Divergence compensation, optimization of the optical-to-optical efficiency, and high beam quality of signal and idler beams of a high-energy mid-infrared ZnGeP2 (ZGP) optical parametric oscillator (OPO) have been demonstrated by use of a Galilean telescope inside the nonplanar fractional-image-rotation enhancement (FIRE) ring resonator. With a small variation of the distance between the lenses of the telescope, the divergences of signal and idler beams could be adjusted. Up to 36 mJ of mid-infrared pulse energy in the 3-5 µm wavelength range is obtained with 92 mJ of pump energy on crystal. The beam quality factors M2 are < 1.5 for the resonant signal beam and the non-resonant idler beam, respectively. Actually, this is an improvement of the beam quality by a factor 3 for the signal and ~2.7 for the idler beam compared without using a telescope inside the FIRE ring resonator.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Mid-infrared optical parametric oscillators (OPOs) operating in the mid-infrared transmission window (3-5-µm wavelength range) are the subject of ongoing research. They have a number of applications in earth monitoring, remote sensing and surgery and they are key components in directional infrared systems for countermeasure against infrared sensors. There is a strong trend to high pulse energies and average powers, where substantial progress has been achieved in recent years [1–7].
One promising material for such devices is zinc germanium phosphide (ZGP) which has the highest nonlinear coefficient (75 pm/V) and the highest thermal conductivity (35 W/mK) of any bulk birefringent crystal being transparent and able to be phase matched at 2 μm . In addition, ZGP is commercially available in high quality. The optical transparency range is 2-12 µm. It shows defect related residual absorption at near infrared wavelength and thus has to be pumped by laser radiation beyond 2-μm, where typical absorption coefficients are in the range of 0.05 – 0.1 cm−1 at a pump wavelength of 2.05 µm leading to approximately 10% absorption of the average pump power for typical crystal lengths between 10 and 20 mm. The thermo-optical coefficient of ZGP is 150·10−6 K−1 at wavelength of 4 µm and hence ~20 times the value of YAG. Therefore, in case of high average power and with high repetition rate systems, thermal lensing in the ZGP crystal can degrade the OPO beam quality considerably .
A significant drawback of ZGP crystals is the low damage threshold of < 1-2 J/cm2 for nanosecond pulses . In high-energy nanosecond pulse systems, the beam diameter has to be increased to stay below the damage threshold of the nonlinear medium, leading to a high Fresnel number of the resonator and, therefore, poor mode discrimination. In high-energy nanosecond pulse systems, the main challenge is to maintain good beam quality while increasing the pulse energy. This problem can be solved by using a master oscillator power amplifier (MOPA) approach, where the output from a low-energy OPO with good beam quality is amplified without significant degradation of the beam quality. Haakestad et al.  reported a ZGP-based MOPA system with up to 212 mJ energy in the 3-5 µm range at 1 Hz repetition rate with a beam quality M2 = 3. However, because MOPA systems tend to be relatively complex, it is of interest to study the performance limitations of simple OPOs. In principle, the mode discrimination of the OPO could be improved by increasing the OPO resonator length, but this would lead to reduced efficiency due to a longer signal buildup time, and would result in increased system size. Another possibility is to use unstable resonators [11–13] or image rotation in a nonplanar ring cavity, such as the RISTRA  or FIRE  resonator. Advantages of these resonators are the generation of more symmetric beams by averaging over pump beam inhomogeneity and insensitivity to mirror tilts, making vibration-tolerant designs such as quasi-monolithic cavities with no mirror adjustment possible . In , an efficient high-energy mid-infrared ZGP OPO based on the nonplanar FIRE resonator has been demonstrated at a repetition rate of 1 Hz. Here, M2-values from 4 to 6 were reported for signal and idler beam using a pump spot diameter of 5.5 mm. Recently, the authors showed that the beam quality of a high-energy mid-infrared ZGP OPO could be improved significantly by use of a negative lens inside the nonplanar FIRE resonator compensating for thermal lensing and gain guiding effects . However, this negative lens leads to a stronger divergent resonant signal beam and we observed a strong focusing effect of the non-resonant idler beam leaving the resonator.
In this paper we demonstrate that divergence compensation, optimization of the optical-to-optical efficiency, and high beam quality of signal and idler beams of a high-energy mid-infrared ZGP OPO could be achieved by use of a Galilean telescope inside the nonplanar FIRE ring resonator. With a small variation of the distance between the lenses of the telescope, the divergence of signal and idler beams could be adjusted. Up to 36 mJ of mid-infrared pulse energy in the 3-5 µm wavelength range is obtained with 92 mJ of pump energy on crystal. The beam quality factor M2 is < 1.5 for the resonant signal beam and the non-resonant idler beam, respectively. This is an improvement of the beam quality by approximately a factor 3 for the signal beam and ~2.7 for the idler beam compared without using a telescope inside the FIRE ring resonator.
2. Experimental setup
The experimental setup is shown in Fig. 1. It consists of a Ho3+:LLF MOPA pump laser , a half-wave plate and a polarizer (allow for attenuating the pump energy without altering the pulse duration and pump beam profile), a telescope to adapt the pump spot size and the monolithic FIRE resonator.
The Ho3+:LLF MOPA pump laser delivers 30 ns long pulses (FWHM) with a pulse energy of up to 180 mJ at a wavelength of 2.053 μm at 100 Hz repetition rate and has a beam quality factor M2 < 1.1. The pump beam was adjusted with a telescope (L1 and L2) and produced a slightly elliptical pump profile with pump beam diameter (FWe−2M) of 5.1 mm × 6.6 mm in x- and y-direction, respectively. The ZGP crystal has a size of 9 × 9 × 16 mm3 and was cut at 56° with respect to the optical axis, which allows type I phase matching. We estimate that less than 1% of the pump energy is clipped by the crystal and a possible diffraction effect is of minor importance. The crystal holder is transversely centered on the pump beam and adjusted in a phase-matching angle for maximum OPO output pulse energy, resulting in a signal wavelength around 3.85 μm and an idler around 4.45 μm. The FIRE resonator with a length of L = 222 mm consists of six flat mirrors in a nonplanar arrangement, providing a fractional image rotation per round-trip of 77.5°. The incident angle for the resonating signal on all six mirrors is 32.7°, identical to the corresponding incidence angle in a RISTRA cavity. The output coupler OC of the FIRE resonator has a reflectivity of 50% for the signal and high transmission for the pump (T >95%) and idler (T > 95%) wavelengths. The input coupler IC and the other four mirrors are identical and have high transmission for the pump (T = 83.6%) and idler (T > 94%), and are highly reflective for the signal (R > 99%). We used an uncoated zero order MgF2 half-wave plate (λ = 3.75 µm) which was turned to minimize the OPO threshold. The Galilean telescope inside the FIRE resonator is formed by a CaF2 bi-concave lens L3 (Thorlabs LD5788-E) with a focal length of −25 mm and a BaF2 plano-convex lens L4 (Thorlabs LA0259-E) with a focal length of 40 mm being commercially available with a broadband AR coating optimized for the 3 - 5 μm spectral range deposited on all surfaces. Both lenses have a diameter of 12.7 mm and were mounted in a ∅1/2” lens tube (Thorlabs SM05V05 and SM05L03) where the distance between the lenses could be varied to adjust the telescope to infinity (collimated), slightly convergent or slightly divergent for the resonant signal beam. This adjustment has been done outside the resonator and the lens tube with a length of ~30 mm has been placed then symmetrically to the ZGP crystal in the opposite arm of the FIRE resonator where the distance between adjacent mirrors M1 and M2 is 50 mm. Both lenses form a Galilean telescope with a magnification of 1.6. Simple mode size calculations using ABCD matrix formalism show that gain guiding effects could be compensated resulting in a fundamental mode size of the resonant signal beam which matches the pump spot diameter, as will be discussed in Section 3. The residual pump behind the OPO is filtered out by two dichroic mirrors highly transmissive (T >96%) for the pump and highly reflective for the signal and idler (R > 99%). It should be pointed out, that all six mirrors are fixed inside the monolithic FIRE block and the resonator is adjusted alone via the pump beam. If the cavity is aligned, the telescope is introduced and slightly moved to start oscillation and tightened (afterwards clamped) and final alignment is done again via the pump beam.
3. Results and discussion
The sum of signal and idler pulse energy as a function of pump energy incident on the ZGP crystal is shown in Fig. 2. The highest pulse energy (optical-to-optical efficiency) of 36.4 mJ (39.7%) was obtained when the telescope was adjusted slightly convergent, followed by 35.8 mJ (39.3%), 35.4 mJ (38.65) and 31.8 mJ (35.1%) with telescope collimated, without telescope and with the telescope slightly aligned divergent, respectively. Threshold was lowest in the configuration without telescope at 21 mJ, followed by 24 mJ, 24.4 mJ and 27 mJ in the configuration with telescope convergent, collimated and divergent, respectively. The higher threshold could be explained by higher round trip losses due to the insertion of the telescope. It seems to be obvious that the configuration with the slightly divergent telescope increases the diffraction losses for the resonant signal beam at the crystal entrance resulting in the highest threshold and lowest slope efficiency of 58.4%. By slightly varying the distance between the lenses of the telescope the phase distribution of the resonant signal beam can be adjusted to the phase distribution of the pump beam which allows an optimization of the efficiency of the nonlinear interaction inside the ZGP crystal.
The beam quality factor of the OPO output, shown in Figs. 3 and 4 for signal and idler, respectively, was determined separately along the transverse x- and y-axis by imaging the near and far field of the OPO output beam onto a PyroCam III camera, extracting the beam size and divergence, and calculating the corresponding M2-value. Based on tests with background subtraction of the raw data from the PyroCam III camera, we estimate an uncertainty of at most 5% in the M2-value. Without the telescope, the M2 of the resonant signal beam increases from ∼2.5 near threshold to ∼4.5 at OPO pulse energy of 30 mJ. The experiments show that the signal beam quality could be considerable improved with the telescope. Here, the M2-factor increases from ~1.1 near the threshold to 1.5 at high pulse energy and does not depend on the exact tuning of the telescope (slightly divergent or convergent or collimated). The signal far field shows a Gaussian-like intensity profile (lower inset in Fig. 3) and a rotationally symmetric like intensity distribution which, we assume, is due to the image rotation of the resonant signal beam after each round trip averaging over pump inhomogeneities. The beam quality of the idler beam at high pulse energy is M2 ∼4 without telescope and is improved to M2 ∼1.5 with the telescope. The lower inset in Fig. 4 shows a Gaussian-like far field intensity profile for the idler beam with telescope, however, the profile has an elliptical contour which is clearly due to the astigmatic pump spot.
In order to present a qualitative theoretical interpretation of the mode behavior without and with telescope, as illustrated in Figs. 3 and 4, Fig. 5 shows the results of mode size calculations for the resonant signal beam in the FIRE ring cavity using laser cavity analysis and design (LASCAD) software . In , we have reported that the M2-factor of the resonant signal beam increases from ∼2.5 near threshold to ∼4.5 at OPO pulse energy of 30 mJ which could be attributed to a quadratic variation of the gain parameter α2 inside the ZGP crystal. With α2 = 0.025 mm−3, the mode size of the multimode signal beam (M2 = 4.5) matches the pump mode size diameter (Fig. 5(a)) of ~5.8 mm and the size of the fundamental mode decreased to 5.8 mm · 4.5-1/2 ~2.7 mm (Fig. 5(b)). If the Galilean telescope is introduced, this simple calculation shows that the size of the fundamental mode of the signal beam is adapted again to the pump mode diameter (Fig. 5(c)). Hence, it is no longer possible for the higher order signal modes to be excited in the crystal.
Additional support to the measurements is provided by the simulation code OPODESIGN  basing on the solution of the paraxial wave equation, e.g , by a Fast Fourier Transform (FFT) method. It propagates the signal, idler and pump waves in a ring cavity with image rotation. A Galilean telescope for image magnification can be activated, the propagation of the waves through the lenses with realistically short focal lengths (e.g. f1 = −25 mm, f2 = + 40 mm) being numerically handled by a Talanov-transformation, see  and the reference given herein. For the crystal, the split-step method  is applied integrating the nonlinear OPO-equations  by a fourth order Runge-Kutta scheme. The spatial walk-off effect is taken into account . The temporal evolution of the pump pulse is realized by sampling slices of round-trip time duration. Computations are done for an area of 25.6 mm × 25.6 mm being sampled with a resolution of 29 × 29 mesh points. In the simulations, the elliptic pump spot profile has been taken into account, as shown in the inset of Fig. 1. Further, a crystal thermal lens with a focal length of 12.5 m has been assumed (asymmetry and aberrations of a real thermal lens are neglected). Figure 6 shows the measured (left hand side) and simulated (right hand side) signal and idler beam diameters versus distance from the ZGP crystal as obtained for the three different telescope tunings: (a) slightly adjusted divergent, (b) collimated and (c) convergent for the resonant signal beam. Actually, the simulation has been done with the following distances between the telescope lenses: (a) 14.75 mm, (b) 15 mm, and (c) 15.25 mm. The pump energy incident on the ZGP crystal was 50 mJ. For the graphical output of the simulations, an area of 12.4 mm × 12.4 mm with 124 × 124 mesh points was adapted identically to the camera chip of the Pyrocam III allowing the use of Spiricon software to determine the beam diameters via the 4 sigma method. In the insets, the peak fluences of the simulations are scaled to the peak fluences of the measurements.
It can be clearly realized that the divergent character of the signal beam is transposed to a convergent idler beam and vice versa. This can be explained by the interaction of signal and idler beam due to the nonlinear behavior of the crystal. When stepwise passing the crystal, the increase of the idler field is proportional to the pump field and the complex conjugate of the signal field; see the OPO equations in . Hence, the defocusing (focusing) effect of the telescope is transformed by the crystal via the signal wave to a focusing (defocusing) behavior of the idler wave. In principal, the simulations show qualitative agreement with the measurements. The diameters of signal and idler beam in the simulations are little higher than the measured ones. The calculated energy of signal and idler is 14 mJ and therefore higher than the measured 10 mJ at pump energy of 50 mJ. Increasing the losses for the signal beam in the simulations would result in less efficiency and smaller spot sizes. However, this is not the scope of the present investigations. Further, the simulations show that there is an optimum with respect to the achievable pulse energy when the telescope is slightly adjusted convergent for the signal beam which is in agreement with our observations.
In conclusion, we have demonstrated that using a Galilean telescope inside a FIRE ZGP OPO improves the beam quality considerably by a factor of 3 for the signal and idler beam. With a small variation of the distance between the lenses of the telescope, the divergence of signal and idler beams could be adjusted. We found an optimum in optical-to-optical efficiency, if the telescope is slightly adjusted convergent for the resonant signal beam. The efficiency is slightly better than without using the telescope. Up to 36 mJ of mid-infrared pulse energy in the 3-5 µm wavelength range is obtained with 92 mJ of pump energy on crystal. The beam quality factor M2 is < 1.5 for the resonant signal beam and the non-resonant idler beam, respectively. Results from a numerical simulation tool are in qualitative agreement with our experimental observations. Future experiments are devoted to the construction of a telescope with tunable distance between the lenses during lasing operation of the OPO in order to optimize the efficiency and divergence of signal and idler beam and maintaining the good beam quality at the same time.
References and links
1. A. Dergachev, D. Armstrong, A. Smith, T. Drake, and M. Dubois, “3.4-µm ZGP RISTRA nanosecond optical parametric oscillator pumped by a 2.05-µm Ho:YLF MOPA system,” Opt. Express 15(22), 14404–14413 (2007). [CrossRef] [PubMed]
2. C. Kieleck, M. Eichhorn, A. Hirth, D. Faye, and E. Lallier, “High-efficiency 20-50 kHz mid-infrared orientation-patterned GaAs optical parametric oscillator pumped by a 2 µm holmium laser,” Opt. Lett. 34(3), 262–264 (2009). [CrossRef] [PubMed]
3. E. Lippert, H. Fonnum, G. Arisholm, and K. Stenersen, “A 22-watt mid-infrared optical parametric oscillator with V-shaped 3-mirror ring resonator,” Opt. Express 18(25), 26475–26483 (2010). [CrossRef] [PubMed]
4. M. Eichhorn, G. Stoeppler, M. Schellhorn, K. T. Zawilski, and P. G. Schunemann, “Gaussian- versus flat-top pumping of a mid-IR ZGP RISTRA OPO,” Appl. Phys. B 108(1), 109–115 (2012). [CrossRef]
5. G. Stöppler, M. Schellhorn, and M. Eichhorn, “Enhanced beam quality for medical applications at 6.45 µm by using a RISTRA ZGP OPO,” Laser Phys. 22(6), 1095–1098 (2012). [CrossRef]
6. M. W. Haakestad, H. Fonnum, and E. Lippert, “Mid-infrared source with 0.2 J pulse energy based on nonlinear conversion of Q-switched pulses in ZnGeP2.,” Opt. Express 22(7), 8556–8564 (2014). [CrossRef] [PubMed]
7. M. Eichhorn, M. Schellhorn, M. W. Haakestad, H. Fonnum, and E. Lippert, “High-pulse-energy mid-infrared fractional-image-rotation-enhancement ZnGeP2 optical parametric oscillator,” Opt. Lett. 41(11), 2596–2599 (2016). [CrossRef] [PubMed]
8. P. G. Schunemann, K. T. Zawilski, L. A. Pomeranz, D. J. Creeden, and P. A. Budni, “Advances in nonlinear optical crystals for mid-infrared coherent sources,” J. Opt. Soc. Am. B 33(11), D36–D43 (2016). [CrossRef]
9. A. Hemming, J. Richards, A. Davidson, N. Carmody, S. Bennetts, N. Simakov, and J. Haub, “99 W mid-IR operation of a ZGP OPO at 25% duty cycle,” Opt. Express 21(8), 10062–10069 (2013). [CrossRef] [PubMed]
10. K. T. Zawilski, S. D. Setzler, P. G. Schunemann, and T. M. Pollak, “Increasing the laser-induced damage threshold of single-crystal ZnGeP2,” J. Opt. Soc. Am. B 23(11), 2310–2316 (2006). [CrossRef]
11. B. C. Johnson, V. J. Newell, J. B. Clark, and E. S. McPhee, “Narrow-bandwidth low-divergence optical parametric oscillator for nonlinear frequency-conversion applications,” J. Opt. Soc. Am. B 12(11), 2122–2127 (1995). [CrossRef]
12. S. Pearl, Y. Ehrlich, S. Fastig, and S. Rosenwaks, “Nearly diffraction-limited signal generated by a lower beam-quality pump in an optical parametric oscillator,” Appl. Opt. 42(6), 1048–1051 (2003). [CrossRef] [PubMed]
13. S. Zou, M. Gong, Q. Liu, and G. Chen, “Low threshold characteristic of pulsed confocal unstable optical parametric oscillators with Gaussian reflectivity mirrors,” Opt. Express 13(3), 776–788 (2005). [CrossRef] [PubMed]
14. A. V. Smith and D. J. Armstrong, “Nanosecond optical parametric oscillator with 90° image rotation: design and performance,” J. Opt. Soc. Am. B 19(8), 1801–1814 (2002). [CrossRef]
15. S. Bigotta, G. Stöppler, J. Schöner, M. Schellhorn, and M. Eichhorn, “Novel non-planar ring cavity for enhanced beam quality in high-pulse-energy optical parametric oscillators,” Opt. Mater. Express 4(2), 411–423 (2014). [CrossRef]
16. M. Schellhorn, G. Spindler, and M. Eichhorn, “Improvement of the beam quality of a high-pulse-energy mid-infrared fractional-image-rotation-enhancement ZnGeP2 optical parametric oscillator,” Opt. Lett. 42(6), 1185–1188 (2017). [CrossRef] [PubMed]
17. M. Schellhorn and M. Eichhorn, “High-energy Ho:LLF MOPA laser system using a top-hat pump profile for the amplifier stage,” Appl. Phys. B 109(2), 351–357 (2012). [CrossRef]
18. LASCAD, LAS-CAD GmbH, http://www.las-cad.com.
19. A. E. Siegman, Lasers (University Science Books, Sausalito, CA, 1986).
20. J. A. Fleck Jr, J. R. Morris, and M. D. Feit, “Time-dependent propagation of high energy laser beams through the atmosphere,” Appl. Phys. (Berl.) 10(2), 129–160 (1976). [CrossRef]
21. G. Arisholm, “General numerical methods for simulating second-order nonlinear interactions in birefringent media,” J. Opt. Soc. Am. B 14(10), 2543–2549 (1997). [CrossRef]
22. J. A. Fleck Jr and M. D. Feit, “Beam propagation in uniaxial anisotropic media,” J. Opt. Soc. Am. 73(7), 920–926 (1983). [CrossRef]