Advances in laser technology have driven the development of laser-wakefield accelerators, compact devices that are capable of accelerating electrons to GeV energies over centimetre distances by exploiting the strong electric field gradients arising from the interaction of intense laser pulses with an underdense plasma. A side-effect of this acceleration mechanism is the production of high-charge, low-energy electron beams at wide angles. Here we present an experimental and numerical study of the properties of these wide-angle electron beams, and show that they carry off a significant fraction of the energy transferred from the laser to the plasma. These high-charge, wide-angle beams can also cause damage to laser-wakefield accelerators based on capillaries, as well as become source of unwanted bremsstrahlung radiation.
The laser-plasma wakefield accelerator is a novel ultra-compact particle accelerator. A very intense laser pulse focused onto plasma can excites plasma density waves. Electrons surfing these waves can be accelerated to very high energies with unprecedented accelerating gradients in excess of 1 GV/cm. While accelerating, electrons undergo transverse betatron oscillations and emit synchrotron-like x-ray radiation into a narrow on-axis cone, which is enhanced when electrons interact with the electromagnetic field of the laser. In this case, the laser can resonantly drive the electron motion, lading to direct laser acceleration. This occurs when the betatron frequency matches the Doppler down-shifted frequency of the laser. As a consequence, the number of photons emitted is strongly enhanced and the critical photon energy is increases to 100’s of keV.
The ion-channel laser (ICL) has been proposed as an alternative to the free-electron laser (FEL), replacing the deflection of electrons by the periodic magnetic field of an undulator with the periodic betatron motion in an ion channel. Ion channels can be generated by passing dense energetic electron bunches or intense laser pulses through plasma. The ICL has potential to replace FELs based on magnetic undulators, leading to very compact coherent X-ray sources. In particular, coupling the ICL with a laser plasma wakefield accelerator would reduce the size of a coherent light source by several orders of magnitude. An important difference between FEL and ICL is the wavelength of transverse oscillations: In the former it is fixed by the undulator period, whereas in the latter it depends on the betatron amplitude, which therefore has to be treated as variable. Even so, the resulting equations for the ICL are formally similar to those for the FEL with space charge taken into account, so that the well-developed formalism for the FEL can be applied. The amplitude dependence leads to additional requirements compared to the FEL, <i>e.g.</i> a small spread of betatron amplitudes. We shall address these requirements and the resulting practical considerations for realizing an ICL, and give parameters for operation at UV fundamental wavelength, with harmonics extending into X-rays.
Modern accelerators and light sources subject bunches of charged particles to quasiperiodic motion in extremely
high electric fields, under which they may emit a substantial fraction of their energy. To properly describe the
motion of these particle bunches, we require a kinetic theory of radiation reaction. We develop such a theory
based on the notorious Lorentz-Dirac equation, and explore how it reduces to the usual Vlasov theory in the
appropriate limit. As a simple illustration of the theory, we explore the radiative damping of Langmuir waves.
The normalised transverse emittance is a measure of the quality of an electron beam from a particle accelerator. The
brightness, parallelism and focusability are all functions of the emittance. Here we present a high-resolution single shot
method of measuring the transverse emittance of a 125 ± 3 MeV electron beam generated from a laser wakefield
accelerator (LWFA) using a pepper-pot mask. An average normalised emittance of εrms,x,y = 2.2 ± 0.7, 2.3 ± 0.6 π-mmmrad
was measured, which is comparable to that of a conventional linear accelerator. The best measured emittance was
εrms,x,=1.1 ± 0.1 π-mm-mrad, corresponding to the resolution limit of our system. The low emittance indicates that this
accelerator is suitable for driving a compact free electron laser.
The ion-channel laser (ICL) has been proposed as an alternative to the free-electron laser (FEL), replacing the
deflection of electrons in an undulator by betatron oscillations in an ion channel. The aim of this study is to
describe the ICL in terms of the well-developed formalism for the FEL in the steady-state, while taking into
account the dependence of the resonance between oscillations and emitted field on the oscillation amplitude. Numerical
solutions for experimentally relevant parameters show similarities and differences between both devices.
The ICL has potential to replace FELs based on magnetic undulators, leading to very compact coherent X-ray
sources. Furthermore, coupling the ICL with a laser plasma wakefield accelerator would reduce the size of a
coherent light sources by several orders of magnitude.
Electron acceleration using plasma waves driven by ultra-short relativistic intensity laser pulses has
undoubtedly excellent potential for driving a compact light source. However, for a wakefield accelerator to
become a useful and reliable compact accelerator the beam properties need to meet a minimum standard. To
demonstrate the feasibility of a wakefield based radiation source we have reliably produced electron beams
with energies of 82±5 MeV, with 1±0.2% energy spread and 3 mrad r.m.s. divergence using a 0.9 J, 35 fs 800
nm laser. Reproducible beam pointing is essential for transporting the beam along the electron beam line. We
find experimentally that electrons are accelerated close to the laser axis at low plasma densities. However, at
plasma densities in excess of 10<sup>19</sup> cm<sup>-3</sup>, electron beams have an elliptical beam profile with the major axis of
the ellipse rotated with respect to the direction of polarization of the laser.
The Advanced Laser-Plasma High-Energy Accelerators towards X-rays (ALPHA-X) programme is developing laserplasma
accelerators for the production of ultra-short electron bunches with subsequent generation of incoherent radiation
pulses from plasma and coherent short-wavelength radiation pulses from a free-electron laser (FEL). The first
quantitative measurements of the electron energy spectra have been made on the University of Strathclyde ALPHA-X
wakefield acceleration beam line. A high peak power laser pulse (energy 900 mJ, duration 35 fs) is focused into a gas jet
(nozzle length 2 mm) using an F/16 spherical mirror. Electrons from the laser-induced plasma are self-injected into the
accelerating potential of the plasma density wake behind the laser pulse. Electron beams emitted from the plasma have
been imaged downstream using a series of Lanex screens positioned along the beam line axis and the divergence of the
electron beam has been measured to be typically in the range 1-3 mrad. Measurements of the electron energy spectrum,
obtained using the ALPHA-X high resolution magnetic dipole spectrometer, are presented. The maximum central energy
of the monoenergetic beam is 90 MeV and r.m.s. relative energy spreads as low as 0.8% are measured. The mean central
energy is 82 MeV and mean relative energy spread is 1.1%. A theoretical analysis of this unexpectedly high electron
beam quality is presented and the potential impact on the viability of FELs driven by electron beams from laser
wakefield accelerators is examined.
The transverse emittance is an important parameter governing the brightness of an electron beam. Here we
present the first pepper-pot measurement of the transverse emittance for a mono-energetic electron beam from a
laser-plasma wakefield accelerator, carried out on the Advanced Laser-Plasma High Energy Accelerators towards
X-Rays (ALPHA-X) beam line. Mono-energetic electrons are passed through an array of 52 μm diameter holes in
a tungsten mask. The pepper-pot results set an upper limit for the normalised emittance at 5.5 ± 1 π mm mrad
for an 82 MeV beam.