Chirped-Pulse Amplifier Laser Systems

Just prior to the advent of the Chirped-Pulse Amplifier (CPA) laser systems, the prevailing architecture for high power solid state laser systems consisted of a monochromatic oscillator that produced a high quality seed pulse of some several hundred psec duration, followed by an amplifier or amplifier chain that multiplied the seed pulse's energy several billionfold. This architecture had topped out with peak power densities of several GW/cm2 in the final amplifier. The path to even higher output intensities in a solid state laser amplifier is blocked by the self-focusing effect that bulk optical materials - ruby, glass, Nd:YAG, Lucite, etc. - present to light propagation at intensities of several GW/cm2 and above.


At such high intensities inside bulk optical matter, the wavefront of an optical pulse experiences a higher refractive index where the local beam intensity is higher.Since the beam is more intense at its center than around its periphery, as such a finite front propagates through a laser rod, its diameter shrinks and drives the beam center intensity even higher, which accelerates the collapse of beam diameter with propagation distance even further. [Fig cpa_fig1, above]

Ultimately, after a certain propagation distance through the laser rod, the beam has collapsed to a diameter so small that the extreme field strengths and absorbed power densities ruin the laser rod. The amplifier laser rod has shattered itself.


[Fig cpa_fig2, above] The architecture of a CPA laser system begins with a Ti:sapphire oscillator that produces a high quality ultra short optical pulse, typically 30-100 fsec in duration and a few nanoJoules of energy. Crucially, the pulse is spectrally broad-band.

This broad-band ultra short seed pulse is then directed through a pulse stretcher grating pair. This is a pair of special optical gratings oriented in such a way that longer wavelength components of the seed pulse travel a shorter path through it than do the shorter wavelengths, and hence emerge from it sooner to enter the amplifier stage. The result is that a temporally stretched version of the seed pulse now heads on to the amplifier. Its duration has been stretched out typically 20,000 times or so, and its amplitude reduced by an identical factor. The leading edge of this flat-topped stretched seed pulse is "redder" than its trailing edge. The instantaneous wavelength of the roughly one nsec long pulse, is swept smoothly and uniformly throughout its duration. In the parlance of electromagnetic wave technology, we say this pulse is chirped.

The broad-band temporally stretched seed pulse is amplified by a broad-band amplifier, usually Ti:sapphire or Nd:glass, some ten billion-fold or more, up to the several GW/cm2 intensity limit imposed by nonlinear optics coefficient values of the amplifier medium itself.

Out of the broad-band amplifier the now energetic chirped pulse is directed through a compressor grating-pair. This is a pair of special reflective optical gratings oriented in such a way that longer wavelength components of the high energy chirped pulse travel a longer path through it than do the shorter wavelengths, and hence emerge from it retarded. The chirp is now undone. The spectrally broad energetic pulse emerging from the compressor optics now has all its spectral components occurring at the same instant. This shortens the duration of the energetic pulse by typically some 20,000 times, and multiplies its peak intensity by the same factor. Because of the extreme optical power densities, from the compressor on the beam is never again directed through bulk optical matter - windows, lenses, etc.. In fact, even air in the beam path is disallowed at the high beam intensities. A 100 fs pulse of a few Joules energy in a few centimeters diameter beam will self-focus in air down to a radius so tight that the consequent extreme electric field strengths in the beam ionize the gas molecules along its path, drawing a long spark in raw air. Hence from the compressor on, the beam path is contained entirely inside vacuum. The final output pulse is steered towards the target and focused with special hardened reflective optics. On target optical power densities of 1.0E+14 W/cm2 to 1.0E+21 W/cm2 excite matter to extreme conditions of temperature for x-ray production, hydrodynamics studies, and other phenomena of interest.

For CPA laser systems to yield the shortest possible output pulse duration and hence the greatest peak output power, the spectrum after amplification must be exactly the same as that before. This is why Ti:sapphire or Nd:glass is used for the laser amplifier. These amplifiers have flat response over a broad spectral range.

When a high-energy laser amplifier fires, waste heat makes the amplifier bulk optical medium warm up slightly and optically distort enough to need a cool down time. An old style laser that produces a TW output pulse by emitting 10 kJ of optical energy in 100 psec, may need an hour or more of cool down time before for the next firing. In contrast, a CPA laser system might achieve a 1TW optical pulse might by delivering a mere 10 J of optical energy in a fleeting 100 fsec. With much less waste heat to make the same peak output power, it takes only a few minutes for the ultra short pulse CPA laser system to cool down and be ready to fire again.


Physics and Space Technology home pageLLNL Home PageLLNL Disclaimers

UCRL-MI-135159