CC BY
4The 30th International Conference on Advanced Laser Technologies ALT'23
LS-O-15
The lifetime value of the terminal Nd:YAG laser level measured from
direct gain saturation observations
V.B.Morozov, A.N.Olenin, D.V.Yakovlev
Physics Faculty of M.V.Lomonosov Moscow State University, 119899 Leninskiye Gory, Moscow, Russia
Nd:YAG have been one of the most popular laser media for several decades due to the successful combination of useful properties of both the activator and the base crystal. Laser generation most often occurs between the Stark sublevels of the 4F3/2^4In/2 transition with the strongest line at a wavelength of about 1.064 ^m. Owing to relatively long radiation lifetime of the upper laser level (~230 ^s) Q-switching is the most commonly used regime in Nd:YAG lasers. This type lasers generate pulses of nanosecond duration (typically 10-30 ns), and are widely used in scientific research, technological, medical, pulse laser ranging and other applications.
Regarding the lifetime value of the lower laser level, there is no common consensus here and various considerations were discussed. Non-radiative transitions within the 4Iii/2 multiplet are characterized by times <<10-8s, while for the transition between 4Iii/2 and the lower 4Ig/2 multiplet ~10-8s is suitable as a first approximation. This is much shorter than the lifetime of the upper laser level and usually does not exceed the pulse duration with Q-switching. That is, such an approximation is quite suitable in many practical cases for the characteristics of an ideal 4-level system. Indeed, this transition in Nd:YAG is usually regarded as a classic example of the 4-level system.
The transition linewidth of ~4.5 cm-1 at a temperature of 300K makes the Nd:YAG crystal suitable also for generating picosecond pulses. With use of passive or active mode locking in Nd:YAG lasers, pulses of from ~10 to ~100 ps durations can be produced. Typical applications of such lasers are scientific research, micromachining, remote precision laser ranging, cosmetology and others.
The generation and amplification of picosecond pulses shorter than the lower laser level lifetime are essentially nonstationary processes. In fact, on this time scale the system operates according to a three-level scheme. In the case of the arrival of an amplified pulse of energy close to the saturation energy, the populations of the upper and lower levels tend to equalize. The amplifying ability is partially restored only after a time corresponding to the lower level depopulation time. Thus, the problems of efficient amplification of ultrashort pulses in two-pass, multi-pass and regenerative amplifiers are essentially nonstationary if the time between successive passes of the amplified pulse through the amplifying medium is comparable to the lower level relaxation time.
Thus, the lifetime of the lower laser level refinement is of both fundamental and practical importance for common activating ions and various crystalline and glass matrices. The transition at 1.064 ^m in Nd:YAG is of obvious interest in this respect, and quite a lot of papers have been published aimed its terminal level lifetime clarification. The results are distributed from 500 ns to 170 ps, with most of the results being significantly just the lower estimate. Such a broad scatter is obviously due to the use of estimate approaches based, typically, on indirect measurements. While the direct population diagnostics of the required multiplet component is an obvious problem.
We have carried out direct measurements based on the diagnostics of the gain recovery dynamics after the passage of a saturating pulse through the amplifying medium. In the experiments, we used a picosecond Nd:YAG laser generating a 25 ps pulses with an energy of 20 mJ and synchronized with it by pumping the side-diode-pumped Nd:YAG amplifier with a crystal aperture of 3 mm and a small-signal gain per pass of >10. The main part of picosecond laser output was used as a pulse that saturates the laser transition in the preliminarily pumped active medium of the amplifier, and a small part was split off, passed through a controlled optical delay line, and passed through the amplifier with crossed polarization as a probe pulse diagnosing the gain. The dependence of the probe pulse amplification on the delay time with respect to the saturating pulse was measured.
Thus, the scheme used made it possible to test the gain magnitude defined by the population difference between precisely those components of the upper and lower multiplets which participate to the laser transition. A detailed description of experiments and results, their comparison with previous data and accounting in practical amplifying stages are planned to be discussed.
