Table of contents

This special issue on laser and plasma accelerators illustrates the rapid advancement and diverse applications of laser and plasma accelerators. Plasma is an attractive medium for particle acceleration because of the high electric field it can sustain, with studies of acceleration processes remaining one of the most important areas of research in both laboratory and astrophysical plasmas. The rapid advance in laser and accelerator technology has led to the development of terawatt and petawatt laser systems with ultra-high intensities and short sub-picosecond pulses, which are used to generate wakefields in plasma. Recent successes include the demonstration by several groups in 2004 of quasi-monoenergetic electron beams by wakefields in the bubble regime with the GeV energy barrier being reached in 2006, and the energy doubling of the SLAC high-energy electron beam from 42 to 85 GeV. The electron beams generated by the laser plasma driven wakefields have good spatial quality with energies ranging from MeV to GeV. A unique feature is that they are ultra-short bunches with simulations showing that they can be as short as a few femtoseconds with low-energy spread, making these beams ideal for a variety of applications ranging from novel high-brightness radiation sources for medicine, material science and ultrafast time-resolved radiobiology or chemistry.

Laser driven ion acceleration experiments have also made significant advances over the last few years with applications in laser fusion, nuclear physics and medicine. Attention is focused on the possibility of producing quasi-mono-energetic ions with energies ranging from hundreds of MeV to GeV per nucleon. New acceleration mechanisms are being studied, including ion acceleration from ultra-thin foils and direct laser acceleration.

The application of wakefields or beat waves in other areas of science such as astrophysics and particle physics is beginning to take off, such as the study of cosmic accelerators considered by Chen et al where the driver, instead of being a laser, is a whistler wave known as the magnetowave plasma accelerator. The application to electron--positron plasmas that are found around pulsars is studied in the paper by Shukla, and to muon acceleration by Peano et al.

Electron wakefield experiments are now concentrating on control and optimisation of high-quality beams that can be used as drivers for novel radiation sources. Studies by Thomas et al show that filamentation has a deleterious effect on the production of high quality mono-energetic electron beams and is caused by non-optimal choice of focusing geometry and/or electron density. It is crucial to match the focusing with the right plasma parameters and new types of plasma channels are being developed, such as the magnetically controlled plasma waveguide reported by Froula et al. The magnetic field provides a pressure profile shaping the channel to match the guiding conditions of the incident laser, resulting in predicted electron energies of 3GeV. In the forced laser-wakefield experiment Fang et al show that pump depletion reduces or inhibits the acceleration of electrons. One of the earlier laser acceleration concepts known as the beat wave may be revived due to the work by Kalmykov et al who report on all-optical control of nonlinear focusing of laser beams, allowing for stable propagation over several Rayleigh lengths with pre-injected electrons accelerated beyond 100 MeV.

With the increasing number of petawatt lasers, attention is being focused on different acceleration regimes such as stochastic acceleration by counterpropagating laser pulses, the relativistic mirror, or the snow-plough effect leading to single-step acceleration reported by Mendonca. During wakefield acceleration the leading edge of the pulse undergoes frequency downshifting and head erosion as the laser energy is transferred to the wake while the trailing edge of the laser pulse undergoes frequency up-shift. This is commonly known as photon deceleration and acceleration and is the result of a modulational instability. Simulations reported by Trines et al using a photon-in-cell code or wave kinetic code agree extremely well with experimental observation.

Ion acceleration is actively studied; for example the papers by Robinson, Macchi, Marita and Tripathi all discuss different types of acceleration mechanisms from direct laser acceleration, Coulombic explosion and double layers. Ion acceleration is an exciting development that may have great promise in oncology. The surprising application is in muon acceleration, demonstrated by Peano et al who show that counterpropagating laser beams with variable frequencies drive a beat structure with variable phase velocity, leading to particle trapping and acceleration with possible application to a future muon collider and neutrino factory.

Laser and plasma accelerators remain one of the exciting areas of plasma physics with applications in many areas of science ranging from laser fusion, novel high-brightness radiation sources, particle physics and medicine.

The guest editor would like to thank all authors and referees for their invaluable contributions to this special issue.

The laser acceleration of proton beams with quasi-monoenergetic features in the energy spectra from microdot targets is investigated by numerical simulation. The formation of these spectral peaks is strongly dependent on the interplay between different ion species in the target. The scaling of the spectral peak's energy, and number of protons in the spectral peak, with both microdot composition and laser intensity is considered. Particular attention is given to determining the proton concentration below which the number of protons in the spectral peak rapidly diminishes. It is shown that at proton concentrations of 1-5n_crit a spectral peak is produced that reaches an energy up to 70% of the maximum proton energy, whilst still containing more protons than would be produced by a conventional target in this energy range.

We investigate theoretically and with three-dimensional particle-in-cell simulations high-quality laser proton acceleration for oblique incidence of the high intensity laser pulse on a double-layer target. The double-layer target is composed of a high-Z ion layer coated by a thin and narrow hydrogen patch. The highest proton energy gain is achieved at a certain incidence angle at which the fast proton maximum energy is much greater than the case of normal incidence. The fast protons form a tilted bunch which propagates at some angle with respect to the normal of the target surface, as determined by the proton energy and the incidence angle.

In a forced laser-wakefield accelerator experiment (Malka et al 2002 Science298 1596) where the length of the pump laser pulse is a few plasma periods long, the leading edge of the laser pulse undergoes frequency downshifting and head erosion as the laser energy is transferred to the wake. Therefore, after some propagation distance, the group velocity of the leading edge of the pump pulse—and thus of the driven electron plasma wave—will slow down. This can have implications for the dephasing length of the accelerated electrons and therefore needs to be understood experimentally. We have carried out an experimental investigation where we have measured the velocity v_f of the 'wave-front' of the plasma wave driven by a nominally 50 fs (full width half maximum), intense (a₀ ≃ 1), 0.815 µm laser pulse. To determine the speed of the wave front, time- and space-resolved refractometry, interferometry and Thomson scattering were used. Although a laser pulse propagating through a relatively low-density plasma (n_e = 1.3 × 10¹⁹ cm⁻³) showed no measurable changes in v_f over 1.3 mm (and no accelerated electrons), a high-density plasma (n_e = 5 × 10¹⁹ cm⁻³) generated accelerated electrons and showed a continuous change in v_f as the laser pulse propagated through the plasma. Possible causes and consequences of the observed v_f evolution are discussed.

The problem of the 'hole-boring' (HB)-type of radiation pressure acceleration of ions by circularly polarized laser pulses interacting with overdense plasmas is considered in the regime where the dimensionless scaling parameter I/ρc³ becomes large. In this regime a non-relativistic treatment of the 'HB' problem is no longer adequate. A new set of fully relativistic formulae for the mean ion energy and 'HB' velocity is derived and validated against one-dimensional particle-in-cell simulations. It is also found that the finite acceleration time of the ions results in large energy spreads in the accelerated ion beam even under the highly idealized conditions of constant laser intensity and uniform mass density.

The dynamics of electric field generation and radial acceleration of ions by a laser pulse of relativistic intensity propagating in an underdense plasma has been investigated using a one-dimensional electrostatic, ponderomotive model developed to interpret experimental measurements of electric fields (Kar S et al 2007 New J. Phys.9 402). Ions are spatially focused at the edge of the charge-displacement channel, leading to hydrodynamical breaking, which in turn causes the heating of electrons and an 'echo' effect in the electric field. The onset of complete electron depletion in the central region of the channel leads to a smooth transition to a 'Coulomb explosion' regime and a saturation of ion acceleration.

A scheme for fast, compact and controllable acceleration of heavy particles in vacuum has been recently proposed (Peano F et al 2008 New J. Phys.10 033028), wherein two counterpropagating laser beams with variable frequencies drive a beat-wave structure with variable phase velocity, leading to particle trapping and acceleration. The technique allows for fine control over the energy distribution and the total charge of the accelerated beam to be obtained via tuning of the frequency variation. Here, the theoretical bases of the acceleration scheme are described, and the possibility of applications to ultrafast muon acceleration and to the prompt extraction of cold-muon beams is discussed.

We discuss on purely analytical grounds the various mechanisms for electron acceleration, which can be relevant in the Petawatt laser pulse regime. This includes a modified version of the well-known wakefield acceleration (dominant for short pulses), stochastic acceleration due to counter-propagating laser pulses (which could be dominant for long pulses) and the relativistic photon mirror or the snow-plow effect (which could lead to enhanced single step acceleration). We discuss other stochastic processes related to the interaction between direct and wakefield acceleration. Approximate analytical criteria for stochastic and enhanced acceleration regimes are established. Our discussion also includes excitation of unstable betatron oscillations. The present analytical models could be relevant to the interpretation of both simulations and experiments.

The modulational instability that occurs during the interaction of a long laser pulse and its own wakefield in an underdense plasma has been studied experimentally and theoretically. Recent experiments using laser pulses that are several times longer than the wakefield period have yielded transmission spectra that exhibit a series of secondary peaks flanking the main laser peak. These peaks are too closely spaced to be the result of Raman instabilities; their origin was found to be photon acceleration of the laser's photons in the wakefield instead. In the experiments described in this paper, a laser pulse of 50–200 fs containing 300–600 mJ was focused on the edge of a helium gas jet on a 25 µm focal spot. The observed transmission spectra show evidence of both ionization blueshift and modulation by the pulse's wakefield. The transmission spectra have also been modelled using a dedicated photon-kinetic numerical code. The modelling has revealed a direct correlation between the spectral modulations and the amplitude of the excited wakefield. By comparing the measured and simulating spectra, the origin of various spectral characteristics has been explained in terms of photon acceleration. The feasibility of using this effect as a wakefield diagnostic will be discussed.

An external magnetic field applied to a laser plasma is shown to produce a plasma channel at densities relevant to creating GeV monoenergetic electrons through laser wakefield acceleration. Furthermore, the magnetic field also provides a pressure to help shape the channel to match the guiding conditions of an incident laser beam. Measured density channels suitable for guiding relativistic short-pulse laser beams are presented with a minimum density of 5 × 10¹⁷ cm⁻³, which corresponds to a linear dephasing length of several centimeters suitable for multi-GeV electron acceleration. The experimental setup at the Jupiter Laser Facility, Lawrence Livermore National Laboratory, where a 1 ns, 150 J, 1054 nm laser will produce a magnetically controlled channel to guide a < 75 fs, 10 J short-pulse laser beam through 5 cm of 5 × 10¹⁷ cm⁻³ plasma is presented. Calculations presented show that electrons can be accelerated to 3 GeV with this system. Three-dimensional resistive magneto-hydrodynamic simulations are used to design the laser and plasma parameters, and quasi-static kinetic simulations indicate that the channel will guide a 200 TW laser beam over 5 cm.

In the experiments reported here, the filamentation of ultrashort laser pulses, due to non-optimal choice of focusing geometry and/or electron number density, has a severely deleterious effect on monoenergetic electron beam production in laser wakefield accelerators. Interactions with relatively small focal spots, w₀ < λ_p/2, and with pulse length cτ ≈ λ_p, incur fragmentation into a large number of low power filaments. These filaments are modulated with a density dependent size of, on average, close to λ_p. The break-up of the driving pulse results in shorter interaction lengths, compared with larger focal spots, and broad energy-spread electron beams, which are not useful for applications. Filamentation of the pulse occurs because the strongly dynamic focusing (small f-number) of the laser prevents pulse length compression before reaching its minimum spot-size, which results in non-spherical focusing gradients.

Nonlinear focusing of a bi-color laser in plasma can be controlled by varying the difference frequency Ω. The driven electron density perturbation forms a co-moving periodic focusing (de-focusing) channel if Ω is below (above) the electron Langmuir frequency ω_p. Hence, the beam focusing is enhanced for Ω < ω_p and is suppressed otherwise. In particular, a catastrophic relativistic self-focusing of a high-power laser beam can be prevented all-optically by a second, much weaker, co-propagating beam shifted in frequency by Ω > ω_p. A bi-envelope equation describing the early stage of the mutual de-focusing is derived and analyzed. Later stages, characterized by a well-developed electromagnetic cascade, are investigated numerically. Stable propagation of the over-critical laser pulse over several Rayleigh lengths is predicted. The non-resonant plasma beat wave (Ω ≠ ω_p) can accelerate pre-injected electrons above 100 MeV with low energy spread.

We present a new concept for a plasma wakefield accelerator driven by magnetowaves (MPWA). This concept was originally proposed as a viable mechanism for the 'cosmic accelerator' that would accelerate cosmic particles to ultra-high energies in the astrophysical setting. Unlike the more familiar plasma wakefield accelerator (PWFA) and the laser wakefield accelerator (LWFA) where the drivers, the charged-particle beam and the laser, are independently existing entities, MPWA invokes the high-frequency and high-speed whistler mode as the driver, which is a medium wave that cannot exist outside of the plasma. Aside from the difference in drivers, the underlying mechanism that excites the plasma wakefield via the ponderomotive potential is common. Our computer simulations show that under appropriate conditions, the plasma wakefield maintains very high coherence and can sustain high-gradient acceleration over many plasma wavelengths. We suggest that in addition to its celestial application, the MPWA concept can also be of terrestrial utility. A proof-of-principle experiment on MPWA would benefit both terrestrial and celestial accelerator concepts.

The generation of electrostatic wakefields by an ordinary mode radiation and a magnetic field-aligned circularly polarized electromagnetic (CPEM) wave in a magnetized electron–positron–ion (e–p–i) plasma is considered. It is found that the presence of the ions is essential for the generation of upper-hybrid wakefields by the ordinary mode radiation, while the magnetic field-aligned electron plasma wakefields are created by the ponderomotive force of CPEM only if either the ions or the external magnetic field is present in an e–p–i magnetoplasma. The electromagnetic wave generated wakefields can trap both the electrons and the positrons and accelerate them to very high energies.

We present a theory for the acceleration of monoenergetic protons, trapped in a self-organized double layer, by short pulse laser irradiation on a thin foil with the specific thickness suggested by the simulation study of Yan et al (2008 Phys. Rev. Lett.100 135003). The laser ponderomotive force pushes the electrons forward, leaving the ions behind until the space charge electric field balances the ponderomotive force at a distance Δ. For the optimal target thickness D = Δ > c/ω_p, the electron sheath is piled up at the rear surface and the sheath electric field accelerates the protons until they are reflected by the inertial force in the accelerated frame. These protons are therefore trapped by the combined forces of the electrostatic field of the electron sheath and the inertial force of the accelerating target. Together with the electron layer, they form a double layer and are collectively accelerated by the laser ponderomotive force, leading to monoenergetic ion production.