The increasing demand for high laser powers is placing huge demands on current laser technology. This is now reaching a limit, and to realise the existing new areas of research promised at high intensities, new cost-effective and technically feasible ways of scaling up the laser power will be required. Plasma-based laser amplifiers may represent the required breakthrough to reach powers of tens of petawatt to exawatt, because of the fundamental advantage that amplification and compression can be realised simultaneously in a plasma medium, which is also robust and resistant to damage, unlike conventional amplifying media. Raman amplification is a promising method, where a long pump pulse transfers energy to a lower frequency, short duration counter-propagating seed pulse through resonant excitation of a plasma wave that creates a transient plasma echelon that backscatters the pump into the probe. Here we present the results of an experimental campaign conducted at the Central Laser Facility. Pump pulses with energies up to 100 J have been used to amplify sub-nanojoule seed pulses to near-joule level. An unprecedented gain of eight orders of magnitude, with a gain coefficient of 180 cm−1 has been measured, which exceeds high-power solid-state amplifying media by orders of magnitude. High gain leads to strong competing amplification from noise, which reaches similar levels to the amplified seed. The observation of 640 Jsr−1 directly backscattered from noise, implies potential overall efficiencies greater than 10%.
Envelope models offer the potential to dramatically reduce the computational overhead of particle-in-cell simulations of laser-plasma interactions. However, the associated approximations inevitably limit their applicability. We here derive the governing equations for an envelope model in order to gauge those limits. The approximations for electron response are shown to be exact in one dimension for the correct initial conditions. For multidimensional geometries, limits are placed on the canonical momentum perpendicular to the laser field, and on the amplitude of the laser field relative to the laser spot size. It is shown that those conditions are readily satisfied for the case of Raman amplification.
Raman backscattering (RBS) in plasma has been proposed as a way of amplifying and compressing high intensity
laser pulses for more than a decade. Not like the chirped pulse ampliffication (CPA) laser system, in which the
laser intensity is limited by the damage threshold of conventional media, plasma is capable of tolerating ultrahigh
laser intensities, together with RBS which is enable to transfer laser energy efficiently from a higher frequency pulse to a lower one, this scheme opens a scenario of the next generation of laser amplifiers. Experimental investigation has been carried out with a long (250 ps) pump pulse and a counter-propagating short (70 fs) probe pulse interacting in an under-dense preformed capillary plasma channel. Energy transfer from the pump pulse to the probe was observed. The guiding property was studied and the energy gain dependence of pump and probe energy were recorded.
The influence of wavebreaking on Raman amplification is investigated. A phenomenological modification is added
to a set of slowly varying envelope equations, and found to give good agreement to particle-in-cell simulations
for cold plasma in the wavebreaking regime.
For warm plasma, good agreement is not found using the warm wavebreaking limit. However, the PIC
simulations show that the decreased wavebreaking limit does have a significant impact on amplification. The
limitations of our model are discussed, and possible future work outlined.
Stimulated Raman backscattering in plasma has been suggested as a way to amplify short laser pulses to intensities
not limited by damage thresholds as in chirped pulse amplification using conventional media. Energy
is transferred between two transverse electromagnetic waves, pump and probe, through the parametric interaction
with a longitudinal Langmuir wave that is ponderomotively excited by their beat wave. The increase of the
plasma temperature due to collisional absorption of the pump wave modifies the dispersion of the Langmuir wave:
firstly, its resonance frequency rises (Bohm-Gross shift), and secondly, Landau damping sets in. The frequency
shift acts in a similar way to a chirp of the pump frequency, or a density ramp: different spectral components of
the probe satisfy the resonance condition at different times. This limits their growth, while increasing the bandwidth
of the amplifier, thus leading to superradiant amplification. Landau damping may shorten the probe pulse,
but reduces the amplification efficiency. We investigate these effects analytically and using numerical simulations
in order to assess their importance in experimental demonstrations, and the possibility of applications.
The role of thermal effects on Raman amplification are investigated. The direct effects of damping on the
process are found to be limited, leading only to a decrease from the peak output intensity predicted by cold
plasma models. However, the shift in plasma resonance due to the Bohm-Gross shift can have a much larger
influence, changing the required detuning between pump and probe and introducing an effective chirp through
heating of the plasma by the pump pulse. This "thermal chirp" can both reduce the efficiency of the interaction
and alter the evolution of the amplified probe, avoiding the increase in length observed in the linear regime
without significant pump depletion.
The influence of this chirp can be reduced by using a smaller ratio of laser frequency to plasma frequency,
which simultaneously increases the growth rate of the probe and decreases the shift in plasma resonance. As
such, thermal effects only serve to suppress the amplification of noise at low growth rates. The use of a chirped
pump pulse can be used to suppress noise for higher growth rates, and has a smaller impact on the peak output
intensity for seeded amplification.
For the parameter ranges considered, Landau damping was found to be negligible, as Landau damping rates
are typically small, and the low collisionality of the plasma causes the process to saturate quickly.
Raman backscattering (RBS) in plasma is an attractive source of intense, ultrashort laser pulses, which has the
potential asa basic for a new generation of laser amplifiers.1 Taking advantage of plasma, which can withstand
extremely high power densities and can offer high efficiencies over short distances, Raman amplification in
plasma could lead to significant reductions in both size and cost of high power laser systems. Chirped laser pulse
amplification through RBS could be an effective way to transfer energy from a long pump pulse to a resonant
counter propagating short probe pulse. The probe pulse is spectrally broadened in a controlled manner through
self-phase modulation. Mechanism of chirped pulse Raman amplification has been studied, and features of
supperradiant growth associated with the nonlinear stage are observed in the linear regime. Gain measurements
are briefly summarized. The experimental measurements are in qualitative agreement with simulations and
High power short pulse lasers are usually based on chirped pulse amplification (CPA), where a frequency chirped
and temporarily stretched "seed" pulse is amplified by a broad-bandwidth solid state medium, which is usually
pumped by a monochromatic "pump" laser. Here, we demonstrate the feasibility of using chirped pulse Raman
amplification (CPRA) as a means of amplifying short pulses in plasma. In this scheme, a short seed pulse is
amplified by a stretched and chirped pump pulse through Raman backscattering in a plasma channel. Unlike
conventional CPA, each spectral component of the seed is amplified at different longitudinal positions determined
by the resonance of the seed, pump and plasma wave, which excites a density echelon that acts as a "chirped"
mirror and simultaneously backscatters and compresses the pump. Experimental evidence shows that it has
potential as an ultra-broad bandwidth linear amplifier which dispenses with the need for large compressor
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 nonlinear regime of Raman amplification has been studied including the combined effects of relativistic and
ponderomotive nonlinearities. The study is important for interaction of mildly relativistic pump and probe laser pulses.
Nonlinear coupled temporal evolution of fields and density in Raman amplification is analyzed. It is shown that the
saturation amplitude and time of the probe pulse in nonlinear regime depends upon the intensity of the electromagnetic
waves and the density of the medium. Further in the nonlinear regime the probe laser pulse gain is severely affected by
changes in both the electromagnetic wave amplitude and the plasma density.