Open Access
Add to BibSonomyAbstract
A fibre-coupled 1.47µm laser diode end-pumped Er:YAG laser system comprising one oscillator and two single pass amplifiers is described. 120mJ pulses at a 30Hz repetition rate and 1.64µm emission wavelength are reported.
© 2014 Optical Society of America
1. Introduction
One major requirement for laser illumination for active imagery or laser range finder is the “eye-safe” property. Laser pulses with an emission wavelength around the 1.5-1.6µm zone are well suited for that purpose. For long range detection (10’s km) energy per pulse in the 100mJ range is mandatory.
Up to now, such laser pulses are obtained with a Q-Switched Nd:YAG laser, with an emission wavelength of 1.06µm, followed by nonlinear wavelength conversion, such as an Optical Parametric Oscillator (OPO). A lot of work has been performed on that subject since the 60’s, leading to interesting performances (see for example [1,2]). Beam quality of high energy OPOs has been also addressed [3] and other nonlinear effects, such as Raman [4], have been demonstrated to be effective. Anyway, these schemes remain multi-step ones. Thus, the interest in direct laser emission at 1.5µm under laser diode pumping is high. Indeed, parallel to the Erbium Doped Fibre Amplifier for telecoms, codoped Er,Yb:Glass lasers were developed under flash pumping, then laser-diode pumping, and eventually became commercial products[5]. But, such lasers are hampered by the very poor thermal properties of the glass, limiting the Q-Switch energy - repetition rate product to values unsuitable for our applications. Despite promising recent results [6], a suitable codoped Er,Yb crystal is not available yet, mainly due to poor ytterbium to erbium transfer efficiency.
For a decade now, the resonant pumping of erbium-doped crystals has been studied. The main advantages, apart for the thermal properties of crystals, are the very small quantum defect and the existence of two pumping bands: at ~1.46-1.48µm and close to 1532nm. Both have been used either for continuous wave or Q-Switch regimes [7,8]. The 1.47µm band is suitable for laser diode pumping as high power 1470 nm laser diodes are becoming more and more available. Due to its narrow linewidth, the 1532nm line is better suited with erbium fibre laser pumping [9,10] or requires spectrally narrowed laser diodes [11]. The highest Q-Switched energy of 100 mJ disclosed to date relies on this pumping scheme and Er:YAG crystals [12] Unfortunately, very few details are known as only the abstract is available.
After some design considerations based on modelling and experimental measurement of thermal lens focal length, we will present the experimental results achieved with our Q-Switched laser system based on laser-diode end-pumping of Er:YAG laser and single pass amplifiers.
2. System design and set-up
The major drawbacks of the resonant pumping of erbium laser are i) the small pump absorption cross section, requiring high intensity pump source; ii) up-conversion processes, requiring low erbium doping level and thus long absorption length (10-100mm range). As a result, an end-pumping scheme is advisable.
Our preferred architecture is represented in Fig. 1. Two fibre-coupled laser-diodes are focused on a single laser crystal through dichroic mirrors for pump/laser beams separation. This scheme is used both for the laser or the amplifier(s).
Fig. 1 Oscillator set-up: LD: fibre coupled laser diode; DM: dichroic mirror; OC: optical coupler; QS: Q-Switch cell; DL: diverging lens. All mirrors are flats.
We performed some modelling of resonantly pumped erbium laser. The aim was to obtain some clues for the optimal crystal lengths and beam diameters. Our in-house code, based on rate equations, models a 3-level Q-Switched laser in two steps: first the pumping, then the laser action. The code takes into account for “2D” spatial effects (radial and longitudinal) and temporal pump saturation and pump beams divergence. The two contra-propagating pump beams are supposed to have the same waist, symmetrically placed with respect to the centre of the laser crystal. The size of the pump beams evolves like a Gaussian beam with a M2 = 100 (corresponding to a 400µm diameter fibre with a 0.22 numerical aperture). The laser beam is supposed to be collimated.
The modelling underlined the stringent balance between laser fluence and efficiency. For example, Fig. 2 presents the calculated output energy from an Er:YAG laser for different transmission of the cavity output coupler and different intra-cavity laser fluence. In practice, the intra-cavity fluence has to be smaller than 10J/cm2 to avoid damages on optical components.
Fig. 2 Calculated output energy for different output coupler transmission (T) and different intra-cavity fluence. Experiment: see section 3.1.
With our 100mJ + target in mind and considering these results and the available laser diodes, we choose to use an oscillator/amplifier architecture.
The final laser system consists in one oscillator and two single-pass amplifiers.
The oscillator, see Fig. 1, relies on an 6x6x60mm3 Er:YAG crystal with an erbium doping level of 0.5%. The laser crystal is end-pumped on its two faces by two fibre coupled laser diode bars. Each pump fibre (400µm diameter, 0.2 numerical aperture) delivers 120W at 1470nm. The pump spectral width is roughly 10nm. Each pump fibre is imaged with a magnification of 4 in the oscillator crystal. Dichroic mirrors (DM) ensure efficient laser reflection and good pump transmission. The optimal output coupler (OC) has a 40% transmission coefficient. All mirrors are flat. The Q-Switching is performed with an acousto-optic cell (QS). The pump are driven with 6ms pulses at 30Hz. This pump pulse-width value was found to be the optimal one for Q-Switch operation.
The pumping architecture of the two single pass amplifiers is identical to the oscillator, with double end-pumping of the amplifier crystals with similar fibre coupled laser diodes, as presented in Fig. 3. Er:YAG crystal are 75mm long. The main difference is the smaller magnification of the pump fibres in the crystals (x3 instead of x4). A smaller pump-beam -and thus laser beam- leads to a higher gain. But, the magnifications presented here for both the oscillator and the amplifier are the smallest one achievable without reaching the damage threshold of the dichroic mirrors. Indeed, these latter components appear to be the weakest one in our laser system. Imaging lenses (IL) are used between the oscillator and each amplifier to adjust the laser beam size and compensate for thermal lensing in the gain mediums. The losses between the oscillator output and the amplifier input are less than a few percent.
The pump absorption range from 75% to 93% depending of the diode spectrum and the Er:YAG crystal length, as it can be seen on Fig. 4 for two of our six laser diodes. This absorption was measured at full pump power without lasing action or input pulse, as expected in Q-Switch regime.
3. Experimental results
3.1 Laser energy and amplifier gain
To permit a step by step energy increase (to avoid exceeding damage threshold of the dichroic mirrors) without changing the laser beam size within the cavity, we delayed the Q-Switch action after the end of the pump pulse, see Fig. 5. Indeed, changing pump pulse width or intensity leads to strong thermal lens variation and not the fix pump scheme with delayed Q-Switch. We experimentally confirmed that the laser beam size only changes for few percent up to a delay of 1ms, while the energy variation is on a ratio of ~3:1.
The oscillator was first optimised. To achieve large enough laser beam inside the Er:YAG crystal that the thermal lens tends to shrink, a diverging lens (DL, f = −750mm) was inserted close by one face of the laser crystal and the corresponding cavity arm elongated to achieve a total cavity length of 76cm. The laser beam waist, measured to be 850µm, matches very well the pump beam waists (800µm). The laser beam size was then adjusted to match the pump beam size within the two amplifiers.
Output energy of the oscillator and both amplifiers are presented in Fig. 6. Up to 55mJ from the oscillator and 120mJ from the laser system was achieved. The laser performance is reported on the Fig. 2, demonstrating a relatively good agreement with the modelling, although the experimental intra-cavity fluence is estimated to be “only” 5J/cm2.
The single-pass gain of the first (second) amplifier was 45% (40%) respectively. Our attempts to double pass the beam in one amplifier were unsuccessful. Indeed, our optical isolator didn’t handle the requested fluence.
The Er:YAG crystal length effect on the amplifier gain was investigated and is reported in Fig. 7, left part. An even longer crystal should provide a slightly larger gain. Single-pass gain was also investigated for different amplifier pump pulse-width (see Fig. 7, right part) demonstrating that the optimal value is 6ms.
Fig. 7 Single-pass gain achieved for: left, different crystal length; right, different pump pulse-width.
3.2 Emission wavelength and pulse duration
The laser emission wavelength is 1645nm both in free running and Q-Switch mode. Many previous experiments report a 1617nm emission wavelength in Q-Switch mode [7,9,10,13]. The different behaviour of our laser reveals a smaller population inversion which may be due to the use of a pump source with brightness lower than fibre laser pumps.
The measured oscillator pulse-width is 100ns FWHM. This quite long value is supposed to be dictated by the cavity length and the smaller gain at 1645nm than 1617nm. No relevant pulse-width change along the amplifier chain is observed.
3.3 Beam characterization
The beam profiles were measured with an InGaAs camera (Raptor Photonics). The beam far-field slightly degrades after the second amplifier (see Fig. 8), may be due to an imperfect mode-matching of the laser beam and the pump beams.
Beam sizes and beam quality parameters were calculated from these images with the second momentum technique, as described in the ISO11146 norm (see Fig. 9). As expected from the far-field measurement, the second amplifier slightly degrades the beam quality. From M2x = 2.5 / M2y = 2.0 at the oscillator output and M2x = 2.6 / M2y = 2.1 at the first amplifier output, the beam quality increases to M2x = 3.7 / M2y = 3.1 at the second amplifier output.
4. Conclusion
120mJ at a 30Hz repetition rate has been demonstrated from an end-pumped 1.47µm laser-diodes erbium YAG laser system. The output beam quality M2 = 3.1-3.7 should be improvable close to 2-2.5 as only the second amplifier degrades it. The main energy limitation of our laser system is the damage threshold of the pump dichroic mirrors. Similarly, multi-pass in the amplifier was prevented by the damage threshold of our oscillator. The expected “next limitation” is the availability of fibre coupled laser diode system with higher power with the same brightness.
Acknowledgment
This work was supported by the Délégation Générale de l’Armement (DGA) under contract n°06 50 212 00 470 92 58.
References and links
1. G. A. Rines, D. M. Rines, and P. F. Moulton, “Efficient, high-energy, KTP optical parametric oscillators pumped with 1 micron Nd-lasers,” in Advanced Solid State Lasers, T. Fan and B. Chai, eds., Vol. 20 of OSA Proceedings Series (Optical Society of America, 1994), paper PO9.
2. M. S. Webb, P. F. Moulton, J. J. Kasinski, R. L. Burnham, G. Loiacono, and R. Stolzenberger, “High-average-power KTiOAsO4 optical parametric oscillator,” Opt. Lett. 23(15), 1161–1163 (1998). [CrossRef] [PubMed]
3. D. J. Armstrong and A. V. Smith, “150-mJ 1550-nm KTA OPO with good beam quality and high efficiency,” Proc. SPIE 5337, 71–80 (2004). [CrossRef]
4. J. T. Murray, W. L. Austin, and R. C. Powell, “2.5 W eye-safe solid-state raman laser,” in Advanced Solid State Lasers, M. Fejer, H. Injeyan, and U. Keller, eds., Vol. 26 of OSA Trends in Optics and Photonics (Optical Society of America, 1999), paper WC4.
5. See for examplewww.kigre.com.
6. A. Jaffrès, B. Viana, P. Loiseau, G. Aka, C. Larat, and E. Lallier, “Actively Q-switch operation of diode-pumped Er3+, Yb3+, Ce3+: Ca2Al2SiO7 single crystal laser at 1.5-1.6 μm,” in CLEO Europe (2013), paper CE-6.2 TUE.
7. S. D. Setzler, M. P. Francis, Y. E. Young, J. R. Konves, and E. P. Chicklis, “Resonantly pumped eyesafe erbium lasers,” IEEE J. Sel. Top. Quantum Electron. 11(3), 645–657 (2005). [CrossRef]
8. I. Kudryashov, N. Ter-Gabrielyan, and M. Dubinskii, “Resonantly diode-pumped Er:YAG laser: 1470-nm versus 1530-nm CW pumping case,” Proc. SPIE 7325, 732505 (2009). [CrossRef]
9. J. W. Kim, J. K. Sahu, and W. A. Clarkson, “High-energy Q-switched operation of a fiber-laser-pumped Er:YAG laser,” Appl. Phys. B 105(2), 263–267 (2011). [CrossRef]
10. J. W. Kim, D. Y. Shen, J. K. Sahu, and W. A. Clarkson, “Fiber-laser-pumped Er:YAG lasers,” IEEE J. Sel. Top. Quantum Electron. 15(2), 361–371 (2009). [CrossRef]
11. H. Fritsche, O. Lux, C. Schuett, S. Heinemann, W. Gries, and H. J. Eichler, “Efficient Er:YAG lasers at 1645.55 nm, resonantly pumped with narrow bandwidth diode laser modules at 1532 nm, for methane detection,” Proc. SPIE 8599, 85990G (2013). [CrossRef]
12. S. D. Setzler, M. Francis, and E. P. Chicklis, “A 100mJ Q-switched 1645nm Er:YAG laser,” Proc. SPIE 6552, 6552177 (2007).
13. J. W. Kim, D. Y. Shen, J. K. Sahu, and W. A. Clarkson, “High-power in-band pumped Er:YAG laser at 1617 nm,” Opt. Express 16(8), 5807–5812 (2008). [CrossRef] [PubMed]
