We demonstrate a PPLN based pump-enhanced, singly-resonant optical parametric oscillator configured in a traveling wave geometry and pumped by a Ti:sapphire laser. The inclusion of a low finesse etalon within the OPO cavity stabilizes the signal frequency, and rotation of the etalon allows this frequency to be systematically hopped from axial mode to nearest neighbor axial mode over the entire free spectral range of the etalon (83GHz). Tuning of the pump frequency allows the signal frequency to be smoothly tuned over a cavity free spectral range. More than 35mW of single frequency idler power was generated in the spectral range 2800–3000nm for 600mW pump power. The superiority of traveling wave over standing wave geometries in these regards is discussed.
©2004 Optical Society of America
Continuous-wave optical parametric oscillators continue to attract much interest as compact and efficient sources for the generation of coherent radiation of high spectral quality (single-frequency) over extensive spectral ranges. A particular advantage of these nonlinear optical devices is their capability of addressing diverse spectral regions, often where alternative approaches to the generation of coherent radiation do not exist. This communication reports further progress towards obtaining continuous frequency coverage at the single-frequency level and over large spectral ranges from low oscillation threshold (and hence compact) parametric devices operating as “sine-wave” oscillators.
The field in general has seen substantial progress in recent years resulting both from the advent of new optically nonlinear materials, in particular periodically-poled crystals, and the development of novel resonant cavity geometries. The singly-resonant optical parametric oscillator in particular has benefited substantially from the higher optical non-linearity accessible through the use of the poled materials and much progress has been made in demonstrating sources that combine wide and continuous tuning ranges with single frequency operation [1–6]. However, these devices still exhibit comparatively high oscillation thresholds and so require powerful pump lasers for their operation. Doubly-resonant optical parametric oscillators exhibit substantially lower oscillation thresholds, and although they have been found useful sources of single frequency radiation, continuous spectral coverage/tuning, stability and ease of operation are all more problematical compared to the singly resonant device [7–11]. The pump-enhanced (sometimes referred to as pump-resonant) optical parametric oscillator , has been widely adopted as an approach that combines significant reduction in the oscillation threshold with the attractive features of the singly resonant oscillator, namely controllable tuning at high spectral resolution [13–16], and, by use of dual cavity techniques [17, 18], tuning over wide spectral ranges. In addition, the singly-resonant characteristic of the PE-SRO means that oscillation of the resonant down-converted wave can always be established once sufficient pump-enhancement has been achieved unlike the case of the doubly-resonant oscillator, where only unique resonator conditions allow oscillation. Also, the detrimental effect of cluster hopping is avoided in the former compared to the latter. As a result the pump-enhanced approach has recently found useful applications in mid-infrared trace gas detection [18–20]. A further approach for lowering the oscillation threshold of optical parametric oscillators is through the use of intra-cavity geometries where the parametric oscillator is located within the optical cavity of the pump laser itself, so that the nonlinear medium is subject to the high intracavity pump field .
At the present time pump-enhanced devices have almost all been based on standing wave cavity geometries [12–16]. Where traveling wave cavity (ring resonator) geometries have been reported [22, 23], studies have been confined to the frequency stability and noise properties of these devices, and issues concerned with their tuning have not been addressed. However, it is the case that in the context of the singly resonant oscillator, the traveling wave cavity is known to offer distinct advantages over the standing wave cavity with regard to reliable tuning [1, 4].
In this communication we report on the development of a pump-enhanced optical parametric oscillator based on the traveling wave geometry with particular emphasis on attaining improved tuning performance through this approach. Key outcomes reported here include: the ability to reliably and systematically mode-hop the oscillator from one cavity mode to its nearest-neighbor cavity mode over a sequence of some 100 modes, corresponding to a frequency range in excess of 80GHz, limited only by the properties of the etalon used; attaining such control using only a low-finesse etalon, thereby minimizing intracavity losses; combining the above extensive spectral coverage with an ability to smoothly tune the selected cavity mode over greater than a free spectral range of the cavity through pump tuning. Overall such a combination of mode hopping and fine-tuning results in a low threshold devices that provides continuous spectral coverage over greater than 80GHz, with single-frequency resolution throughout. Although mode hopping over an extended spectral range (~15cm-1) has been demonstrated through a combination of an air spaced intracavity etalon and a fanned grating in PPLN , this was in an SRO without pump-enhancement, and the frequency change due to a single mode hop event was typically 0.1cm-1 corresponding to ~5 cavity mode spacing thus leaving substantial gaps in spectral coverage. Continuous tuning over 40GHz by synchronous control of etalon tilt and signal cavity length has been reported  using a traveling wave cavity, but again in a system without pump-enhancement. If such a tuning scheme were to be implemented in a pump-enhanced device, a dual cavity [17,18] would be required. We believe that the results in this paper are a significant advance in demonstrating for the first time, to our knowledge, systematic nearest-neighbor mode hopping over extensive spectral ranges in the context of any CW-OPO, pump-enhanced or otherwise.
2. Design of optical parametric oscillator
The continuous-wave, pump-enhanced, singly-resonant (on the signal wave) optical parametric oscillator (PE-OPO) is illustrated in Fig. 1. The ring resonator is designed to resonate both the incoming pump wave so as to provide the required pump enhancement as well as one of the down-converted waves, in the present arrangement the signal wave at 1100nm. The nonlinear crystal is periodically-poled lithium niobate (PPLN), of dimensions 19mm×0.5mm×11mm (length×thickness×width), with 8 gratings arranged across the width, with periods varying from 21µm to 22.4µm. The crystal was mounted inside of a specially-designed oven controlled by a temperature controller (Wavelength Electronics model LFI 3751) to 0.01°C, and was operated over a temperature range from 150 to 190°C, so avoiding photorefractive damage. Both crystal faces, which are cut normal to the propagation direction, are coated for antireflection at the pump (reflectivity<0.5%), (resonant) signal (reflectivity<0.8%) and idler (reflectivity<1.0%). In the present characterisation experiments the PE-OPO was pumped by a single-frequency Ti:sapphire laser (Microlase Model MBR-110), but pump power requirements (see below) are consistent with the use ultimately of diode-lasers for this purpose. The bow-tie ring cavity is formed by two plano reflectors and two focussing mirrors (concave, radius of curvature 50mm) in the arrangement shown in Fig. 1 (distances M1-M2, M2-M3, and M4-M1~100mm, distance M3-M4 ~65mm, free spectral range of cavity ~750MHz). Because of the non-zero angle of incidence (~17°) on the focusing mirrors, the cavity is astigmatic. However, analysis shows that this has little effect on the beam waist located at the center of the crystal, which has comparable dimensions in both the sagittal and tangential planes, namely around 45µm (full width 1/e2) for the pump wavelength around 850nm and around 55µm for the signal wavelengths around 1100nm. The large beam waist located mid point between mirrors M1 and M2 is however more astigmatic, but what is of greater concern here from the point of view of frequency control is its size at the signal wavelength, which is calculated to be around 640µm×360µm. Mirrors M1, M2, M3 were coated for high reflectivity at the pump wavelength (reflectivity > 99.9% at 805nm) and signal wavelength (reflectivity >99.9% at 1100nm), and to be transmitting at the idler wavelength (transmission >99.6% at 2800 nm). Mirror M4 was similar to the above, except that as the input coupling mirror for the pump its transmission at the pump wavelength was chosen to be 4%. The incoming pump light was focused through mirror M4 (with focusing lens f=100mm) so as to form a beam waist within the PPLN crystal. The focusing conditions were adjusted while observing the resonant enhancement of the intracavity pump field as the cavity was scanned through its resonant condition in order to optimize the coupling of pump power into the fundamental cavity mode. (We estimate that in this way some 83% of the incoming pump light was so coupled). The ring resonator was then locked onto resonance with the incoming pump field, which was phase modulated at 50MHz by an external phase modulator for this purpose, using the Pound-Drever-Hall technique  with the error signal fed to the piezo-electric transducer on which mirror M2 was mounted. Enhancement factors, prior to parametric oscillation, of the order of 9.5 were routinely achieved by this means. This enhancement factor was determined by measuring the leaking field through mirror M3 whilst the ring cavity was held on resonance with the pump field, and also by ascertaining the impedance matching of the cavity by analyzing the ratio of pump light rejected from the cavity when on and off the resonance condition. A measured enhancement factor of 9.5 indicates a cavity round trip loss of 6.5% and a finesse of 75 at the pump wavelength; comparable to parameters reported elsewhere . With enhancement factors of this order, external pump powers required to reach oscillation threshold are reduced by close to an order of magnitude compared to single pass, singly-resonant devices, so opening up opportunities for compact, all solid-state systems sources with low power requirements. Since the signal cavity as well as the pump cavity is locked by this means, the signal/idler output acquires the frequency stability (expressed as a fraction of the mean frequency) associated with the pump laser (in this case <100kHz rms).
A low-finesse etalon (~10% reflectivity coating on both faces at the signal wavelength of 1100nm, antireflection coated at the pump wavelength of 805nm, fabricated in YAG so as to be transparent at all three wavelengths involved, ~1mm thick, free spectral range 83 GHz) was inserted at the position of the large beam waist in arm M1-M2 of the cavity. It was mounted on a galvanometer driver to allow tuning through tilting.
Coarse tuning was through selection of the grating period or through temperature tuning. With regard to the latter, with a grating period of 21.2µm, the idler wave tuned from 2800nm to 3000nm for a temperature change from 150°C to 190°C (temperature tuning rate of 180GHz/°C).
3. Characterisation of optical parametric oscillator
Figure 2 shows the output power in the idler wave (wavelength) as a function of external pump power (wavelength 800nm). Oscillation threshold was around 250mW corresponding to a circulating intracavity power of 2.3W for the measured enhancement factor of 9.5. When pumping at 625mW, 40mW of idler output was generated, corresponding to a total down conversion efficiency of around 25%. (We believe that this efficiency may be improved through closer impedance matching of the ring cavity to the incoming pump light since in the present arrangement 17 % of the incident pump light was reflected from the cavity).
The presence of the etalon in the ring cavity ensures that the resonant signal wave maintains a stable single frequency. Figure 3(a) shows the mode hop tuning of the OPO as the etalon is tilted within the ring cavity while the pump frequency is held constant. Since the etalon is coated for antireflection at the pump wavelength, the only effect of this tilting on the resonant pump field is a change in its optical path through the etalon. The ring cavity is however held on resonance with the pump wave throughout by compensatory changes in cavity length via the piezo crystal mounted mirror M2 using the well established Pound- Drever-Hall locking scheme. On the other hand, the 10% reflectivity associated with each of its faces at 1100nm provides the mode selection of the signal wave. The notable feature here is the demonstration of reliable and systematic hopping from any particular (axial) mode to its adjacent (axial) mode (the axial mode spacing being 750MHz) over almost the entire free spectral range of the etalon (83GHz), in excess of 100 mode hops in all.
Figure 3(b) shows a magnified section of part of the hopping range shown in Fig. 3(a), to more clearly illustrate the regularity of the hopping from adjacent mode to adjacent mode. That this behavior can be repeated over at least five adjacent free spectral ranges of the etalon, corresponding to a total etalon rotation angle (starting from close to normal incidence, but avoiding the “flash” condition associated with exact normal incidence) of 6° highlights the advantage of being able to use such a low finesse etalon, with associated low insertion loss, for reliable mode control in the case of the ring resonator (taking into consideration the characteristics of the etalon and the size of the beam waist at its location, walk-off losses are estimated to be about 1% for a rotation angle corresponding to tuning over one free spectral range). This is in contrast to the situation encountered in the standing wave configurations . It may be seen from Fig. 3(a) that for a limited range (B to E) in the systematic hopping over the full free spectral range of the etalon (A to F) the OPO transfers to oscillating on an adjacent etalon mode, thereby covering the spectral range C to D before returning to the main sequence at point E. This behavior is reproducible occurring at similar signal wavelengths, centered on 1095.10nm (standard air), in all of the five free spectral ranges scanned. This wavelength corresponds to a known line in water vapor , and moreover is the only known absorption line associated with atmospheric absorption occurring in the spectral range scanned. We discuss this issue further in section 4 below. At point F the adjacent mode of the etalon becomes closer to the center of the phase-matched bandwidth and hence the signal frequency jumps by one free spectral range of the etalon (F to A=FSRetalon).
Figures 4(a) and 4(b) show the dependence of the signal wave frequency on that of the pump wave as the latter is tuned while keeping the etalon angle fixed. From fig. 4a it may be seen that, as expected, it is possible to tune the OPO over most of the frequency interval between adjacent axial modes (750MHz) for only a modest tuning range of the pump laser (since signal and pump share a common cavity the tuning range associated with the signal wave is given by the tuning range of the pump multiplied by the pump wavelength divided by the signal wavelength). Continuous tuning exceeding the mode spacing may be attained by synchronous rotation of the etalon. Figure 4(b) shows the effectiveness of the etalon in controlling the signal frequency when the pump laser is tuned over a more substantial range.
The “fine” tuning range associated with inter-mode scanning displayed in Fig. 4(a) is not here resolved, but it is apparent that the etalon maintains effective control of the signal frequency through mode selection, until the center of the phase match bandwidth shifts, due to pump tuning, so as to favor the adjacent free spectral range of the etalon (adjacent etalon mode) as that point where cavity mode selection occurs. (Note that under the present phase matching conditions the center frequency of the signal gain curve experiences a shift that is approximately five times the shift in the pump frequency). The anomalous point relates to water absorption of the signal field, as discussed at the end of the previous paragraph. An alternative method to pump tuning, as described above, for (ultimately) transferring the spectral range over which mode selection occurs to an adjacent free spectral range of the etalon is to employ temperature tuning. Given that the tuning rate with temperature under the present conditions is 180GHz/°C, temperature changes of the order of 0.5°C are hence required for this purpose. It is also apparent that the present arrangement for frequency selection does not place too great a demand on temperature stability, since etalon control of the oscillator frequency is maintained for a shift in the center frequency of the phase match bandwidth of up to half the free spectral range of the etalon (~40GHz), corresponding to a temperature change of around 0.2°C.
We consider first of all the discrimination towards mode selection exhibited by the intracavity etalon. Etalon transmission is described by the following standard expression in the case of plane waves of infinite extent:
where the R’s are the reflectivities of the two surfaces of the etalon (in this case equal at 0.1), and δ is the round-trip phase shift within the etalon. The axial mode spacing of the signal cavity in the present case is 750MHz, so from the above it can be seen that for a free spectral range of the etalon of 83GHz, effective mode selection is attained with a differential discrimination (in terms of a change in the etalon’s transmission) between adjacent modes of 0.04%. The phase match bandwidth associated with the PPLN crystal under the present operating conditions is estimated to be 1200GHz (full-width). Since the associated saturated gain equals the signal cavity round trip loss, previously estimated to be of the order of 2%, under steady-state conditions, the estimated change in the gain between selected mode and the next adjacent mode is 3×10-6 % for operation at gain curve center, rising to 3.5×10-4 % for operation 40GHz from center (this latter corresponding to the maximum frequency excursion from gain curve center attainable by mode selection with the etalon). It can be seen that the present etalon is hence well suited to ensuring selection of that cavity mode lying closest to its transmission maximum throughout the hopping range explored, thus ensuring, as is indeed observed, that the hopping proceeds systematically to nearest neighbor cavity mode throughout, and hence thereby ensuring a continuous tuning capability.
As described in Section 3 above, a gap in the tuning range occurs in the vicinity of a known water absorption line. The location of this gap corresponds, within the absolute accuracy of the wavemeter (+/-3ppm), to the tabulated wavelength  of this water absorption line (1095.10nm, under standard air conditions). Further, the absorption loss on this line due to the presence of atmospheric water vapor in the cavity is estimated to be of order of magnitude 6×10-3 %, and as such is within a factor of 2 of the expected diminution in the gain (10-2 %) associated with operation on an adjacent etalon mode rather than on that mode closest to gain center. The extent of the gap is also comparable with the pressure-broadened line width of the line (10GHz). Purging the cavity with dry air or an inert gas should eliminate this gap in the continuous tuning.
The ability to attain systematic hopping in the case where the etalon selectivity is sufficient to discriminate against changes in the parametric gain alone is indicative of the superiority of the traveling wave (ring) cavity over the standing wave cavity. Our previous experience is that such reliable hopping is not possible with the latter class of cavity, certainly for an etalon of such low reflectivity . In this case, although the overall trend with etalon tilt is similar to that reported here, on a mode-to-mode basis hopping is not always to the nearest neighbor mode, a not infrequent behavior being for the hop to miss out a number of modes in between, thereby detracting from the ability of such a geometry to provide truly continuous spectral coverage. We believe that such inferior performance may be ascribed to two causes. Firstly the standing wave cavity has greater susceptibility to parasitic etalon effects, in particular those involving the highly reflecting cavity mirrors at the ends of the cavity. Since these mirrors are normal to the optical axis, unlike in a ring resonator, strong parasitic etalon effects can arise between one or other of them and any single surface normal to the optical axis where the antireflection coating is less than perfect. On the other hand, in the case of the ring resonator two such surfaces are necessary, they must be parallel to one another, and even then both surfaces are likely to be weakly reflecting unlike the combination in the standing wave cavity of a strong with a weak reflector. Secondly the standing wave cavity when oscillating on only a single axial mode has a sequence of nodes and antinodes along its optical axis, in contrast to the case of the traveling wave cavity where no such internal pattern occurs. The positions of the nodes and antinodes change as the optical parametric oscillator is tuned, and this may in itself provide an additional and parasitic frequency selection mechanism. The location of an antinode at some internal (crystal, say) surface within the cavity produces a large electric field amplitude on the surface and this has the potential for introducing greater loss than in the case when a node, say, is located on the surface. (Just such a mechanism has been used in the past for axial mode selection in low gain lasers).
We have demonstrated that the use of a ring resonator, in a single cavity configuration that resonates both pump and signal wave, has allowed precise and reproducible axial mode control of a pump-enhanced optical parametric oscillator following the insertion of a suitable etalon. In the present case in particular the demonstration of some 100 sequential mode hops between nearest neighbor modes ensures complete and reliable spectral coverage throughout a bandwidth in excess of 80GHz. The availability of just a modest range of pump tuning (~1GHz), so as to tune the down converted signal wave over the mode spacing associated with the signal cavity, when combined with spectral interleaving provided by the above mode hopping, allows this full bandwidth to be covered. On the other hand a wider range of pump tuning (~10GHz) accompanied by extended synchronous tuning of the etalon tilt angle would allow uninterrupted tuning throughout bandwidths in excess of 10GHz, the limitation in this case being the magnified (×5) change of the center frequency of the phase match bandwidth over change of the pump frequency ultimately leading to the frequency jumping a free spectral range of the etalon. In order to circumvent this latter limitation, synchronous tuning of the phase-match condition (either by temperature tuning of the non linear crystal or the use of a fanned grating ) or alternatively the implementation of a dual cavity scheme , is required. Such an arrangement could also be used to obviate the requirement for pump tuning; independent tuning of the signal cavity, along with synchronous tuning of the etalon located in this cavity, is now possible since the (separate) pump enhancement cavity can be maintained resonating the fixed frequency pump.
References and links
1. U. Stroβner, J.-P. Meyn, R. Wallenstein, P. Urenski, A. Arie, G. Rosenman, J. Mlynek, S. Schiller, and A. Peters, “Single-frequency continuous-wave optical parametric oscillator system with an ultra wide tuning range of 550 to 2830nm,” J. Opt. Soc. Am. B 19, 1419–1424 (2002). [CrossRef]
2. M. E. Klein, D.-H. Lee, J.-P. Meyn, K.-J. Boller, and R. Wallenstein, “Singly resonant, continuous-wave optical parametric oscillator pumped by a diode laser,” Opt. Lett. 24, 1142–4 (1999). [CrossRef]
3. 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, 713–15 (1996). [CrossRef] [PubMed]
4. W. R. Bosenberg, A. Drobshoff, J. I. Alexander, L. E. Myers, and R. L. Byer, “93% pump depletion, 3.5W continuous-wave, singly resonant optical parametric oscillator,” Opt. Lett. 21, 1336–38 (1996). [CrossRef] [PubMed]
5. L. E. Myers, R. C. Eckardt, M. M. Fejer, R. L. Byer, and J. W. Pierce, “Multigrating quasi-phase-matched optical parametric oscillator in periodically poled LiNbO3, “Opt. Lett. 21, 591–3 (1996). [CrossRef] [PubMed]
6. M. van Harpen, S. L. Hekkert, S. E. Bisson, and F. J. M. Harren, “Wide single-mode tuning of a 3.0–3.8µm, 700mW, continuous-wave Nd:YAG-pumped optical parametric oscillator based on periodically poled lithium niobate,” Opt. Lett. 27, 640–642 (2002). [CrossRef]
7. R. Al-Tahtamouni, K. Bencheikh, R. Storz, K. Schneider, M. Lang, J. Mlynek, and S. Schiller, “Long-term stable operation and absolute frequency stabilisation of a doubly resonant parametric oscillator,” Appl. Phys. B 66, 733–39 (1998). [CrossRef]
8. 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, 2102–16 (1995). [CrossRef]
9. 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, 397–9 (1999). [CrossRef]
10. F. G. Colville, M. J. Padgett, and M. H. Dunn, “Continuous-wave, dual-cavity, doubly resonant, optical parametric oscillator,” Appl. Phys. Lett. 64, 1490–92 (1994). [CrossRef]
11. A. J. Henderson, P. M. Roper, L. A. Borchowa, and R. D. Mead, “Stable, continuously tunable operation of a diode-pumped doubly resonant optical parametric oscillator,” Opt. Lett 25, 1265–1266 (2000). [CrossRef]
13. K. Schneider, P. Kramper, S. Schiller, and J. Mlynek, “Toward an optical synthesizer: a single-frequency parametric oscillator using periodically poled LiNbO3,” Opt. Lett. 22, 1293–95 (1997). [CrossRef]
14. 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 continuous-wave optical parametric oscillator,” Opt. Lett. 26, 1430–1432 (2001). [CrossRef]
15. 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 RbTiAsO4,” Opt. Express 11, 134–140 (2003), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-11-2-134 [CrossRef] [PubMed]
16. M. Scheidt, B. Beier, R. Knappe, K.-J. Boller, and R. Wallenstein, “Diode-laser-pumped continuous-wave KTP optical parametric oscillator,” J. Opt. Soc. Am. B 12, 2087–94 (1995). [CrossRef]
17. 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, 341–43 (2000). [CrossRef]
18. F. Muller, A. Popp, F. Kuhnemann, and S. Schiller, “Transportable, highly-sensitive photoacoustic spectrometer based on a continuous-wave dual-cavity optical parametric oscillator,” Opt. Express 11, 2820–25 (2003), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-11-22-2820 [CrossRef] [PubMed]
19. A. Popp, F. Muller, S. Schiller, G. v. Basum, H. Dahnke, P. Hering, M. Murtz, and F. Kuhnemann, “Ultra-sensitive mid-infrared cavity leak-out spectroscopy using a cw optical parametric oscillator,” Appl. Phys. 75, 751–54 (2002). [CrossRef]
20. F. Kuhnemann, K. Schneider, A. Hecker, A. A. E. Martis, W. Urban, S. Schiller, and J. Mlynek, “Photoacoustic trace gas detection using a cw single-frequency optical parametric oscillator,” Appl. Phys. B 66, 741–45 (1998). [CrossRef]
21. M. Ebrahimzadeh, G. A. Turnbull, T. J. Edwards, D. J. M. Stothard, I. D. Lindsay, and M. H. Dunn, “Intracavity continuous-wave singly resonant optical parametric oscillators,” J. Opt. Soc. Am. 16, 1499–1511 (1999). [CrossRef]
22. D. Chen, D. Hinkley, J. Pyo, J. Swenson, and R. Fields, “Single-frequency low-threshold continuous-wave 3-µm periodically poled lithium niobate optical parametric oscillator,” J. Opt. Soc. Am. B 15, 1693–97 (1998). [CrossRef]
23. F. Hanson, P. Poirier, and M. A. Arbore, “Single-frequency mid-infrared optical parametric oscillator source for coherent laser radar,” Opt. Lett. 26, 1794–96 (2001). [CrossRef]
24. S. E. Bisson, K. M. Armstrong, T. J. Kulp, and M. Hartings, “Broadly tunable, mode-hop-tuned cw optical parametric oscillator based on periodically poled lithium niobate,” Appl. Opt. 40, 6049–6055 (2001). [CrossRef]
25. R. W. P. Drever, J. L. Hall, F. V. Kowalski, J. Hough, G. M. Ford, A. J. Munley, and H. Ward, “Laser Phase and Frequency Stabilisation Using an Optical Resonator,” Appl, Phys, B-Lasers and Optics 31, 97–105 (1983). [CrossRef]
26. HITRAN data base, http://cfa-www.harvard.edu/HITRAN/.