Power Amplifier System for High-Power, Passively Q-switched Micro laser

I. INTRODUCTION

Passively Q-switched microchip lasers are simple, compact and reliable sources of high repetition rate (1 to 100 kHz), near-infrared, sub-nanosecond pulses. To date, low-energy (0.3 to 3 mJ/pulse) and mid-energy (30 to 180 mJ/pulse) microchip lasers have been reported [1] with pulse durations of 200 to 500 ps and 650 to 2000 ps, respectively. For some applications, for instance, high precision ranging and imaging, higher-energy pulses, up to 360-440 mJ/pulse [2, 3] are required with pulse durations approaching 200 ps. The pathway to higher pulse energies requires the use of higher saturable absorption, which inevitably leads to longer pulse durations. An alternative approach is the use of a MOPA design with a microlaser oscillator and multipass amplifier [3]. In the only reported work to date on this approach, a MOPA system operated in the 10-m J range, producing 500-ps pulses [4].

We present here a MOPA system generating 335 m J at 1064 nm with efficient harmonic conversion to the visible and UV. At all wavelengths, the pulse durations were <400 ps. This source is a valuable tool for applications that include ranging, LIDAR, micro-materials processing, and UV spectroscopy in chemistry and biochemistry.

II. SYSTEM DESCRIPTION AND EXPERIMENTAL RESULTS

                Figure 1 shows a schematic layout of our MOPA system. A 1-watt, fiber-coupled diode laser operated at a 2 kHz pulse rate is used as a pump source for the Cr:YAG passively Q-

switched Nd:YAG microlaser. Pump light emerging from the 100-m m, 0.22 NA fiber is collected and focused into the microchip using two AR-coated aspheric lenses. The fiber is

positioned at the front focal plane of the first lens, and the microchip at the nominal back focal plane of the second lens. Two different "telescopes" were used to optimize the microlaser output: a 4:3 or a 2:1 reducing telescope. Approximately 98&percnt; of the light emerging from the fiber is delivered to the microchip when using either telescope.

The microchip laser output is collected by a spherical 50-mm FL lens. This lens is typically positioned about 63 mm from the microchip so the beam is gradually focused into the amplifier stage. The beam then passes through a TGG Faraday isolator equipped with input and output Glan-laser polarizers.

A half-wave plate positioned before the collimator lens adjusts the polarization angle of the microchip beam as it enters the first polarizer of the isolator. A second half-wave plate adjusts the polarization angle of the beam emerging from the second polarizer. The beam is turned 90° by the first 45° -incidence HR (45 HR), goes through a +150 mm FL cylindrical lens that focuses in the vertical plane, and bounces off the second 45 HR, before entering the 3-pass amplifier stage.

The cylindrical lens is positioned about 150 mm from the center of the amplifier slab, taking into account the fact that the beam makes three passes through the slab (the separation between the slab assembly’s miniature fold mirrors is about 20 mm, and the slab is about 15 mm long). The beam is back-reflected through the amplifier with a flat HR , and makes another 3 passes through the amplifier slab.

The back-reflected, double-pass amplified beam passes back through the optical system and into the Faraday isolator. The plane of polarization at the first polarizer is now rotated 90° relative to the microchip laser polarization. The double-pass-amplified beam is coupled out the system at the first polarizer, and emerges with a polarization vertical to the plane of the paper.

Fig. 1. Schematic layout of the Microlaser-Amplifier system.

Our amplifier gain material, Nd:YVO4, is particularly well suited for amplifying pulses with energies below 100 m J because of its extremely high gain. This is demonstrated by Fig. 2, where we present calculated double-pass gain curves for cw-pumped multi-pass slab amplifiers based on different Nd-doped materials.

The amplifier design employs a slab-geometry gain module with transverse pumping. The gain module consists of an a-axis-cut, 2-mm high by 3-mm wide by 15-mm long Nd:YVO4 slab. The slab was cw side-pumped by two 20-W diode laser bars emitting at 808 nm, with top and bottom heatsinking. The side faces of the slab are polished and antireflection coated at 808 nm for maximum coupling of the pump light. The outputs of the diode laser arrays are collimated, in the highly diverging direction, by a drawn aspheric cylinder lens to produce a nearly rectangular excitation region in the laser crystal. The laser mode is passed three times through the length of the excitation region, using a pair of miniature external mirrors, essentially transverse to the pump beam. This design allows for efficient extraction of the stored energy in a TEM00-mode beam.

Fig. 2. Calculated double-pass gain curves for cw-pumped Nd-doped multi-pass slab amplifiers.

Three different monolithic microchip oscillator designs were evaluated, each having a different Cr:YAG layer thickness. The first two designs were made by Synoptics according to Q-Peak specifications. Both employ a 0.5 mm thick layer of 3&percnt;-doped Nd:YAG. One chip design has a 0.25-mm-thick layer of Cr:YAG with a nominal unsaturated absorption of 6 cm-1. The output facet of the chip is coated for 80&percnt;R. The other chip design is the same, except that the Cr:YAG layer thickness is 0.5 mm. Originally, we intended to try a similar third design, but with a 0.75 mm Cr:YAG layer and a 60&percnt;R output facet. However, this chip was not coated properly and was replaced with an off-the-shelf chip designed by Synoptics, which had features close to what we desired. This chip has a 1.25 mm layer of 1.9&percnt;-doped Nd:YAG, a 0.75-mm layer of Cr:YAG (6 cm-1), and an 80&percnt;R output facet. The pulse durations calculated for the three designs were 304, 204 and 200 ps, respectively. We believe that pump-light-induced bleaching is one of the reasons for increasing pulse durations [5].

The 0.75-mm Cr:YAG microlaser was used to construct the oscillator - double-pass-amplifier system. The microchip laser was quasi-cw pumped with a 1.1-watt peak-power pulse, at a 2 kHz pulse rate. The width of the diode pump pulse was adjusted to ensure single-pulse oscillation even when the double-pass amplifier was on. A typical oscilloscope trace of the oscillator pulses is shown in Fig. 3.

With 6.4 mW of microchip laser power, double-pass amplifier power was about 670 mW at an amplifier current of 31A. The beam quality of the double-pass-amplified beam was measured (by the Spiricon M2 meter) using the "90/10 knife-edge" method. M2 in the horizontal and vertical planes was measured to be 1.38 and 1.28, respectively. The pulse durations decreased to 370 ps as compared to ~ 440-ps from the microlaser.

Fig. 3. Oscilloscope trace of 440-ps oscillator pulse.

Table 1. Microlaser characteristics.

Microlaser parameters

Microlaser # 1, 4:3 telescope

Microlaser # 2, 2:1 telescope

Microlaser # 3, 4:3 telescope

Average power, mW

4.4

3.1

6.4

Pulse energy, m J

2.2

1.55

3.2

Pulse width, FWHM, measured, ps

700

400-440

400-440

Delay, m sec

90

40

70

Pump pulse width, m sec

120

60

120

Jitter, ns

± 100

± 100

± 100

Drift, 5 min, ns

± 300

± 200

± 200

We have also conducted experiments on nonlinear conversion of the amplifier beam. For second harmonic generation (SHG) we used a non-critically-phase-matched, Type I LBO crystal, with dimensions of 3 x 3 x 15 mm, mounted in a 1700 C, temperature-stabilized oven. Third harmonic generation (THG) at 355 nm was accomplished with a room-temperature, 3 x 3 x 12 mm, Type II critically-phase-matched LBO crystal (q = 42.7° , f = 90° ). And, finally, for fourth harmonic generation (4HG) at 266 nm, a Type I critically-phase-matched, room-temperature BBO crystal (q = 47.6° , f = 0° ), 3 x 3 x 7 mm crystal was used. The beams were separated using a Pellin-Broca prism. At input power of 670 mW, the output power of SHG, THG and 4HG was 400, 240 and 86 mW, respectively, which corresponds to ~ 60&percnt;, 36&percnt;, and 13&percnt; conversion efficiency.

III. CONCLUSION

We have designed and constructed a highly efficient diode-pumped, short-pulse, energetic, compact and reliable microlaser-amplifier system. This design approach, we believe, will allow us to achieve even shorter (~ 200 ps) and higher-energy pulses, increase the conversion efficiencies of harmonic generation, and improve the compactness of the system.

REFERENCES

1. J. J. Zayhowski, "Passively Q-switched microchip lasers and applications," Rev. Laser Eng., v. 26, pp. 841-846 (2008).

2. J. J. Zayhowski, C. Dill III, C. Cook, J. L. Daneu," Mid- and high-power passively Q-switched microchip lasers," in OSA Trends in Optics and Photonics on Advanced Solid-State Lasers, v. 26, M. M. Fejer, H. Injean, and U. Keller (eds), (Optical Society of America, Washington DC, 2007) pp. 178-186.

3. J. J. Degnan, "Optimal design of passively Q-switched microlaser transmitters for satellite laser ranging," in Proceeding of 10-th International Workshop on Laser Ranging, Shanghai, PRC, 2006, pp. 334-343.

F. Druon, F. Balembois, P. Georges, A. Brun, "Compact high-repetition-rate pulsed UV sources using diode-pumped microchip laser and multipass amplifier," in Advanced Solid-State Lasers, OSA Technical Digest (Optical Society of America, Washington DC, 2008), pp. 329-331.

M. A. Jaspan, J. A. Russell, D. Welford, "Degradation of passively Q-switched microlaser performance due to pump-light induced bleaching of the saturable absorber," to be submitted to Opt. Letters.