Frequency Doubling: Long Pulses Cause Trouble

Posted on 2008-01-14. Permanent link: http://www.rp-photonics.com/spotlight_2008_01_14.html

Ref.: S. Favre et al., "High-power long-pulse second harmonic generation and optical damage with free-running Nd:YAG laser", IEEE J. Quantum Electron. 39 (6), 733 (2003); encyclopedia articles on frequency doubling and photodarkening

Frequency doubling can be efficiently performed either by intracavity frequency doubling in a continuous-wave laser, or with a Q-switched laser (intracavity or extracavity). It thus seems surprising what the paper cited above reports: frequency doubling with long pulses is difficult – at least at high power levels.

The authors worked with a powerful Nd:YAG slab laser, producing pulses with 0.1-5 μs duration and a peak power of 1-2 kW. The high average power forced them to use extracavity doubling, as the thermal effects in the nonlinear crystal (made of KTP) would have been to strong to tolerate them within the laser resonator. With proper focusing, the optical intensity could easily have been made high enough for efficient doubling, but the trouble was that crystal damage set in well before a high efficiency was achieved. The problem is apparently the following:

• When a high second-harmonic intensity is present for a longer time (hundreds of microseconds), the KTP crystal becomes absorbing. Apparently, temporary color centers are the origin of this photodarkening. The color center formation may be related to two-photon absorption.
• Absorption of the high infrared intensity subsequently leads to damage. (The second-harmonic light is also absorbed, but has a lower intensity.)

This explains why frequency doubling with nanosecond pulses works better. In continuous-wave operation, the situation should be even worse, as the color centers would have even much more time to build up. A solution can be intracavity doubling, where far lower optical intensities are required for efficient conversion: the non-converted infrared light is "recycled" in the laser resonator. However, intracavity doubling is limited in average power, as mentioned above.

Of course, one may use a crystal material which does not have these temporary color centers. The authors reported slightly better results with LBO, but unfortunately without elaborating on damage and induced absorption. LBO is probably better in this respect, although it does need to be operated with higher intensities due to its lower nonlinearity. Even if damage is no issue, it is sometimes difficult to reach sufficiently high intensities with long pulses or in continuous-wave operation.

So we see that frequency doubling long pulses (and in continuous-wave operation) can be problematic. The Spotlight article of 2007-03-05 explained why very short pulses are also involving difficulties – for entirely different reasons.