam was then self-guide it through a gas cell in preparation for a staged plasma wakefield accel- erator. These were one of the first measurements at such intensities of 4 × 1017 W cm−2 where the beam was injected into a gas cell. To test the usability of this beam to drive an accelerator stage, the guiding efficiency was measured and compared to particle-in- cell simulation. The simulations modelled the propagation of multi-Gaussian low beam quality laser beams at densities 0.25−0.75 × 1018 cm−3. Self-focusing of imperfect laser beams below the critical power is reported and simulation show potential 100s MeV en- ergy increase and a strong argument for beam quality optimisation is made by showing that the potential energy could be increased even further close to 1 GeV. The same Astra-Gemini laser at 150 TW was used to accelerate a bunch of elec- trons in a single gas cell and to optimise betatron radiation. The unique properties of the betatron radiation including a high peak photon flux of 7.5 ± 2.6 × 108 ph mrad−2 and a synchrotron spectrum with critical energy of 14.6 ± 1.3 keV was used in imaging industrial samples. The obtained data was post-processed to remove source based imaging artefacts such as bremsstrahlung hits obscuring the sample. The samples included a topography XCT sample for performance validation, a pouch cell battery and composite cylinder with a kink band failure. The results took advantage of phase- contrast enhancement show-casing its advantage compared to conventional XCT ma- chines. Finally, a machine learning algorithm, based on Bayesian optimisation, was im- plemented on the 5TW Astra laser to optimise electron and x-ray properties with 1Hz repetition rate. This work prepared the diagnostics and extracted the physical quantities used for the Gaussian process regression. It was used to investigate the optimisation process itself, the correlation of parameters for the enhancement of x- ray brightness with ionisation injection, based on N2 doped He-gas. A brilliance of 4.1 ± 1.0 × 1020 ph s−1 mm−2 mrad−2 0.1%BW−1 with a critical energy of 2.7 ± 0.3 keV can be reported at such low power as of 5.6 ± 0.2 TW. Furthermore, the reported energy of the electrons also exceeds similar laser power experiments.
Content Version: Open Access
Issue Date: Dec-2020
Date Awarded: Mar-2021
URI: http://hdl.handle.net/10044/1/88322
DOI: https://doi.org/10.25560/88322
Copyright Statement: Creative Commons Attribution NonCommercial No Derivatives Licence
Supervisor: Najmudin, Zulfikar
Sponsor/Funder: Science and Technology Facilities Council (Great Britain)
John Adams Institute of Accelerator Science
Funder's Grant Number: ST/J002062/1
Department: Physics
Publisher: Imperial College London
Qualification Level: Doctoral
Qualification Name: Doctor of Philosophy (PhD)
Appears in Collections:Physics PhD theses

This item is licensed under a Creative Commons License Creative Commons