We demonstrate, for the first time, 10 W, Er-fiber laser pumped, pure CW, thermally diffusion doped, polycrystalline Cr2+:ZnS laser operating at 2380 nm. We also show Littrow-grating, “single-knob”, wavelength tuning of the laser spanning 1940–2780 nm spectral range with the maximum output power of 7.4 W near the central wavelength of 2400 nm and above 2 W over 1970–2760 nm wavelength range. The laser performs with 40% real optical- and 43% slope efficiency, and shows no output power roll-off up to the highest available incident pump power of 27 W.
©2009 Optical Society of America
The existence of large number of strong vibronic absorption lines of organic molecules within the middle-infrared (mid-IR) spectral range of 2–20 μm makes this “molecular fingerprint” wavelength region extremely attractive for many scientific and practical applications. Virtually unlimited number of possibilities exist for such fields of study as eye-safe efficient laser surgery, laser radars, remote sensing of atmospheric constituents, non-invasive medical diagnostics, high-resolution molecular spectroscopy. Of a particular interest is the 2–5μm wavelength region where transition metal doped II-VI semiconductor gain media are feasible for direct generation of mid-IR laser radiation at room temperature (RT) [1–6]. Much attention has been devoted to development of Cr2+:ZnSe lasers due to excellent lasing and optical properties of this material in the mid-IR spectral region. The laser action of Cr2+ :ZnSe in continuous wave (CW) regimes of operation with efficiency exceeding 60% , power levels in excess of 1.8 W , gain switched pulsed lasing with average output power of up to 18.5 W , and CW wavelength tunability spanning 1880-3100 nm  have been reported. Recent progress  in fabrication of high-quality Cr2+ :ZnSe laser crystals has resulted in demonstration of multi-Watt output powers of pure CW, Er/Tm-fiber laser pumped Cr2+:ZnSe laser systems [12,13].
The closest alternative to Cr2+:ZnSe is the chromium-doped zinc sulfide (Cr2+:ZnS) laser gain media . This material naturally possesses higher optical damage threshold  (1.5 J/cm for ZnS, 0.5 J/cm for ZnSe), higher thermal conductivity  (27 W/m·K for ZnS, 19 W/m·K for ZnSe), and lower thermooptic coefficient dn/dT [15,16] (46×10-6 Kr-1 for ZnS, 70×10-6 Kr-1 for ZnSe), thus leading to weaker thermal lensing effects. Furthermore, Cr2+ :ZnS is more suitable for Er-fiber laser pumping at 1.56μm wavelength, which is located closer to the maximum of Cr2+ :ZnS absorption band of 1.68 μm [1,3]. However, primarily due to unavailability of high-quality Cr2+:ZnS crystals, only a few low-power CW laser systems based on this media have been demonstrated previously [4,17,18].
In this work we report the results of our preliminary research on power-scaling of CW Cr2+:ZnS laser systems. We demonstrate 10 W pure CW Cr2+:ZnS laser system pumped by a 26 W Er-fiber laser. The achieved optical efficiency on the incident pump power is 40% and the slope efficiency is 43%, respectively, with no noticeable output power roll-off at high pump powers. In a slightly modified (dispersive) cavity the laser also demonstrates an overall wavelength tunability range spanning 1940–2780 nm, with the output power exceeding 2 W level over wavelength range of 1970–2760 nm, and above 5 W level within 2120–2590 nm.
2. Experimental setup
The experimental setup is shown schematically in Fig. 1. The high-power Cr2+:ZnS laser is based on a rectangular 7 × 7 × 2.6 mm (7 mm is length) polycrystalline, thermally-diffusion-doped Cr2+:ZnS gain element mounted in the laser cavity at the Brewster angle. The Cr2+ :ZnS sample is pumped by 1.56 μm Er-fiber laser through the 2.6 × 7 mm facets. The absorption coefficient of this homogeneously-doped laser crystal at 1.56 μm is about 5.8 cm-1, which corresponds to the unsaturated absorption of the pump radiation of about 98%. To fully explore the power scaling capabilities of this gain material we have chosen a compact, astigmatically-compensated Z-cavity laser design . The cavity consists of the following major optical elements: two 50 mm spherical mirrors (AR at 1500–1600 nm and HR at 2000–3000 nm, respectively); flat end mirror (or highly-efficient plane ruled Littrow grating); and a set of removable CaF2 output couplers with measured transmissions of 8.6%, 26.9%, 46.4%, 69.5%, and 74.7% (all AR-coated on the output surface). The pump beam is focused into the gain element by a 75-mm AR-coated pump lens through one of the spherical mirrors. This focal length of the pump lens was found experimentally to be optimal for obtaining the highest output power (among 50 mm, 75 mm, 80 mm, and 100 mm lenses). The estimated pump beam waist radius for the optimal pump lens in the crystal is wp ≈ 55 μm, while the laser mode waist radius is slightly smaller and equals to wL ~ 41 μm.
For efficient thermal control of the gain element, the Cr2+:ZnS crystal is mounted between two thermally-connected copper cold plates (on the 7 × 7 mm surfaces) and the whole copper block is cooled with a 20 W water-cooled Peltier element. Such an arrangement also allows for investigation of the laser performance at different temperatures of the laser crystal.
The 30 W Er-fiber pump laser, available for these experiments, generates an 1.6 mm diameter, high-brightness (M 2 ≈ 1.1), randomly polarized laser beam, making the use of a polarization beam attenuator impractical. Thus the pump power was controlled by the current of the diodes of the pump laser. Due to the nature of the non-polarization maintaining fiber of the pump laser, the output beam polarization state strongly depends on the laser output power and fluctuates unpredictably. Consequently, certain amount of the pump radiation is always reflected on the Brewster input surface of the Cr2+:ZnS gain element, as shown schematically in Fig. 1. The lost portion of the pump radiation Pref was constantly monitored during the experiments to correctly compute the true incident pump power. This computed effective incident pump Pin = P 0 - Pref is always referred to in all further analysis.
3. Experimental results
3.1. Output power, efficiency, and passive losses
The power input-output characteristics of the Cr2+:ZnS laser obtained with 8.6% and 26.9% output couplers (OC) are shown in Fig. 2. The highest obtained output power was 10.5 W at 26 W incident pump with 43% slope- and 40% real optical efficiencies, respectively, at ≈ 2380 nm output wavelength. Although we conducted additional input-output experiments with higher OC transmissions, the 26.9% OC was found to provide the highest output power at the maximum available pump. The graph insert in Fig. 2 shows the dependencies of slope efficiency and maximum obtained output power on the transmission of the output coupler. From this plot one can see that the highest slope efficiency of 47% was obtained with 46.4% OC. Further increase of the OC transmission leads to gradual decrease of the slope efficiency and maximum output power. Therefore, 26.9% OC was considered to be the optimal one for the available pump intensity and existing intracavity passive loss level (see bellow), and all further experiments were performed with this output coupler.
In order to obtain an estimate of the total intracavity passive losses, we performed Findlay-Clay and Caird analysis of the laser performance using the experimental data for the first 3 output couplers with T = 8.6%, 26.9%, and 46.4% (while the slope efficiency was growing with T). The Findlay-Clay and Caird plots are shown in Fig. 3. The Findlay-Clay method gives a round-trip passive loss of about 12%, which corresponds to the single-pass loss of about 8.2%/cm (effective “absorption” coefficient of 0.18 cm-1). The Caird plot shows a very close value of the round-trip loss of 10.3%, which gives approximately 7.2%/cm (or 0.16 cm-1) for the single-pass loss. This loss in our laser system is mainly attributed to the surface scattering due to poor polishing of the crystal facets: the crystal for these experiments was hand-polished on a sequence of fiber-termination abrasive papers (15 μm, 5 μm, 1 μm Al oxide followed by 0.3 μm diamond paper) and considered to be “inspection polished”. It is noteworthy that the estimated Cr2+:ZnS losses are more than twice as small as previously reported values of 14%/cm  and 25%/cm , respectively, for Cr2+:ZnS single crystals. Therefore, we believe that high-quality polishing of the gain element surfaces will allow us to further improve the laser efficiency, as well as its output power and wavelength tunability.
where: Eph-photon energy; Pabsth -threshold absorbed pump power; T-transmission of OC; L = 0.12–estimated passive loss; n–refractive index of the gain medium; wL and wp-laser mode and pump beam radii in the gain element, respectively; τ = 4.3μs-upper-state lifetime. The calculation gives the values of σ ≈ 11.3 × 10-19 cm2, 7.3 × 10-19 cm2, and 8.5 × 10-19 cm2 for OC transmissions T = 8.6%, 26.9%, and 46.4%, respectively. For comparison, the earlier spectroscopic measurements give the value of the stimulated emission cross section of Cr2+:ZnS ranging from 7.5 × 10-19 cm2 in Ref.  to 14 × 10-19 cm2 in Ref. .
3.2. Wavelength tunability
To explore wavelength tunability performance of the high power Cr2+:ZnS laser, the flat end mirror was replaced with a flat, Au-coated, 300 g/mm, replica diffraction grating mounted in the Littrow configuration, as shown in Fig. 1. The choice of the grating parameters was dictated by the maximum damage threshold of its surface at the highest achievable intracavity laser mode intensity and the grating efficiency curve. The results of the wavelength tuning experiment are shown in Fig. 4 and the experiment was conducted as follows.
Initially, the laser was aligned at the minimum threshold pump power. Then the pump power was increased to its maximum and the cavity was further adjusted to obtain the maximum achievable output power, which was found to be 7.4 W at 2410 nm with the optimal 26.9% OC. The laser wavelength at the maximum output power was different from the central wavelength of the non-dispersive cavity version of the laser due to efficiency profile of the diffraction grating. The entire wavelength tuning experiment was performed at the maximum pump by means of the grating rotation in its dispersion plane. No other cavity adjustments were ever made during this experiment (thus “one-knob” wavelength tuning).
One can see from Fig. 4 that the maximum tuning range spans 1940–2780 nm. The output power exceeds 2 W level over a wavelength range of 1970–2760 nm, and 5 W level within 2120–2590 nm. Note that the irregular variations of the output power and some non-smoothness of the tuning curve are due to the pump noise caused by pump beam polarization fluctuations, as explained in Section 2. It is noteworthy that the wavelength tuning range is limited at the lower end by rapidly growing transmissions of the OC and the input (pump) mirror, while at the upper end it is limited by rapidly increasing transmission of the second spherical mirror. Furthermore, it must be mentioned here that the grating efficiency is bellow 98% at all wavelengths, thus leading to additional intracavity loss which decreases the maximum output power.
3.3. Thermal effects
Due to relatively high thermooptic coefficients in II-VI semiconductors [15,16], thermal lensing effects play extremely important role in construction and design of high-power mid-IR lasers based on such gain media as Cr2+:ZnS and Cr2+:ZnSe. In the case of Cr2+:ZnSe, strong lensing effects have been the main limiting factor on obtaining high output power and inspired the researchers for development of advanced thin-disk laser systems . Nevertheless, we have recently demonstrated that even with strong thermal lensing it is possible to use the conventional rectangular slab geometry of the Cr2+:ZnSe gain element for obtaining multiwatt output powers [12,13] exceeding 10 W level. This work represent the first successful attempt to obtain similar high levels of output powers for the Cr2+:ZnS gain media in the same Brewster slab geometry as previously reported for Cr2+:ZnSe lasers.
To perform a qualitative analysis of the thermal lensing effects, we observed the spatial distribution of the transmitted pump beam Pt on a Er-ceramic visualizer screen, as shown schematically in Fig. 1. The screen was located at a distance of 8 cm from the Cr2+:ZnS gain element. The images of the transmitted pump beam profile for several output powers (2 W, 15 W, and 20 W) at lasing conditions are shown in Fig. 5. The last image shows the transmitted pump beam distribution at 20 W when the lasing is blocked.
One can see that initially, at low pump, the pump beam is transmitted without distortion and is circularly symmetric. As the pump power is increased, the thermal lens splits the single incident pump beam into two separate lobs. When at high pump power the lasing is blocked, the transmitted pump beam becomes completely distorted. If then the laser is unblocked, the lasing will not start again until the crystal is cooled by reducing the pump power to a level of about 15 W. This experiment clearly shows that the complex distortion of the pump beam is not a result of some internal imperfections of the gain element or non-planarity of its surfaces, but rather is due to strong non-symmetric thermal lens created by the intense astigmatic pump beam in the Brewster-cut gain element.
This pump beam distortion also strongly influences the laser transverse mode state and leads to a complex transverse mode distribution. Initially, at low pump powers, the laser generates a circularly-symmetric TEM00 mode output beam. As the pump power is increased up to 10 W, the output beam becomes horizontally elliptical, and finally, at the highest pump power the laser switches to a TEM01 mode of operation. Generally, the behavior of the output beam transverse distribution repeats the evolution of the pump beam shown in Fig. 5. The TEM01 transverse mode output beam profile, measured at a distance of about 2.6 m from the laser OC, while the laser was operated at the maximum output power, is shown in Fig. 6. It is noteworthy that while it is possible to adjust the laser cavity for TEM00 transverse mode operation, its output power is significantly reduced (by about 40%) due to a stronger mode mismatch between the low-mode-volume TEM00 laser mode and strongly distorted pump beam in the gain crystal.
The fluorescence lifetime of Cr2+:ZnS laser media drops quickly as the laser crystal is heated above RT, unlike in the case of its Cr2+:ZnSe analog [1, 18]. Consequently, proper thermal management of the Cr2+:ZnS gain element is essential for obtaining high output powers.
In order to investigate the influence of the temperature of Cr2+:ZnS gain element on the laser output power, the following experiment was conducted. The laser (in non-dispersive cavity with the 26.9% OC) was adjusted to the maximum output power at the maximum pump and then the thermal stabilization was turned off. Then the output power was recorded as a function of the temperature of Cr2+:ZnS crystal. The result of this experiment is shown in Fig. 7, superimposed with the dependence of Cr2+:ZnS fluorescence lifetime on its temperature .
On can see that as the gain element temperature rises from 8.5°C to about 21°C, the output power gradually decreases from 10 W to 8.6 W (by 14% from the maximum). Then we observe a sudden drop of the output power to 5 W level (50% from the maximum). Afterwards the output power continues to reduce gradually to about 2.4 W as the temperature reaches 45°C. It is noteworthy that the temperature was measured at the crystal holder, and thus the actual temperature of the gain element region where the laser radiation was generated is higher than the measured value. The gradual decrease of the output power with temperature is associated with the decreasing of fluorescence lifetime due to thermal quenching of Cr2+:ZnS media with rising temperature. However, the sharp fall of the output power at about 21°C can be explained by a nonlinear behavior of the laser system using the following qualitative model.
The slope efficiency and lasing threshold depend on the temperature of the Cr2+:ZnS gain element. In addition to the slow effect of decreasing emission lifetime with growing temperature, a decrease of output power and laser efficiency can be caused by the appearance of a strong non-symmetric thermal lens, which significantly reduces the cavity Q-factor and degrades the mode-matching conditions. On the other hand, the laser oscillation works as additional chilling mechanism of the gain element: due to fast transfer of the pump energy into output radiation the temperature of the gain element is reduced with growing output power (e.g., we directly observed about 6°C temperature rise of Cr2+:ZnS crystal when the cavity was blocked at full pump power). Therefore, this negative feedback mechanism results in a strongly nonlinear behavior and optical bistability of the laser system. A careful realignment of the laser cavity at 21°C allows to return the output power back to about 8.5 W, however, similar behavior is observed at other temperatures, e.g. near 13°C, 32°C, and 37° as one can see in Fig. 7. It is noteworthy that similar observations of optical bistability caused by thermal effects were reported for other types of chromium-doped laser systems .
In this work we have demonstrated 10 W pure CW, Cr2+:ZnS laser system pumped by a 1.56 μm Er-fiber laser. We obtained optical efficiency on the incident pump power as high as 40% and the slope efficiency of about 43% with no noticeable output power roll-off at high pump powers, which indicates that much higher output powers are achievable with appropriate, more powerful pump sources. The laser shows broadband high-power wavelength tunability spanning 1940–2780 nm, with the output power exceeding 2 W level over wavelength range of 1970–2760 nm, and above 5 W level within 2120–2590 nm. This achieved level of output power corresponds to more than 14-fold increase over the best previously reported result for CW Cr2+:ZnS laser , and almost 2-fold increase as compared to our published record results for the Cr2+:ZnSe CW laser with the same pump source .
We acknowledge support from the National Science Foundation Grants No. ECS-0424310, EPS-0447675, EPS-0814103, and BES-0521036.
References and links
1. L. D. DeLoach, R. H. Page, G. D. Wilke, S. A. Payne, and W. F. Krupke, “Transition metal-doped zinc chalcogenides: spectroscopy and laser demonstration of a new class of gain media,” IEEE J. Quantum Electron. 32, 885–895 (1996). [CrossRef]
2. R. H. Page, K. I. Schaffers, L. D. DeLoach, G. D. Wilke, F. D. Patel, J. B. Tassano, S. A. Payne, W. F. Krupke, K. T. Chen, and A. Burger, “Cr2+-doped zinc chalcogenides as efficient, widely tunable mid-infrared lasers,” IEEE J. Quantum Electron. 33/4, 609–619 (1997). [CrossRef]
3. S. B. Mirov, V. V. Fedorov, K. Graham, I. S. Moskalev, I. T. Sorokina, E. Sorokin, V. Gapontsev, D. Gapontsev, V. V. Badikov, and V. Panyutin, “Diode and fibre pumped Cr2+:ZnS mid-infrared external cavity and microchip lasers,” IEE Optoelectronics 150(4), 340–345 (2003). [CrossRef]
4. S. B. Mirov, V. V. Fedorov, K. Graham, and I. S. Moskalev, “Erbium fiber laser-pumped continuous-wave microchip Cr2+:ZnS and Cr2+:ZnSe lasers,” Opt. Lett. 27(11), 909–911 (2002). [CrossRef]
5. U. Hommerich, X. Wu, V. R. Davis, S. B. Trivedi, K. Grasza, R. J. Chen, and S. Kutcher, “Demonstration of room temperature laser action at 2.5μm from Cr2+:Cd0.85Mn0.15Te,” Opt. Lett. 22, 1180–1182 (1997). [CrossRef] [PubMed]
6. J. Mckay, K. L. Schepler, and G. C. Catella, “Efficient grating tuned mid-infrared Cr2+:CdSe laser,” Opt. Lett. 24, 1575–1577 (1999). [CrossRef]
7. G. J. Wagner, T. J. Carrig, R. H. Page, K. I. Schaffers, J. O. Ndap, X. Ma, and A. Burger, “Continuous-wave broadly tunable Cr2+:ZnSe laser,” Opt. Lett. 24, 19–21 (1999). [CrossRef]
8. G. J. Wagner and T. J. Carrig, “Power scaling of Cr2+:ZnSe lasers,” in OSA Trends Opt. Photonics, Adv. Solid State Lasers , vol. 50, pp. 506–510 (2001).
9. T. J. Carrig, G. J. Wagner, W. J. Alford, and A. Zakel, “Chromium-doped chalcogenides lasers,” in Proc. SPIE , vol. 5460 of Solid State Lasers and Amplifiers, pp. 74–82 (2004). [CrossRef]
11. S. Mirov, V. Fedorov, I. Moskalev, and D. Martyshkin, “Recent progress in transition metal doped II-VI mid-IR lasers,” IEEE J. Sel. Top. Quantum Electron. 13(3), 810–822 (2007). [CrossRef]
13. I. S. Moskalev, V. V. Fedorov, S. B. Mirov, P. A. Berry, and K. L. Schepler, “12-Watt CW Polycrystalline Cr2+:ZnSe Laser Pumped by Tm-Fiber Laser,” in Advanced Solid-State Photonics, p. WB30 (OSA, Denver, Colorado, 2009).
14. D. M. Simanovskii, H. A. Schwettman, H. Lee, and A. J. Welch, “Midinfrared Optical Breakdown in Transparent Dielectrics,” Phys. Rev. Lett. 91(10), 107,601–1 (2003). [CrossRef]
15. D. N. Nikogosyan, Properties of Optical and Laser-Related Materials: A Handbook, chap. 5.5, pp. 278–294 (John Wiley & Sons, 1997).
16. R. J. Harris, G. T. Johnston, G. A. Kepple, P. C. Krok, and H. Mukai, “Infrared thermooptic coefficient measurement of polycrystalline ZnSe, ZnS, CdTe, CaF2, and BaF2 , single crystal KCI, and TI-20 glass,” Appl. Opt. 16, 436–438 (1977). [CrossRef] [PubMed]
17. I. T. Sorokina, E. Sorokin, S. B. Mirov, V. V. Fedorov, V. Badikov, V. Panyutin, and K. Schaffers, “Broadly tunable compact continuous-wave Cr2+:ZnS laser,” Opt. Lett. 27, 1040–1042 (2002). [CrossRef]
18. I. T. Sorokina, E. Sorokin, S. B. Mirov, V. V. Fedorov, V. Badikov, V. Panyutin, A. DiLieto, and M. Tonelli, “Continuous-wave tunable Cr2+:ZnS laser,” Appl. Phys. B, Laser Opt. 74, 607–611 (2002). [CrossRef]
19. H. W. Kogelnik, A. Dienes, and C. V. Shank, “Astigmatically Compensated Cavities for CW Dye Lasers,” IEEE J. Quantum Electron. QE-8(3) (1972).
20. P. F. Moulton, “An investigation of the Co:MgF2 laser system,” IEEE J. Quantum Electron. QE-21, 1582–1595 (1985). [CrossRef]
21. P. F. Moulton, “Spectroscopic and laser characteristics of Ti:Al2O3,” J. Opt. Soc. Am. B 3, 125–133 (1986). [CrossRef]
22. K. L. Schepler, R. D. Peterson, P. A. Berry, and J. B. McKay, “Thermal Effects in Cr2+:ZnSe Thin Disk Lasers,” IEEE J. Sel. Top. Quantum Electron. 11, 713–720 (2005). [CrossRef]
23. M. A. Noginov, B. D. Lucas, and M. Vondrova, “Optical bistability in a Cr:LiSrGaF6 laser,” J. Opt. Soc. Am. B 19, 1999–2006 (2002). [CrossRef]