Laser beams can be strongly compressed in a plasma by stimulated Raman backscattering in a time short compared to the time scale for filamentation instabilities to develop. Such a compression should make feasible multi-MJ multi-exawatt-laser pulses technologically challenging by other means. The compression efficiency can reach nearly 100% at the Langmuir wave breaking limit. [S0031-9007(99) PACS numbers: 52.40.Nk, 42.65.Yj, 52.35.Mw Achievable laser intensities have increased remarkably during recent years mainly due to the method of chirped pulse amplification (CPA). Petawatt (PW 10 15 W) powers and kJ energies have been attained by stretching, amplifying, and then recompressing laser pulses [1]. Certain modern applications, however, require even higher powers and energies. An example of such an application is the fast igniter approach to inertial confinement fusion [2]. Higher power pulses could have applications for radiography of dense materials. Important applications are already envisioned for exawatt (EW 10 18 W) MJ pulses. The higher total power in the pulse presents a practical difficulty to CPA techniques, because very large gratings would be needed to handle large power and heat loads.The rationale of the CPA technique of stretching and recompressing is to avoid nonlinear effects which tend to spoil the beam quality. These nonlinear effects appear in the refraction coefficient at power densities ϳGW͞cm 2 . Longitudinal recompressing by ϳ10 3 times allows CPA to reach power densities ϳTW͞cm 2 , without focusing in vacuum. For kJ 0.5 psec pulse, the cross section of the amplifier and compressor must be ϳ2000 cm 2 , so that ϳ50-cm-diameter beam and ϳ1-m-diameter gratings are required [1]. A higher recompression factor of 10 4 would allow CPA to get a 10 kJ, 0.5 psec pulse using gratings about a meter in diameter. For higher energies, larger, more expensive gratings are required.This Letter examines an alternative to the CPA approach, wherein the laser beams are amplified to power densities much higher than a GW͞cm 2 , even without stretching-recompressing and focusing outside the amplifying medium. The medium capable of bearing such high power densities and heat loads is plasma, but intense pulses in plasmas are subject to numerous instabilities. In particular, the modulational instability can lead to filamentation of the laser beam followed by ponderomotive expulsion of electrons from the filaments. The critical power for such a cavitation is P cr ഠ 17͑v͞v p ͒ 2 GW, where v is laser frequency and v p is plasma frequency [3]. Larger P cr occurs for larger v͞v p , but then the laser-plasma coupling becomes poor and hence the compression efficiency decreases. For a reasonably large frequency ratio, say v͞v p 10, one has P cr ഠ 1.7 TW.Highly overcritical powers are needed, but instabilities stand in the way of reaching those powers.The difficulties could be overcome by ultrafast compression of laser beams, such that highly overcritical laser powers are attained before the instabilities develop. Stimulat...

The recently proposed scheme of so-called "fast compression" of laser pulses in plasma can increase peak laser intensities by 10(5) [Phys. Rev. Lett. 82, 4448 (1999)]. The compression mechanism is the transient stimulated Raman backscattering, which outruns the fastest filamentation instabilities of the pumped pulse even at highly overcritical powers. This Letter proposes a novel nonlinear filtering effect that suppresses premature backscattering of the pump in a noisy plasma layer, while the desired amplification of a sufficiently intense seed persists with a high efficiency. The effect is of basic interest and also makes it robust to noise the simplest technologically fast compression scheme.

Laser energies and powers, significantly much higher than available now through the most advanced chirped pulse amplifiers, might be achieved in much smaller devices. The working medium in such devices is plasma, capable of tolerating ultrahigh laser intensities within times shorter than it takes for filamentation instabilities to develop. The ultrafast amplification mechanism that outruns filamentation instabilities is the transient Raman backscattering of a laser pump in plasma. In principle, this mechanism is fast enough to reach nearly relativistic pumped pulse intensities, like 1017 W/cm2 for λ=1 μm wavelength radiation. Such a nonfocused intensity would be 105 times higher than currently available. This mechanism also produces complete pump depletion. Many amplifiers with expensive and fragile meter-size gratings might then be replaced by a single amplifier comprised of a 1 cm size plasma layer. Raman instabilities of the pump to noise, as the pump traverses plasma layer towards the seed pulse, can be suppressed by detuning the resonance appropriately, even as the desired amplification process persists with high efficiency due to nonlinear resonance broadening. Moreover, since the peak intensity scales like 1/λ2, even much higher laser intensities might become feasible when appropriate x-ray pump lasers are developed.

Backward Raman amplification (BRA) in plasmas holds the potential for longitudinal compression and focusing of powerful x-ray pulses. In principle, this method is capable of producing pulse intensities close to the vacuum breakdown threshold by manipulating the output of planned x-ray sources. The minimum wavelength limit of BRA applicability to compression of laser pulses in plasmas is found.

No abstract