Comparison of continuous-wave optical parametric oscillators based ...

2102

J. Opt. Soc. Am. B / Vol. 20, No. 10 / October 2003

Stothard et al.

Comparison of continuous-wave optical parametric oscillators based on periodically poled LiNbO3 and periodically poled RbTiOAsO4 pumped internal to a high-power Nd:YVO4 laser David J. M. Stothard, P.-Y. Fortin, Alison Carleton, Majid Ebrahimzadeh, and Malcolm H. Dunn The J. F. Allen Physics Research Laboratories, School of Physics and Astronomy, University of St. Andrews, St. Andrews, Fife, KY16 9SS, Scotland, UK Received November 9, 2002; revised manuscript received May 12, 2003 We present and compare two continuous-wave optical parametric oscillators based on the nonlinear crystals periodically poled LiNbO3 (PPLN) and periodically poled RbTiOAsO4 (PPRTA) pumped internal to a highpower, all-solid-state Nd:YVO4 laser. At a diode pump power of 12 W, similar total extracted idler powers of over 440 mW were obtained from each system even though the nonlinearity of PPRTA is significantly less than that of PPLN. The spatial, power, and frequency stability of the PPRTA-based system outperformed those of the PPLN system. We attribute the poor performance of PPLN in these parameters to its susceptibility to thermal lensing. © 2003 Optical Society of America OCIS codes: 140.3580, 140.3600, 160.4330, 190.4400, 190.4970.

1. INTRODUCTION Continuous wave (cw) optical parametric oscillators (OPOs) have now been established as practical and efficient sources of broadly tunable midinfrared radiation.1–7 Cw, pump-enhanced, singly resonant OPOs (PESROPOs)7–10 and doubly resonant OPOs (DR-OPOs)6,11,12 in particular offer excellent narrow linewidth, singlefrequency output, and, with careful cavity design, continuous fine-tuning characteristics that make them potentially ideal for high-resolution spectroscopic applications.6,10 Their practical field application has, however, been limited by their need for high-quality, single-frequency laser pump sources, complex electronic cavity-stabilizing mechanisms, and a low-mechanicalnoise operating environment, particularly important in the case of the DR-OPO. Externally pumped, singlyresonant OPOs (EPSR-OPOs) eliminate the singlefrequency-pump-source and multifrequency-resonancestability requirements of the PESR-OPO and DR-OPO, but exhibit high OPO threshold [typically 3–10 W1,13 of pump power from a laser which itself needs to be externally pumped], and, unless a single-frequency pump laser is utilized, produce a spectral output less suitable for high-resolution spectroscopy because the multifrequency nature of the pump laser is transferred to the idler. Very low external power thresholds are possible by placing the (singly resonant) OPO within the cavity of a high-finesse pump laser (intracavity SR-OPO) (ICSROPO), thereby taking advantage of the very high circulating field found there.2,4 This technique obviates the high thresholds exhibited by the EPSR-OPO and the veryhigh-cavity stability, single-frequency pump laser and complex experimental arrangement mandated by the DROPO and PESR-OPO. The threshold pump power re0740-3224/2003/102102-07$15.00

quirement of the diode-pumped ICSR-OPO is particularly favorable when a comparison is made between it and the external pump power required to drive the laser that is itself pumping other cw OPO geometries, most notably the EPSR-OPO and, to a lesser extent, the PESR-OPO. Indeed, we have previously demonstrated an intracavity OPO pumped internal to an all-solid-state Nd:YVO4 laser that exhibited OPO threshold at an external-diode pump power of only 310 mW without the need for any external cavity-control electronics, and, at a pump power of 1 W, produced over 70 mW of extractable, multifrequency idler power.4 The advent of quasi-phase matching (QPM) materials has had a profound and widespread effect in the field of nonlinear optics. Periodically poled LiNbO3 (PPLN), in particular, has offered enhanced effective nonlinearities and the flexibility of grating-engineered phase matching. A highly significant result in the application of QPM nonlinear materials of cw OPOs has been the demonstration, through the use of PPLN, of externally pumped cw SROPOs that exhibit thresholds of the order of a few watts, hence placing them within the pumping domain of diodepumped solid-state lasers.1 Although the establishment of QPM as a truly practical technique has largely been due to the successful poling of PPLN, the material is subject to certain significant limitations. The maximum available aperture through which poling can be induced is currently 1 mm; therefore, crystal apertures are ultimately limited to this thickness. The material is also susceptible to undesirable thermal effects, such as thermally induced lensing and phase mismatching, even at modest pump power levels.14 It is also well established that PPLN is subject to photorefractive damage when operated at low temperature. This mandates the use of © 2003 Optical Society of America

Stothard et al.

this material in an oven at high temperature, adding to system complexity and instability, particularly in the case of ICSR-OPO, PESR-OPO, and DR-OPO where thermal air currents can perturb the circulating fields. On the other hand, the phosphate and arsenate families of nonlinear crystals exhibit low susceptibility to these deleterious thermally induced effects, and they do not suffer from photorefractive damage, which permits room-temperature operation. The phosphate and arsenate families of crystals also require significantly less coercive field for the poling effect to occur, which permits the fabrication of wider apertures, typically 3 mm. As such, these crystals are attractive material candidates for use in higher power cw ICSR-OPOs for which the high circulating intensities preclude the use of PPLN because of thermal effects. An obstacle to the exploitation of the periodically poled arsenates and phosphates in externally pumped cw SROPOs is the reduced effective QPM nonlinearity of 8–10 pm/V15 compared with 15–17 pm/V in PPLN.16 In addition, the interaction lengths currently available in these materials are limited by growth to ⬃20 mm, which falls short of those that are now routinely available for PPLN (50 mm). For these reasons, very high oscillation thresholds (⬎10 W) are to be anticipated for cw SR-OPOs based on periodically poled RbTiOAsO4 (PPRTA), which, although within the pump capability range of high-power, cw, diode-pumped, solid-state or fiber lasers, leads to unacceptable efficiency loss if the material is to be pumped in an external SR-OPO configuration. This is particularly evident when one takes into account the fundamental pump power requirements of the pump laser source. However, these obstacles are overcome in cw ICSR-OPOs17 and PESR-OPOs,18 for which the high pump power thresholds can be readily surpassed and practical room-temperature operation can be achieved even in crystals with limited interaction lengths. Although PPRTA has been utilized in the ICSR-OPO configuration before,17 we report here what we believe to be the first demonstration of a cw ICSR-OPO based on this material pumped internal to an all-solid-state laser system. In particular, we compare and contrast the performance of this material with that of PPLN in the context of intracavity devices.

Vol. 20, No. 10 / October 2003 / J. Opt. Soc. Am. B

2103

2. EXPERIMENT The optical layout of the ICSR-OPO is shown schematically in Fig. 1. The pump source for the SR-OPO was a temperature-stabilized, 15-W, fiber-coupled, diode-laser array (SDL part number SDL-3460). The focused diodelaser pump light entered the Nd:YVO4 crystal by way of a plane 1.064-␮m-high reflector that was antireflection coated at the diode pump wavelength. Experimental observation and numerical modeling showed that the thermal lens induced in the gain crystal by the diode pump light gave the plane 1.064-␮m-high reflector an effective radius of curvature of ⬃150–200 mm. The aperture was included to aid alignment but was also useful in suppressing oscillation on higher-order laser modes. Antireflection-coated lens L3 (focal length 63 mm) served to form a pump waist with M2 (200-mm radius of curvature) in the nonlinear crystal, and also desensitized the cavity from thermally induced lensing effects that formed in the nonlinear crystal as the circulating pump field was increased. To facilitate temperature tuning of the signal and idler, and to avoid the effects of photorefractive damage, the PPLN crystal was placed in a servo-controlled oven. The PPRTA was mounted on a thermoelectrically stabilized copper plate. Both crystals were triple-band antireflection-coated at the pump, signal, and idler wavelengths. The double highreflector (high-reflection at pump and signal wavelengths, antireflection at idler wavelength) mirror M2 completes the high-finesse pump cavity. The signal cavity is defined by M2, the dichroic beam splitter BS, and mirrors M3 (plane) and M4 (200-mm radius of curvature). The beam splitter and mirrors M3 and M4 were antireflectioncoated at the pump wavelength and broadband highreflection coated at the signal wavelengths. Although the signal cavity could be configured in the absence of M3 and with the curved mirror M4 on axis with the signal field reflected from the beam splitter, M3 was included to eliminate any interferometric feedback of the 1.064-␮m pump radiation into the pump cavity by way of the signal cavity. We found that such feedback effects induced large intensity fluctuations in the circulating pump field, exacerbating the onset of relaxation oscillations. The curvature and separation of mirrors M2 and M4 was cho-

Fig. 1. Schematic of the all-solid-state, cw, intracavity SR-OPO with a fiber-coupled 15-W laser diode as the primary pump source. L1–L3, lenses; M1–M4, mirrors; BS, beam splitter.

2104


Stothard et al.

Fig. 2. Extracted one-way idler output power and one-way circulating intracavity pump power at 1.064 ␮m as a function of the external pump power from the diode laser for the cases of PPLN and PPRTA.

sen to form a concentric cavity for the signal in order to produce the appropriate signal waist at the center of the crystal. The distance between the laser pump head and the intracavity lens was chosen such that the cavity remained stable over a large range of effective radii of curvature of the mirror M1 to accommodate the thermal lens in the Nd:YVO4 crystal and to mode match the 1.064-␮m circulating field efficiently into the diode-pumped volume. The PPLN crystal19 was 0.5 mm ⫻ 11 mm ⫻ 25 mm in size and had eight gratings written within it over the range 28.5–29.9 ␮m in 0.2-␮m steps. The pump field was focused down into the crystal to form a waist of ⬃50 ␮m but whose size varied between this value and ⬃80 ␮m as a result of thermal lensing effects. The PPRTA crystal20 dimensions were 3 mm ⫻ 11 mm ⫻ 20 mm with only one grating at a period of 39.6 ␮m to phase match the signal and idler at 1.525 and 3.525 ␮m, respectively, at 20 °C. The pump field was focused to form a waist in the crystal of 50–60 ␮m. The PPLN was operated at a temperature of 150 °C to match the signal and idler wavelengths to that of the PPRTA operated at room temperature. The PPRTA was affixed to a thermoelectrically stabilized copper block that allowed the crystal temperature to be varied over the range 0–100 °C to facilitate temperature tuning, although the temperature was not reduced below 10 °C in order to avoid condensation on the crystal facets.

3. RESULTS The power characteristics of the two devices are summarized in Fig. 2. In it are shown the one-way extracted idler power (through M2) and the intracavity circulating field, each as a function of diode-laser pump power for the two devices. In terms of the circulating 1.064-␮m field, the SR-OPO threshold in the case of PPLN was lower than that of the PPRTA (30 W compared to 65 W), in line

with its higher nonlinear coefficient and longer interaction length. When one looks at SR-OPO thresholds in terms of external pumping power, however, it is the PPRTA that has the lower threshold as a result of a higher slope efficiency shown by the intracavity field: 4.2 W compared with 8.0 W in the case of the PPLN. The higher slope efficiency of the circulating field in the case of the PPRTA is ascribed to the higher resistance of this material to the formation of thermal lens effects; and consequently, a much-reduced aperturing of the laser cavity mode in comparison with PPLN. (As the circulating field in the PPLN was increased, a significant thermal lens was formed; this effect was observed to cause the circulating mode to expand, resulting in it then being apertured by the thickness of the crystal.) Complementary experiments carried out on an ‘‘empty-cavity’’ (i.e., one lacking the nonlinear crystal) configuration of the Nd:YVO4 laser indicated that, in addition, changing thermal lensing associated with the Nd:YVO4 gain medium itself also influenced performance, an effect which of course persists even when thermal lensing in the nonlinear medium has been minimized through the use of PPRTA. Further refinements in cavity design to obviate the effects of thermal lensing in the laser gain medium would be advantageous. Figure 2 shows that both devices produced a singlepass idler power of ⬃220 mW for 12 W of diode-laser pump power. This was the power extracted through mirror M2; taking into account the signal and idler wavelengths, the idler losses associated with the coatings of both this optic and that of the nonlinear crystal, and also that the idler was produced in both directions, we obtain a total downconverted power of 1.61 W, assuming 5% loss at each coating. This compared with 5.5 W of 1.064-␮m power extractable through M2 when it was replaced by an optimal output coupler of 10% transmission. There was almost no discernible evidence of the OPO clamping the intracavity field, as is evident from the fig-

Stothard et al.

ure. This is attributed to the changing mode sizes of both the pump and signal waists in the nonlinear crystal as the thermal lens in this and the laser gain crystal are constantly evolving as the diode-laser pump power is increased. In effect, the cavity geometry is dynamic as pump power and circulating field are varied, so the OPO has a different effective threshold as these parameters change. This makes an estimate of the downconversion efficiency extremely difficult as this parameter depends on the threshold of the laser and the OPO as well as the downconverted power produced for a given operating pump power.21 The laser and OPO thresholds are easily measured as a function of pump power, but as the cavity geometry has significantly changed because of thermal

Fig. 3. Sum–frequency-mixed spatial mode quality for the PPLN and PPRTA SR-OPOs pumped at 12 W. Top trace, PPLN; bottom trace, PPRTA.


2105

lensing at higher pump powers, a comparison between measured thresholds and downconverted power at these enhanced pump powers is not meaningful. It is interesting to note that the increase in idler power extracted from the two devices falters beyond diode-laser pump powers of ⬃12 W, and yet the circulating field continues to increase. We consider that this is due to the thermal lens’ formed in the respective nonlinear media destabilizing the signal cavity, which has no lens within it to act as optical ballast, whereas the pump cavity does. The inclusion of such a lens within the signal cavity may lead to improved idler output powers if the diode-laser pump power can thereby be increased significantly without the destabilization of the signal cavity. A rather striking difference between the thermal lensing behavior of the two crystals was evident by observing the quality of the spatial mode in each system. Figure 3 shows the spatial mode of the nonphase-matched, sum– frequency-mixed light emitted by the PPLN and PPRTA devices, as measured by a CCD camera. The two crystals exhibited very different behavior in this respect. Soon after exceeding OPO threshold, a very poor spatial mode was observed in the case of PPLN, a mode that moreover was constantly ‘‘breathing,’’ and hence very difficult to collapse down to a fundamental Gaussian mode. (The trace shown here is a typical ‘‘snapshot’’ at one point of the observed dynamic behavior of the mode, which was constantly changing in time.) The PPRTA system, in contrast, exhibited excellent mode quality that remained stable up to pump powers of ⬃13.5 W. The instability of the thermal lens in the PPLN crystal is evident from the behavior of the extracted idler power as measured over an extended time period. Figure 4 shows such a measurement over a period of three hours. These traces were measured with a thermal power meter, so any transients are lost as a result of the time response of the power meter thermopile. It is evident from the figure that the power stability of the PPRTA system was roughly twice that of the PPLN SR-OPO. The poor performance of the PPLN system is caused both by the unstable thermal lens in the PPLN crystal and by thermal currents occurring at the facets of the crystal as air was heated by the oven and rapidly accelerated over the crystal surface. This problem is not present in the case of PPRTA as it was operated at room temperature. The temperature of the PPLN oven was reduced from 150 °C to 100 °C; this caused the RMS noise to fall from ⬃20% to ⬃17%. Clearly, the thermal lens formed in the PPLN is the most significant factor in the output stability of the device. The pump field rejected from the upper-left face of the beam splitter (Fig. 1) was focused onto a fast photodiode to observe the transient dynamic behavior of the two devices, and these results appears as Fig. 5. The intracavity pump field in the case of PPLN exhibited a continuous train of pseudorandom relaxation oscillation pulses, whereas the circulating field in the PPRTA system was true cw. We recall that while intracavity systems of this type that employ laser gain media with a long upper-state lifetime are inherently stable, they are extremely sensitive to the onset of relaxation oscillations if the cavity is perturbed.22 One source of perturbations in the present

2106


Stothard et al.

matched, sum–frequency mixing of the pump and signal waves was measured to obtain the signal and idler wavelengths. The results are shown in Fig. 6. We see that over the range 10–100 °C, the idler tuned over the range 3538–3404 nm (133 nm) and the signal over the range 1522–1548 nm (26 nm). The PPRTA idler tuning rate with temperature of ⬃1.5 nm/ °C compares with ⬃2.1 nm/ °C in the case of PPLN. Also shown on the graph are three sets of theoretical temperature tuning curves. The solid curve uses the Sellmeier equation of Fennimore et al.23 and the dn/dt expression of Karlsson et al.24 The dotted curve uses the Sellmeier equation of Fradkin-Kashi et al.25 and again the dn/dt expression of Karlsson et al. The signal and idler wavelength pair predicted by the Sellmeier equation of Fradkin-Kashi et al. at room temperature was in excellent agreement with experimentally observed results. To yield better agreement between theory and experiment, the dn/dt coefficients were modified for best fit. The full expression for refractive index as a function of tempera-

Fig. 4. Long-term idler stability of the two devices pumped at 12 W. Top trace, PPLN; bottom trace, PPRTA.

case arose as a result of interferometric feedback of the pump field off the beam splitter by way of mirror M4 (M3 not being incorporated in the cavity in this case), under whose effects even the PPRTA system was observed to suffer from relaxation oscillations. However, with this feedback eliminated by the inclusion of mirror M3 (antireflection coated at 1064 nm), as discussed above, the OPO exhibited excellent transient stability. The PPLN system still suffered from relaxation oscillations even with the inclusion of M3, and we theorize that the ‘‘breathing’’ of the pump and signal mode through the erratic thermal lens formed in the PPLN crystal and the thermal air currents discussed above had sufficient effect to trigger the onset of these pulse trains. We now turn our attention to the tuning characteristics of the PPRTA ICSR-OPO. As the grating periods written on the PPLN crystal were identical to those used in a previously described system,4 the temperature tuning behavior of the PPLN ICSR-OPO is not replicated here. As the PPRTA had only a single grating written within it, tuning of the signal and idler was restricted to varying the temperature of the crystal. The PPRTA crystal was mounted on a thermoelectrically stabilized copper block whose temperature could be varied over the range 10–100 °C. As the diagnostics available in the laboratory were insensitive to the midinfrared idler wavelength, and only partially reliable at the near-infrared signal wavelength, the wavelength of the red light produced by nonphase-

Fig. 5. Transient dynamics of the intracavity pump field for the PPLN and PPRTA ICSR-OPOs pumped at 12 W. Top trace, PPLN; bottom trace, PPRTA. Both traces have a time base of 5 ms/div and are zero’ed at ⫺4 divisions.

Stothard et al.


Fig. 6.

Temperature tuning characteristics of the PPRTA ICSR-OPO.

ture and wavelength so deduced is given below, along with its coefficients. n⫽

冑

B

A⫹ 1⫺

冉冊 C

2

D

⫹ 1⫺

␭

冉冊 E

2

⫺ F␭ 2

␭

⫹ 共 T ⫺ rt 兲共 G␭ ⫺3 ⫹ H␭ ⫺2 ⫹ I␭ ⫺1 ⫹ J 兲 10⫺4 where A ⫽ 2.182064,

F ⫽ 0.008921,

B ⫽ 1.307519,

G ⫽ ⫺1.45046,

C ⫽ 0.228244,

H ⫽ 4.88661,

D ⫽ 0.354743,

I ⫽ ⫺4.52143,

E ⫽ 9.0190959,

2107

J ⫽ 0,

and rt refers to room temperature (20 °C), T is crystal temperature, and n the refractive index at the propagating wavelength ␭.

4. CONCLUSIONS We have demonstrated two ICSR-OPOs based on PPLN and PPRTA and operated internal to a high-power, diodelaser-pumped, Nd:YVO4 laser. Although the PPRTA had a lower nonlinear coefficient and was shorter in length in comparison with the PPLN, it exhibited comparable performance in terms of extracted idler power, but much superior temporal stability of output power. Thermal lensing effects in the PPLN crystal resulted in poor spatial stability of the sum–frequency-mixed light (hence also the downconverted light) as well as the circulating pump field, and also caused sufficient perturbation to the intracavity pump and signal fields to result in driving the pump laser into a quasi-continuous stream of re-

laxation oscillations. In contrast, a stable fundamental spatial mode was observed in the PPRTA ICSRO-OPO, and monitoring of the circulating pump field in this case showed that the system was operating in the absence of relaxation oscillations once the effects of interferometric feedback had been eliminated (by the inclusion of M3), thus demonstrating the crucial advantage of this material in realising true cw operation of an intracavity OPO. Although the temperature tuning of the PPRTA ICSROPO yielded less tuning in the idler wavelength than was the case with the PPLN system, the ability to choose the PPRTA grating period to obtain those wavelengths required for particular applications makes this a relatively minor drawback. Indeed, a reduced rate of change of idler wavelength with crystal temperature is advantageous in relaxing tolerances placed on the control of crystal temperature. Also, the future development of multiple fanned gratings within a single crystal of PPRTA is to be anticipated, hence providing very broad and tunable spectral coverage of the idler and signal wavelengths, comparable indeed to those previously demonstrated in PPLN systems. In a system such as this, careful cavity design and robust performance in the face of thermal lensing in intracavity elements is crucial. The poor thermal performance of PPLN in comparison with PPRTA was largely attributed to its propensity for thermal lensing resulting from its higher absorption coefficient at 1.064 ␮m and its larger refractive-index dispersion with temperature. An estimation of the downconversion efficiency of this system was not possible because of the dynamic nature of the pump cavity geometry as the diode-laser pump power was increased. Although this system with its present cavity design failed to give downconversion power comparable to that extractable at the pump wavelength with an appropriate choice of output coupling, the ⬎440 mW of useful idler output power demonstrated by the PPRTA

2108


Stothard et al.

system holds the promise of finding new and interesting applications. Corresponding author D.J.M. Stothard may be reached by e-mail to [email protected].

REFERENCES AND NOTES 1.

2. 3.

4.

5.

6.

7.

8.

9. 10.

11.

12.

W. R. Bosenberg, A. Drobshoff, J. I. Alexander, L. E. Myers, and R. L. Byer, ‘‘Continuous-wave singly resonant optical parametric oscillator based on periodically poled LiNbO3 ,’’ Opt. Lett. 21, 713–715 (1996). F. G. Colville, M. H. Dunn, and M. Ebrahimzadeh, ‘‘Continuous-wave, singly resonant, intracavity parametric oscillator,’’ Opt. Lett. 22, 75–77 (1997). T. J. Edwards, G. A. Turnbull, M. H. Dunn, M. Ebrahimzadeh, and F. G. Colville, ‘‘High-power, continuous-wave, singly resonant, intracavity optical parametric oscillator,’’ Appl. Phys. Lett. 72, 1527–1529 (1998). D. J. M. Stothard, M. Ebrahimzadeh, and M. H. Dunn, ‘‘Low pump threshold, continuous-wave, singly resonant, optical parametric oscillator,’’ Opt. Lett. 23, 1895–1897 (1998). U. Strossner, J. P. Meyn, R. Wallenstein, P. Urenski, A. Arie, G. Rosenman, J. Mlynek, S. Schiller, and A. Peters, ‘‘Singlefrequency continuous-wave optical parametric oscillator system with an ultrawide tuning range of 550 to 2830 nm,’’ J. Opt. Soc. Am. B 19, 1419–1424 (2002). G. M. Gibson, M. Ebrahimzadeh, M. J. Padgett, and M. H. Dunn, ‘‘Continuous-wave optical parametric oscillator based on periodically poled KTiOPO4 and its application to spectroscopy,’’ Opt. Lett. 24, 397–399 (1999). A. Popp, F. Muller, F. Kuhnemann, S. Schiller, G. von Basum, H. Dahnke, P. Hering, and M. Murtz, ‘‘Ultra-sensitive mid-infrared cavity leak-out spectroscopy using a cw optical parametric oscillator,’’ Appl. Phys. B 75, 751–754 (2002). G. A. Turnbull, D. McGloin, I. D. Lindsay, M. Ebrahimzadeh, and M. H. Dunn, ‘‘Extended mode-hop-free tuning by use of a dual-cavity, pump-enhanced optical parametric oscillator,’’ Opt. Lett. 25, 341–343 (2000). K. Schneider and S. Schiller, ‘‘Narrow-linewidth, pumpenhanced singly-resonant parametric oscillator pumped at 532 nm,’’ Appl. Phys. B 65, 775 (1997). E. V. Kovalchuk, D. Dekorsy, A. I. Lvovsky, C. Braxmaier, J. Mlynek, A. Peters, and S. Schiller, ‘‘High-resolution doppler-free molecular spectroscopy with a continuouswave optical parametric oscillator,’’ Opt. Lett. 26, 1430– 1432 (2001). I. D. Lindsay, G. A. Turnbull, M. H. Dunn, and M. Ebrahimzadeh, ‘‘Doubly resonant continuous-wave optical parametric oscillator pumped by a single-mode diode laser,’’ Opt. Lett. 23, 1889–1891 (1998). R. Al-Tahtamouni, K. Bencheikh, R. Storz, K. Schneider, M. Lang, J. Mlynek, and S. Schiller, ‘‘Long-term stable opera-

13.

14.

15.

16.

17.

18.

19. 20. 21.

22.

23.

24.

25.

tion and absolute frequency stabilization of a doubly resonant parametric oscillator,’’ Appl. Phys. B 66, 733–739 (1998). P. Gross, M. E. Klein, T. Walde, K. J. Boller, M. Auerbach, P. Wessels, and C. Fallnich, ‘‘Fiber-laser-pumped continuouswave singly resonant optical parametric oscillator,’’ Opt. Lett. 27, 418–420 (2002). G. A. Turnbull, T. J. Edwards, M. H. Dunn, and M. Ebrahimzadeh, ‘‘Continuous-wave singly-resonant intracavity optical parametric oscillator based on periodically-poled LiNbO3 ,’’ Electron. Lett. 33, 1817–1818 (1997). L. K. Cheng, L. T. Cheng, J. Galperin, P. A. M. Hotsenpiller, and J. D. Bierlein, ‘‘Crystal Growth and Characterization of KTiOPO4 Isomorphs from the Self-Fluxes,’’ J. Cryst. Growth 137, 107–115 (1994). L. E. Myers, R. C. Eckardt, M. M. Fejer, R. L. Byer, W. R. Bosenberg, and J. W. Pierce, ‘‘Quasi-phase-matched optical parametric oscillators in bulk periodically poled LiNbO3 ,’’ J. Opt. Soc. Am. B 12, 2102–2116 (1995). T. J. Edwards, G. A. Turnbull, M. H. Dunn, M. Ebrahimzadeh, H. Karlsson, G. Arvidsson, and F. Laurell, ‘‘Continuous-wave singly resonant optical parametric oscillator based on periodically poled RbTiOAsO4 ,’’ Opt. Lett. 23, 837–839 (1998). I. D. Lindsay, D. J. M. Stothard, C. F. Rae, and M. H. Dunn, ‘‘Continuous-wave, pump-enhanced optical parametric oscillator based on periodically-poled RbTiOAsO4 ,’’ Opt. Express 11, 134–140 (2003). The PPLN crystal was supplied by Crystal Technology, Inc., 1040 East Meadow Circle, Palo Alto, Calif. 94303-4230., PPLN Crystal. The PPRTA crystal was supplied by Crystal Associates, Inc., PPRTA Crystal. G. A. Turnbull, M. H. Dunn, and M. Ebrahimzadeh, ‘‘Continuous-wave, intracavity optical parametric oscillators: An analysis of power characteristics,’’ Appl. Phys. B 66, 701 (1998). G. A. Turnbull, D. J. M. Stothard, M. Ebrahimzadeh, and M. H. Dunn, ‘‘Transient dynamics of CW intracavity singly resonant optical parametric oscillators,’’ IEEE J. Quantum Electron. 35, 1666–1672 (1999). D. L. Fenimore, K. L. Schepler, D. Zelmon, S. Kuck, U. B. Ramabadran, P. VonRichter, and D. Small, ‘‘Rubidium titanyl arsenate difference-frequency generation and validation of new Sellmeier coefficients,’’ J. Opt. Soc. Am. B 13, 1935–1940 (1996). H. Karlsson, F. Laurell, P. Henriksson, and G. Arvidsson, ‘‘Frequency-doubling in periodically poled RbTiOAsO4 ,’’ Electron. Lett. 32, 556 (1996). K. Fradkin-Kashi, A. Arie, P. Urenski, and G. Rosenman, ‘‘Characterization of optical and nonlinear properties of periodically-poled RbTiOAsO4 in the midinfrared range via difference-frequency generation,’’ Appl. Phys. B 71, 251– 255 (2000).