All-solid-state parametric Raman anti-Stokes laser at 508 nm

We report a parametric anti-Stokes Raman laser using potassium gadolinium tungstate, generating output chiefly at the first anti-Stokes at 508 nm. The compact 4.5 cm long device is pumped by a Q-switched 532 nm laser and uses an off-axis Stokes resonator to provide non-collinear phase matching between the pump and the generated Stokes and anti-Stokes fields. Anti-Stokes output energies up 0.27 mJ were obtained at a conversion efficiency from the pump of 0.46%. Secondand third-order anti-Stokes lines at 486 nm and 465 nm were also observed. ©2008 Optical Society of America OCIS codes: (140.3550) Lasers and laser optics: Lasers, Raman, (190.4410) Nonlinear optics: parametric processes. References and links 1. J. A. Piper and H. M. Pask, “Crystalline Raman lasers,” IEEE J. Sel. Top. Quantum Electron. 13, 692-704 (2007). 2. T. T. Basiev and R. C. Powell, “Solid-state Raman lasers,” in Handbook of Laser Technology and Applications, C. E. Webb et al., eds. (Institute of Physics, UK, 2003), pp. 469–497. 3. H. M. Pask, “The design and operation of solid-state Raman lasers,” Prog. Quantum Electron. 27, 1–56 (2003). 4. P. Cerny, H. Jelinkova, P. G. Zverev, and T. T. Basiev, “Solid state lasers with Raman frequency conversion,” Prog. Quantum Electron. 28, 113–143 (2004). 5. R. Chiao and B. P. Stoicheff, “Angular dependence of maser-stimulated Raman radiation in calcite,” Phys. Rev. Lett. 12, 290-293 (1964). 6. A.A. Kaminskii, K. Ueda, H. J. Eichler, Y. Kuwano, J. Kouta, S. N. Bagaev, T. H. Chyba, J. C. Barnes, G. M. A. Gad, T. Murai, and J. Lu, “Tetragonal vanadates YVO4 and GdVO4 – new efficient χ(3)-materials for Raman lasers,” Opt. Commun. 194, 201-206 (2001). 7. A. A. Kaminskii, H. J. Eichler, K. Ueda, N. V. Klassen, B. S. Redkin, L. E. Li, J. Findeisen, D. Jaque, J. García-Sole, J. Fernández, and R. Balda, “Properties of Nd-doped and undoped tetragonal PbWO4, NaY(WO4)2, CaWO4, and undoped monoclinic ZnWO4 and CdWO4 as laser-active and stimulated Raman scattering-active crystals,” Appl. Opt. 21, 4533-4547 (1999). 8. A. K. McQuillan, W. R. L. Clements, and B. P. Stoicheff, “Stimulated Raman emission in diamond: Spectrum, gain, and angular distribution of intensity,” Phys. Rev. A 1, 628-635 (1970). 9. A. A. Kaminskii, C. L. McCray, H. R. Lee, S. W. Lee, D. A. Temple, T. H. Chyba, W. D. Marsh, J. C. Barnes, A. N. Annanenkov, V. D. Legun, H. J. Eichler, G. M. A. Gad, and K. Ueda “High efficiency nanosecond Raman lasers based on tetragonal PbWO4 crystals,” Opt. Commun. 183, 277-287 (2000). 10. S. O. Konorov, E. E. Serebryannikov, A. M. Zheltikov, P. Zhou, A. P. Tarasevitch, and D. von der Linde, “Generation of femtosecond anti-Stokes pulses through phase-matched parametric four-wave mixing in a photonic crystal fibre,” Opt. Lett. 29, 1545-1547 (2004). 11. C. Reiser, T. D. Raymond, R. B. Michie, and A. P. Hickman, “Efficient anti-Stokes Raman conversion in collimated beams,” J. Opt. Soc. Am. B 6, 1859–1869 (1989). 12. A. Z. Grasyuk, L.L. Losev, A. P. Lutsenko, and S. N. Sazonov, “Raman parametric generation of antiStokes radiation under conditions of amplification of an external Stokes signal,” Sov. J. Quantum Electron. 20, 529-532 (1990). 13. A. Z. Grasiuk, S. V. Kurbasov, and L. L. Losev, “Picosecond parametric Raman laser based on KGd(WO4)2 crystal,” Opt. Commun. 240, 239-244 (2004).. 14. R. B. Andreev, V. A. Gorbunov, S. S. Gulidov, S. B. Papernyl, and V. A. Serebryakov, “Role of parametric effects in generation of higher components of stimulated Raman scattering in gases,” Sov. J. Quantum Electron. 12, 35-37 (1982). 15. A. Z. Grasyuk, L. L. Losev, A. P. Lutsenko, and S. N. Sazonov, “Parametric Raman anti-Stokes laser,” Sov. J. Quantum Electron. 20, 1153-1155 (1990). #103202 $15.00 USD Received 24 Oct 2008; revised 16 Nov 2008; accepted 17 Nov 2008; published 9 Jan 2009 (C) 2009 OSA 19 January 2009 / Vol. 17, No. 2 / OPTICS EXPRESS 810 16. P. A. Roos, L. S. Meng, S. K. Murphy, and J. L. Carlsten, "Approaching quantum-limited cw anti-Stokes conversion through cavity-enhanced Raman-resonant four-wave mixing," J. Opt. Soc. Am. B 21, 357-363 (2004). 17. R. W. Boyd, Nonlinear Optics (Academic Press, Inc., San Diego, 1992). 18. M. C. Pujol, M. Rico, C. Zaldo, R. Sole1, V. Nikolov, X. Solans, M. Aguilo, and F. Dıaz, “Crystalline structure and optical spectroscopy of Er-doped KGd(WO4)2 single crystals,” Appl. Phys. B 68, 187–197 (1999). 19. I. V. Mochalov, “Laser and nonlinear properties of the potassium gadolinium tungstate laser crystal KGd(WO4)2:Nd (KGW:Nd),” Opt. Eng. 36, 1660-1669 (1997). 20. V. V. Ermolenkov, V. A. Lisinetskii, Y. I. Mishkel', A. S. Grabchikov, A. P. Chaikovskii, and V. A. Orlovich, “A radiation source based on a solid-state Raman laser for diagnosing tropospheric ozone,” J. Opt. Tech. 72, 32-36 (2005). 21. For example: R. Wallenstein, “High power all-solid-state laser source for direct-write large-screen laser projection displays,” in Proceeding of the Lasers and Electro-Optics Society 12th Annual Meeting, 158-159 (1999). 22. Y. Chen, W. Hou, H. Peng, H. Zhang, L. Guo, H. Zhang, D. Cui, and Z.Y. Xu, “Intracavity frequency doubling of an active Q-switched Nd:YAG laser with 2.25 W output power at 473 nm,” Opt. Commun. 270, 58-62 (2007). 23. S. H. Ding, X. Y. Zhang, Q. P. Wang, F. F. Su, S. T. Li, S. Z. Fan, Z. J. Liu, J. Chang, S. S. Zhang, S. M. Wang, and Y. R. Liu, “Theoretical and experimental research on the multi-frequency Raman converter with KGd(WO4)2 crystal,” Opt. Express 13, 10120-10128 (2005), http://www.opticsinfobase.org/abstract.cfm?URI=oe-13-25-10120. 24. S. N. Karpukhin and A. I. Stepanov, “Generation of radiation in a resonator under conditions of stimulated Raman scattering in Ba(NO3)2, NaNO3, and CaCO3 crystals,” Sov. J. Quantum Electron. 16, 1027-1031 (1986). 25. R. P. Mildren, M. Convery, H. M. Pask, J. A. Piper, and T. McKay, “Efficient, all-solid-state, Raman laser in the yellow, orange and red,” Opt. Express 12, 785-790 (2004), http://www.opticsinfobase.org/abstract.cfm?URI=oe-12-5-785. 26. C. He and T. H. Chyba, “Solid-state barium nitrate Raman laser in the visible region,” Opt. Commun. 135, 273-278 (1997).


Introduction
Stimulated Raman scattering in crystals forms the basis of laser wavelength conversion to wavelengths not easily generated via other means.Though Stokes generation has been the subject of numerous papers covering a wide variety of wavelengths and temporal formats [1][2][3][4], generation of anti-Stokes has received much less attention.Almost all work to date has centered on anti-Stokes generation using sub-nanosecond lasers focused into the Raman medium, offering little control of spatial or spectral characteristics of the output [5][6][7][8][9].Anti-Stokes generation from ultrafast laser pulses in photonic crystal fibers is an alternative approach that is also receiving attracted significant attention [10].Of much more practical interest are converters of pulsed nanosecond lasers to an anti-Stokes beam of high beam quality.Compact anti-Stokes lasers are particularly interesting for generating blue laser output from the second harmonic of neodymium lasers for applications such as colour holography, optical countermeasures and laser display.
In contrast to Stokes generation, anti-Stokes generation is a four-wave mixing process.The conical Stokes and anti-Stokes output often observed when propagating a high intensity pump beam through a Raman medium is a consequence of phase matching requirements of four wave mixing, with the output cone angles relating directly to the dispersion in the Raman medium (see e.g.[5,8]).Low divergence anti-Stokes output is achievable by pumping with both a pump and a separately-generated Stokes beam, ensuring that both beams are wellcollimated and propagate at the required phase matching angle.This has been demonstrated for picosecond pump pulses in high pressure hydrogen [11,12] and in a potassium gadolinium tungstate (KGW) crystal [13]; however, the complexity of these ultrashort pulsed and twogain-element systems is not well suited to most applications.
A more simple approach is to generate a Stokes and anti-Stokes beam from a nanosecond pump laser within a single Raman medium.On-axis generation of anti-Stokes with conversion efficiency up to 6% and beam quality factor M 2 ~26 have been observed in a 2.3 m-long hydrogen cell pumped using 30 ns-long pulses at 532 nm [14].Much higher beam quality has been obtained, however, using a so-called parametric Raman anti-Stokes laser.A resonator with its axis tilted from the pump axis generates a Stokes beam at the predetermined angle for phase-matching of anti-Stokes generation [15].Using high-pressure hydrogen gas as the Raman medium and a 530 nm pulsed pump laser, a peak (instantaneous) conversion efficiency of 6% and overall pulse conversion efficiency of 2% to the 508 nm first anti-Stokes was obtained with near diffraction limited beam quality.Conversion efficiencies as high as several tens of percent were predicted for improved divergences of the pump and Stokes beams [16].
Achieving a parametric Raman anti-Stokes laser using a solid-state medium is an important step towards realizing compact and rugged wavelength up-converters.Crystalline Raman media also have higher thermal conductivity and Raman gain coefficient than gaseous counterparts, which allow the converters to be made much smaller and at higher average powers.However, we are unaware of any reports of solid-state parametric anti-Stokes Raman lasers operating in the nanosecond regime to date.
Here we report a parametric anti-Stokes Raman laser in KGW pumped by a commercial Q-switched 532 nm laser.We observe generation of up to 3 anti-Stokes orders (508 nm, 486 nm and 465 nm) with first anti-Stokes pulse energies up to 0.27 mJ at 0.46% conversion efficiency from the pump.The spectral, spatial and temporal characteristics of the output are presented and we discuss methods for increasing the conversion efficiency.ω ω ω − = so that the mixing process is resonant with the crystal vibration, and we access the resonantly-enhanced Of course other values of j, k, and l describe processes that will take place, but only those that are phase matched sufficiently closely will have appreciable yield.In this paper, we are interested in the parametric generation of anti-Stokes radiation 1 E + from a pump field 0 E and Stokes field 1 This process can generate a substantial amount of anti-Stokes radiation if we provide a strong field at the Stokes and fundamental wavelengths, and simultaneously satisfy the phase matching condition

Starting
Owing to dispersion in Raman crystals, this condition is not generally satisfied for collinear beams, but can easily be satisfied if a small angle is introduced between the Stokes and fundamental waves.Once a significant field has been generated at the anti-Stokes wavelength, the process can cascade, with parametric generation of second anti-Stokes 2 In this paper, we focus on an external cavity Raman laser using the Raman-active crystal KGW.The pump wavelength of 532 nm propagating along the N P axis generates a strong resonated Stokes field at 559 nm by the 901 cm -1 Raman shift.By tilting the resonator axis relative to the propagation direction of the pump radiation, we introduce a small angle between these two fields in order to phase match for first anti-Stokes generation (see in Fig. 1 below).For beams polarised parallel to the KGW N m axis, the angle required to phase-match the anti-Stokes process of Eq. ( 1.3) can be calculated using the Sellmeier coefficients measured by Pujol et al [18]: this angle is calculated using the cosine rule to be 25.7 mrad (internal angle), with the anti-Stokes radiation expected to be generated at -23.2 mrad.The tolerance on this angle is very tight: for a crystal of length l = 2.5 cm, the phase matching will be completely lost during one pass for Δk = π/l = 125 m -1 , which corresponds to an angular tolerance of just 0.11 mrad.For comparison, the angular tolerance for type-I frequency-doubling of 1064 nm radiation in a 2 cm long KTP crystal is 5 mrad.The anti-Stokes generation can in principle proceed with high efficiency in a pulsed Raman laser.In a Raman resonator with single pass pumping and resonated Stokes, the phase matching condition for anti-Stokes is met only with all beams co-propagating with the pump beam, and so anti-Stokes can be generated only by the forward-propagating Stokes field inside the resonator.SRS on the other hand occurs also for counter-propagating pump and Stokes beams, and so the backward-propagating Stokes field sees SRS gain.We can therefore consider the forward pass to redistribute energy between the fields, creating the anti-Stokes field while simultaneously amplifying the Stokes field and depleting the pump field [15,17]; the Stokes field is also amplified on its backwards pass.Simulations by Grasyuk et al. [15] conclude that in theory, conversion efficiencies of order 30% are possible, although in practice, they found that efficiency was strongly limited by the poor beam quality of the pump and of the generated Stokes cavity field.Only a fraction of the Stokes and pump beam are in the proper alignment, and so the total possible efficiency is immediately curtailed.For a crystalline Raman medium, the angular tolerance is similarly strict, and so these limitations will also apply here.

Experiment
The parametric anti-Stokes Raman laser, shown in Fig. 2 N p axis, polarized in the vertical plane parallel to the N m axis of the KGW crystal in order to access its strong 901.5 cm -1 Raman mode (stationary gain coefficient at 1064 nm of g 0 = 6 cm/GW [19]).A Stokes beam was generated at an angle θ -1 offset from the pump axis in either the vertical and horizontal planes by placing tilted resonator mirrors M1 and M2 adjacent the crystal as shown in Fig. 2. M1 and M2 are identical plano-concave mirrors of radius of curvature 2 m with the planar side anti-reflection coated in the visible and the curved side having dichroic coating highly reflective at 550 -650 nm and highly transmitting for the pump wavelength and anti-Stokes wavelengths (see transmission spectrum in Fig. 3).Note that to obtain highest anti-Stokes output one of the mirrors was placed with the plane surface closest to the Raman crystal.Initial characterizations undertaken with the more stable concave -concave configuration were found to yield only approximately two-thirds of the conversion efficiency.Investigation of performance as a function of θ -1 up to angles of 100 mrad was easily achieved by varying the tilt of M1 and M2 using standard mirror mounts.The mirrors were placed as close as practicable to the Raman crystal forming a wavelength converter of overall length 4.5 cm.The output was characterized for the θ -1 value that yielded maximum anti-Stokes output, using a collimated pump beam with a radius of either 0.5 or 1.0 mm.The walk-off between the pump and Stokes beams (θ -1 l / n m = 0.55 mm) over the length of the crystal l, was thus sufficiently small to allow anti-Stokes generation over most of the crystal length for both pump beam diameters.
The output wavelengths were measured using a fiber spectrometer (Ocean Optics, USB2000), pulse shapes of the individual components were measured (after dispersing the collinear Stokes beams using a 1200 line/mm grating) using a fast-photodiode (Thorlabs, DET2-SI) and a 500 MHz digital storage oscilloscope (Tektronix TDS3054).

Results
Anti-Stokes generation was investigated for pump energies up to 60 mJ in a 1.0 mm pump beam radius, which corresponds to peak power densities of approximately 150 MW/cm 2 .We initially aligned the Stokes resonator parallel to the pump beam and gradually increased the angle in the horizontal plane.No anti-Stokes output was observed until θ -1 was increased to approximately 45 mrad, and maximum 1 st anti-Stokes output was obtained at 58.0 (± 1.6) mrad, which corresponds to an internal crystal angle θ ' -1 = 28.1 (± 0.8) mrad.For pump energies well above the threshold for Stokes generation, up to three anti-Stokes orders were observed at 508 nm, 486 nm and 465 nm as shown in the output spectrum of Fig. 3.The anti-Stokes beams emerged at angles spaced by 40-50 mrad and were incident on a screen positioned 450 mm from the exit face of the KGW crystal; the output pattern is shown in Fig. 4a, and the measured beam positions listed in Table 1.Note that all the Stokes orders were collinear, with the axis defined by the resonator.The polarization of each output beam was parallel to the pump laser.The dominant production channel for the 2 nd and 3 rd anti-Stokes is assumed to be four wave mixing of the pump, Stokes, and 1 st or 2 nd anti-Stokes respectively.The phase-matching diagram for the 2 nd anti-Stokes is shown in Fig. 4(b); note that with the resonator angle optimized for producing first anti-Stokes (thus fixing three of the four vectors), the 2 nd anti-Stokes process retains a small phase mismatch Δk that will limit the efficiency of higher order anti-Stokes generation.We calculated theoretical phase match angles using Eq.(1.4), and inferred the expected positions of the beams on the screen, also shown in fairly consistent with the observations, with minor disagreement attributed to departures in the dispersion of our KGW crystal from the published Sellmeier coefficients [18].
At approximately 10 mJ of pump energy, the Stokes and anti-Stokes output both rose above threshold, and output energy of the first anti-Stokes increased with pump energy as shown in Fig. 5 (Curve A).The rate of increase is approximately quadratic for pump energies in the range 10 -40 mJ and approximately linearly thereafter.The maximum output energy was 0.27 mJ at 0.46% conversion efficiency from the pump.The 2 nd anti-Stokes output energy was much lower with 7 μJ observed at maximum pump energy (conversion efficiency 0.012%).The 3 rd anti-Stokes beam was visible but below the measurement resolution of the energy meter (< 1μJ).
Pump Energy (mJ) We also investigated anti-Stokes generation for the Stokes resonator inclined in the vertical (N p -N m ) plane.We observed a similar threshold to the horizontal inclined resonator (10-12 mJ) and at the maximum pump energy of 60 mJ the output energy was 0.20 mJ, about 75% of that obtained in the horizontal plane (Curve B, Fig. 5).The output beam pattern was essentially identical to that observed in Fig. 4a rotated through 90°.Note that the output beams were again vertically polarized, but due to the inclined propagation direction there was a component of the Stokes polarization in the N p direction.Thus a lower conversion efficiency may be expected due to reduced effective nonlinear coupling (that contains a smaller offdiagonal χ (3) element).Higher conversion efficiency was obtained using a smaller 0.5 mm pump beam diameter, though the maximum output energy of 55 μJ was limited by optical damage to the input mirror dichroic coating (Curve C, Fig. 5).The threshold for anti-Stokes was 4 mJ, and output increased much more linearly than for the 2.0 mm pump diameter.At maximum output, the conversion efficiency was 0.45%.In all the above cases, the divergence of the Stokes beam was much higher than the pump and anti-Stokes beams.We suggest that this was the dominant factor limiting the conversion efficiency, as it was also in the hydrogen anti-Stokes laser work by Grasyuk et al. [15].  2 for the case of the Stokes resonator inclined in the vertical (y-z) plane (so output pattern rotated by 90° compared to Fig. 4a).The pump and Stokes were symmetrical with the M 2 values 2.5 and 38 respectively.In contrast, the anti-Stokes was highly asymmetrical with M y 2 = 3.3 and M x 2 approximately 7.5 times higher.The much larger divergence for the Stokes compared to the anti-Stokes in the vertical plane suggests that only a narrow cone of Stokes rays fall within the angular acceptance for phase-matching.The cause of the asymmetry and the prospects for increasing conversion efficiency by decreasing the Stokes divergence is discussed in more detail below.(t = 0), with the onset of 1 st anti-Stokes approximately 1 ns thereafter.The pulse duration of the anti-Stokes was 4 ns at full-width-half-maximum.The pulse shape and timing of the 2 nd anti-Stokes is essentially identical to the first anti-Stokes, though at much lower intensity.Note that the anti-Stokes pulse peaks slightly after that of the Stokes, pump and depleted pump beams: we attribute this to the dependence of the anti-Stokes generation on the spatial evolution of the Stokes field in the resonator.

Discussion and conclusions
The output power demonstrated herein is currently too low for many applications; efficiency is approximately a factor of 5 times lower than observed in hydrogen [15] and two orders of magnitude below the theoretical maximum.The major cause of the low efficiency is attributed to partial phase matching owing to the very large divergence of the first Stokes beam.From the measured M 2 value (Table 2), the Stokes beam divergence is 6.4 mrad, approximately 15-times the divergence of the pump beam.This is a consequence of the stability and high Qfactor of the resonator.Since the angular acceptance is much smaller than both the divergence of the pump and first-Stokes beams, it is deduced that at most ~1/15 th of the Stokes light can be available for conversion to the anti-Stokes.An additional major energy diversion from the anti-Stokes is by the cascade loss of energy from the 1 st Stokes to up to four observed higher Stokes orders.Improved spatial properties of the Stokes beam and the mitigation of higher Stokes order can be both addressed through improved Stokes resonator design, i.e. via the mirror spectral characteristics, curvatures (e.g.unstable resonators [20]) and/or variable reflectivity profile.By resonator optimization, it is expected that conversion efficiencies more than 20 times higher (i.e.> 10%) might be achieved.
An alternative and potentially interesting method for reducing the constraints on the Stokes beam divergence is by using the birefringent property of KGW to provide phase matching for collinearly-propagating beams.Though such schemes would rely on off-diagonal components of the χ (3) tensor, which are not well known, it is possible in principle to greatly relax the dependence of efficiency on the beam quality and collimation of the beams, as well as allow tighter focusing without sacrificing spatial overlap.
The above results demonstrate upconversion of a pump laser to anti-Stokes wavelengths in an all-solid-state and compact device.The device offers a simple approach to conversion of a 532 nm pump laser to one or more blue wavelengths, which may have practical advantages over schemes such as nonlinear sum-frequency generation plus an optical parametric oscillator [21] or by second-harmonic generation of the 0.94 μm Nd line [22].Finally, it is worth pointing out that the anti-Stokes parametric Raman laser is well suited to the generation of three or more wavelengths simultaneously, of key interest for blue -green -red multiwavelength applications.The cascade of energy into the Stokes orders can be readily controlled via the output coupler spectral characteristics as discussed previously in numerous refs (e.g.[23]) and conversion efficiencies up to 60% have been reported (e.g.[24][25][26]).Thus, including the residual pump beam, three wavelengths are easily obtained from one pump laser and an anti-Stokes parametric Raman laser converter.
then describe the cascaded SRS processes generating the higher Stokes orders.
, consisted of a 25 mm long KGW crystal of 10x10 mm cross-section, anti-reflection coated for wavelengths 530 -650 nm and pumped by the second-harmonic of a Q-switched Nd:YAG laser of pulse duration approximately 8 ns, and repetition rate of 10 Hz.The pump beam was propagated along the #103202 -$15.00USD Received 24 Oct 2008; revised 16 Nov 2008; accepted 17 Nov 2008; published 9 Jan 2009 (C) 2009 OSA

Fig. 5 .
Fig. 5. Output characteristics of the first anti-Stokes.Curves have pump beam radius and the phase-matching performed in the plane as follows: (A) 1.0 mm; N p -N g plane, (B) 1.0 mm; N p -N m plane and (C) 0.5 mm; N p -N m plane.
Measurements of the beam quality factors for the pump, Stokes and anti-Stokes are compared #103202 -$15.00USD Received 24 Oct 2008; revised 16 Nov 2008; accepted 17 Nov 2008; published 9 Jan 2009 (C) 2009 OSA in Table

Fig. 6 .
Fig. 6.Temporal pulse shapes for the pump, residual pump, first Stokes, first anti-Stokes and second anti-Stokes.The temporal characteristics of the pump and output fields are shown in Fig.6, along with the pulse shape of the depleted 532 nm beam transmitted by the anti-Stokes laser.The onset of Stokes generation occurred near the half-maximum of the leading edge of the pump pulse #103202 -$15.00USD Received 24 Oct 2008; revised 16 Nov 2008; accepted 17 Nov 2008; published 9 Jan 2009 (C) 2009 OSA 19 January 2009 / Vol.17, No. 2 / OPTICS EXPRESS 818

Table 1 .
Measured horizontal positions at a distance of 450 mm from the exit face of the KGW, compared with the calculated values.

Table 1 ,
taking into account the deviation and change in angle caused by the inclined curved resonator mirrors.All calculated angles are #103202 -$15.00USD Received 24 Oct 2008; revised 16 Nov 2008; accepted 17 Nov 2008; published 9 Jan 2009 (C) 2009 OSA

Table 2 .
Measured M 2 values for the pump, Stokes, and anti-Stokes beams.