Characteristics of an Fe:ZnSe laser with a new pump scheme were presented. An Fe:ZnSe crystal was placed inside the cavity of the Er:YLF laser pumped by a bar of laser diodes (LD) emitting at 975 nm. The high power 2.66 μm wave ~50 ns pulses were generated inside the cavity due to passive Q-switching of the Er:YLF laser by an Fe:ZnSe saturated absorber. These pulses pumped the Fe:ZnSe laser. As a result, the pulses with an energy of ~2 μJ at a wavelength around 4 μm and time duration of about 50 ns were generated. The laser operated in periodic-pulsed pump regime at room temperature with a repetition rate up to 200 Hz. The Q-switching pulse repetition rate during the pump pulse ran up to 2.5 kHz.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
An Fe:ZnSe laser is the most perspective one for 4-4.5 μm spectral range [1–3] where one of the atmospheric-transmission bands exists. The laser is also interesting for medicine, spectroscopy, metrology.
The best results with this laser were achieved at cooling Fe:ZnSe crystal by liquid nitrogen [4,5]. This relates to the drop of internal quantum efficiency of laser transition at increasing crystal temperature due to activation of multiphonon non-radioactive recombination. Recently we have implemented the pulsed Fe:ZnSe laser with pulse energy as high as 7.5 J at crystal temperature of 220 K by means of using thermoelectric cooling . However the laser threshold of the Fe:ZnSe laser at room temperature (RT) is too high to achieve lasing in quasi continue wave regime. Until now, RT Fe:ZnSe lasing has been achieved only at pumping by high power nanosecond pulses of other lasers operated in the 2.5-3 μm spectral range. One of these lasers is a pulsed electrodischarge HF laser. By using this pumping, a 1.67 J pulse Fe:ZnSe laser was implemented at RT . In periodic-pulsed regime, the average Fe:ZnSe laser power can achieve 2.4 W during a 1 c train of pulses . These lasers are inconvenient for practical use because of the toxicity of the medium used and the cumbersome structure of the device.
The other extensively used pump laser is Q-switched Er:YAG laser . The Er:YAG laser operated in the regime of passive Q-switching by Fe:ZnSe crystal can emit a train of giant laser pulses with a total energy of 0.5 J . Until now such Er-lasers have operated with lamp pumping which is not suitable for the most applications. Even though the Q-switching regime of Er:YAG operation was obtained at laser diode (LD) pumping, the energy of short pulses was not enough for pumping Fe:ZnSe laser at RT .
In recent years, Er:YLF crystals have been used to achieve low-threshold lasing at LD pumping [12–14]. Nanosecond pulses of 3 mJ in energy were achieved with a passive Q-switched Er:YLF laser . But until now, the efficiency of such laser has been only ~1%. In  we proposed a new pumping scheme for the Fe:ZnSe laser operating at RT which is based on LD pumping. In this scheme a Fe:ZnSe crystal plate, placed inside the cavity of the Er:YLF laser, operates as passive switcher and active medium simultaneously. The Fe:ZnSe lasing was achieved only by using additional high quality resonator for Fe:ZnSe crystal. As a result, nanosecond pulses with energy of 0.5 μJ or less were obtained.
In the present paper we further develop this pump method. Unlike in previous work, in the current one we have used a new Fe:ZnSe crystal with higher Fe2+ concentration (~2.81018 cm−3), larger absorption being ~25% per one pass at Er:YLF laser wavelength (2.66 μm) and only one cavity for both the Er:YLF and Fe:ZnSe lasers. As a result we have obtained higher pulse energy (2 μJ) of the Fe:ZnSe laser. We also present the laser spectra and the data of divergence in detail.
Note that such an approach has been used earlier but with different laser and switcher crystal [16,17]. In optical scheme utilizing Nd-laser, three-mirror cavity and LiF crystal with F2+ color centers operated simultaneously as passive Q-switch and active laser medium, 75 percent efficiency of Nd3+ to F2+:LiF radiation conversion was achieved.
Figure 1 shows the experimental setup.
The Er:YLF crystal had the Er concentration of 15%. Its dimensions were 28 mm in length (along the crystal c-axis), 4 mm in width and 2 mm in thickness. The polished surfaces of 2x4 mm2 had antireflective coating at 2.7 μm and were oriented perpendicular to the optical axis of the Er:YLF laser cavity. The transmission of the Er:YLF crystal with the antireflective coating along its length of 28 mm was about 87% at 4.4 μm. The crystal was glued by epoxy to the Au-mirror on sapphire substrate attached to the Cu heat conductor. The 3 mm thick MgF2 plate was glued to the back surface of the Er:YLF crystal and served as a heat spreader.
The Er:YLF crystal was pumped through the surface of 4x28 mm2 along thickness of 2 mm (along the crystal a-axis) by a laser diode bar operating in continuous wave (CW) regime with maximum power of 53 W at 975 nm. The output power of the LD bar passed through the optical system and the power reflected from the unit with the Er:YLF crystal were measured by using a power meter Gentec TPM-300. This device was also used for measuring of the average output power of the Er:YLF and Fe:ZnSe laser beams. The optical system including collimating cylindrical lens, spherical lens and focusing cylindrical lens formed an excitation area of 28 mm in length and about 50 μm in width. Pump losses in the optical system was measured as 10%. Besides, according to the measurement, about 35% of the incident radiation pump power was not absorbed in Er:YLF crystal due to its small optical density. Hence the Er:YLF crystal was excited along whole thickness of 2 mm and maximum absorbed power was only ~30 W. To decrease heating of the Er:YLF crystal we used chopper disc placed between the collimating lens and the spherical lens in the optical system. The duty factor was 0.1. The maximum pulse repetition rate (ν) was 200 Hz.
The Er:YLF laser cavity was formed by the Al spherical concave mirror with the curvature radius of 100 mm and the OC1 or OC2 having high reflectivity at wavelength around 2.7 μm and being partly transparent at λ > 3.8 μm. When the Fe:ZnSe crystal plate was inside the cavity, the OC2 was used. Additional mirrors M1 or M2 (exit mirrors for the Fe:ZnSe laser) were also used in some experiments to change spectral characteristics of the Fe:ZnSe laser. The additional mirrors were coated on the 2 mm thick sapphire substrates. They were placed near the OC2. Reflection spectra of the OC1, OC2 and additional mirrors M1, M2 used in this work are presented in Fig. 2(a). The reflection of the Al mirror was ~98% at λ = 2.7 μm and 4 μm. Physical cavity length was about 100 mm.
The 2 mm thick Fe:ZnSe crystal plate was cut from a crystal grown by the physical vapor transport method using a seed . Its transmission spectrum is presented in Fig. 2(b). From this spectrum, the Fe2+ concentration was estimated as 2.8·1018 cm−3. Absorption at λ = 2.7 μm is about 28%. That is 3 times higher than the one we have used in our previous work . The plate surfaces were polished and parallel to each other with accuracy of 30”. The plate surface normal was parallel to the cavity axis. All optical elements could be aligned independently.
Time performances were analyzed by photodiodes (IBSG Co., Ltd.) and digital oscillograph (Tektronix TDS 2024B) with time resolution of about 5 ns. To reject the Er:YLF laser radiation at λ ≈2.7 μm and select 4 μm radiation a filter similar to the OC2 was used before a photodiode PD-1 with a photoelectric threshold of 4.8 μm. The filter had high reflectivity at λ ≈2.7 μm. It was placed at 30° angle to the optical axis. Reflected radiation was recorded by a PD-2 with a photoelectric threshold at λ ≈3.6 μm. That is how we could record pulses of both the Er:YLF laser and the Fe:ZnSe laser simultaneously. Laser spectra were recorded by using a monochromator based on grating with 150 grooves per 1 mm and a photodiode PD-1 placed at the output slit. Laser divergence was measured at the distance of 300 mm from the OC2. Laser intensity distribution was recorded by the photodiode PD-1 with receiving area of 0.2x0.2 mm2. Photodiode signal was averaged over 16 pump pulses.
3. Results and discussion
Figure 3 shows the dependences of the absorbed peak pulse power in the Er:YLF crystal on a CW LD current and a LD output power. These dependences were calculated based on measured data of pump losses in the optical system and reflected pump. The dependences are not exactly linear because the width of the LD emission line increases and its maximum shifts to long wave side with increasing LD current. As a result, the covering of pump spectrum with the Er:YLF crystal absorption spectrum changes with a LD current.
Oscillograms of Er:YLF laser output pulses without the Fe:ZnSe crystal inside the cavity and with the OC1 are presented in Fig. 4(a).
The LD current threshold at the pump pulse duration of ~5 ms (ν = 20 Hz) was 11.5 A that corresponds to the average absorbed pump power of ≈0.4 W. At the pump current of 62 A, the Er:YLF laser pulse consisted of two parts (see ). The laser wavelength during the first part of the pump pulse was around 2.66 μm. Then the laser wavelength jumped to 2.71 μm. The threshold of lasing at 2.71 μm was about 33 A. Such a wavelength jumping is typical for a Er:YLF laser and has been studied in . It is a result of filling of low laser level. Unfortunately we were unable to achieve lasing at 2.81 μm observed in [12,14] because of the threshold of such lasing was too high for our set-up.
Figure 5(a) shows the dependence of average output power of the Er:YLF laser on the average absorbed pump power at ν = 100 Hz. The threshold increased to ~0.6 W with increasing ν. Maximum average power of the laser at the pulse repetition rate of 100 Hz was about 66 mW. At the duty factor of 0.1 it corresponds to the peak power of 0.66 W. Laser slope efficiency was ~2.8%, much less than slope efficiency of 15% achieved with using of a multipass slab scheme in . One of the reasons of such a low efficiency is likely to be the bad covering of a laser mode volume with an excitation volume. The use of the OC2 with higher reflectivity (99.4%) instead of the OC1 (90%) decreased appreciably the average laser output power while increased intracavity power.
After placing the Fe:ZnSe crystal plate inside the Er:YLF laser cavity, the short pulse regime was achieved. Figure 4(b) shows oscillogram with 5 short pulses per one pump pulse at LD current I = 62 A. Sometimes 6 pulses appeared per a pulse. However, these short pulses occurred only during the first 2-2.5 ms of the long pump pulse. This is a result of the wavelength jumping which takes place both at free-run and Q-switch regimes . The threshold of the Q-switch regime at 2.66 μm was about Ith = 30 A while at 2.71 μm it was higher than the maximum pump used.
The time delay of the first short laser pulse in relation to the pump pulse decreased with increasing the LD current and was about 0.4 ms at I = 62 A. The average time interval between two consecutive short pulses was the same. It means that the repetition rate of the Q-switched pulses is about 2.5 kH. The dependence of average output power of the Fe:ZnSe laser on the average absorbed pump power at ν = 100 Hz is presented in Fig. 5(b).
One of the short pulses recorded by a PD-1 without any filter is presented in Fig. 6(a) (at monochromator entrance slit). The pulse consisted of two components. While only one component was observed at the output slit of the monochromator adjusted on λ = 3.95 μm. We believe that the one component corresponds to the Er:YLF laser while the other more intense time-delayed component is the Fe:ZnSe laser pulse. Both these pulses were separated by using a PD-1 with the filter and a PD-2 (see Fig. 1). Figure 6(b) demonstrates the result of such a separation. The Fe:ZnSe laser pulse is delayed by 20-30 ns. This delay is one of the strong arguments that this pulse is a result of lasing. The pulse was formed during 30-45 around trips of the cavity.
Laser spectra are presented at Fig. 7. Without any additional mirrors, the spectrum contains the line near 2.66 nm due to Er:YLF lasing and two broad lines near 3.97 μm and 4.45 μm which are related to Fe:ZnSe lasing. The line width of the Er:YLF laser was determined by the monochromator slits which were 1 mm. At these slits, the spectral resolution of the monochromator was 0.02 μm. The Fe:ZnSe laser spectrum correlates with the reflectivity spectrum of the OC2. To change the Fe:ZnSe laser spectrum we used the additional mirrors M1 and M2. The Fe:ZnSe laser spectra with the M1 and M2 have a dip in the 4.2-4.3 μm range due to intracavity absorption by atmospheric CO2. Note that the spectrum depends strongly on spectral characteristics of the mirrors. Besides the deep CO2 notch in the spectrum may be formed only if radiation beam passes a long distance. In our case the long distance is multiple rounds of the cavity. These features evidence lasing nature of the Fe:ZnSe radiation.
The laser spectrum observed is unusually broad. Although the Fe:ZnSe laser medium is characterized by the broad amplification band from 3.77 to 5.05-μm the typical laser line has a width of ~100 nm . At CW regime the line width may be even less 1 nm . At pulse regime with high threshold excess, the spectrum may be wider. Besides the wide (2 mm) excitation area may be divided into several independent lasers with different spectra.
The number of the short pulses per one pump pulse might be decreased by decreasing the pump pulse duration. In our case we achieved it by increasing rotating velocity of the chopper disk (see Fig. 1). At that the pump pulse repetition rate ν was increased also. Figure 8 demonstrates oscillograms of laser output at different ν. They were recorded by a PD-1 with the filter. Therefore they are the oscilograms of the Fe:ZnSe laser.
At ν = 200 Hz, typically one short pulse occurred per a pump pulse at I = 62 A. Two short pulses were observed at ν = 100 Hz. Peak short pulse power was not stable from pulse to pulse (see Fig. 8 at ν = 200 Hz). Peak power deviation from the average value of peak power might be 30% at I = 62 A.
At long pump pulse (≥ 1 ms), the laser short pulse peak little depended on the LD current. Increase of the pump power led to increasing a number of generated short pulses during the pump pulse. Hence the average output power increased with increasing the LD current or the average absorbed pump power (see Fig. 5(b)). The average power of the Fe:ZnSe laser operating at periodic-pulsed regime with ν = 100 Hz and the duty factor of 0.1 was measured as 0.4 mW ± 0.1 mW. Taking into account that two short pulses were generated per a pump pulse at ν = 100 Hz we estimated the short pulse energy of the Fe:ZnSe laser beam as 2 μJ. The limit pulse repetition rate of the short pulses is determined by the time interval between the consecutive pulses and at I = 62 A it may be as high as 2.5 kHz. At that the laser should operate in a continuous wave regime. However we observed that the short pulses appeared only during the first 2-2.5 ms of a long duration pump pulse (see Fig. 4(b)). The reason of such a laser break is likely to relate to complicated excitation kinetics of electron levels of Er ions during pumping that is evident in the laser wavelength jumping (see Fig. 4(a)).
Figure 9 shows a distribution of the Fe:ZnSe laser pulse energy in the horizontal and vertical planes at a distance of 300 mm from the output coupler. As evident from the figure the Fe:ZnSe laser operated in a multimode regime. The full width at half maximum (FWHM) of the energy distribution in the horizontal plane is larger than one in the vertical plane. It is associated with an asymmetrical distribution of the pump. The FWHM of angle distribution energy of the Fe:ZnSe laser was 31 mrad and 28 mrad for the horizontal and the vertical planes respectively. These values are more than two times higher than the ones for the Er:YLF laser. Better directivity of the Er:YLF laser radiation is explained by additional mode selection due to large gain length. The directivity of the Fe:ZnSe laser is likely to be improved by using two Fe:ZnSe plates, one placed near the OC and other placed near the high reflectivity mirror.
Small difference in the x- and y-directivity of the Fe:ZnSe emission is an additional argument for lasing. If the output beam was formed by amplified spontaneous emission (ASE) its directivity would be determined by geometrical size of Fe:ZnSe excited area. The ASE x-directivity should be much worse than the y-directivity because of strong difference in the depth (2 mm) and the width (0.05 mm) of the excited area.
Future improvement of laser characteristic is expected to relate to realizing the CW regime of the Er:YLF laser, increasing its efficiency and pump power.
We studied a new optical schema containing a LD side-pumped Er:YLF crystal and a Fe:ZnSe crystal plate placed in one cavity. The Fe:ZnSe crystal plate operated as a passive switcher for the Er:YLF laser. On the other hand, the generated giant pulses pumped the Fe:ZnSe crystal and their energy converted into energy of nanosecond pulses of the Fe:ZnSe laser. Average energy of the Fe:ZnSe laser pulses was ~2 μJ and its time duration was ~50 ns. The Fe:ZnSe laser spectrum was in the 3.9-4.45 μm range and depended on the spectral characteristics of the output coupler. Divergence of the Fe:ZnSe laser radiation was two times higher than that of the Er:YLF laser. The total angle of divergence at half maximum of the Fe:ZnSe laser was 31 mrad in the pump horizontal plane and 28 mrad in the vertical plane. The laser operated in the periodic-pulsed regime of pumping at room temperature. The maximum pump pulse repetition rate was 200 Hz and the duty factor was 0.1. Though the pulse energy and average power of the laser achieved in this work still are not high enough, the scheme allows improving these laser parameters by means of increasing LD power.
Presidium Program of the Russian Academy of Sciences (No. 7 “Actual questions of photonics, sounding of nonuniform mediums and materials”); Competitiveness Program of the National Research Nuclear University (MEPhI) (02.a03.21.0005).
1. J. J. Adams, C. Bibeau, R. H. Page, D. M. Krol, L. H. Furu, and S. A. Payne, “4.0-4.5-µm lasing of Fe:ZnSe below 180 K, a new mid-infrared laser material,” Opt. Lett. 24(23), 1720–1722 (1999). [CrossRef] [PubMed]
2. V. A. Akimov, A. A. Voronov, V. I. Kozlovsky, Y. V. Korostelin, A. I. Landman, Y. P. Podmar’kov, and M. P. Frolov, “Efficient IR Fe:ZnSe laser continuously tunable in the spectral range from 3.77 to 4.40 μm,” Quantum Electron. 34(10), 912–914 (2004). [CrossRef]
3. V. V. Fedorov, S. B. Mirov, A. Gallian, V. V. Badikov, M. P. Frolov, Y. V. Korostelin, V. I. Kozlovsky, A. I. Landman, Y. P. Podmar’kov, V. A. Akimov, and A. A. Voronov, “3.77-5.05-μm tunable solid-state lasers based on Fe2+-doped ZnSe crystals operating at low and room temperatures,” IEEE J. Quantum Electron. 42(9), 907–917 (2006). [CrossRef]
4. V. I. Kozlovsky, Y. V. Korostelin, Y. P. Podmar’kov, Y. K. Skasyrsky, and M. P. Frolov, “Middle infrared Fe2+:ZnS, Fe2+:ZnSe and Cr2+:CdSe lasers: new results,” J. Phys.: Conf. Ser. 740, 012006 (2016).
5. S. Vasilyev, I. Moskalev, M. Mirov, V. Smolski, D. Martyshkin, V. Fedorov, S. Mirov, and V. Gapontsev, “Progress in Cr and Fe doped ZnS/Se mid-IR CW and femtosecond lasers,” Proc. SPIE 10193, 101930U (2017).
6. M. P. Frolov, Y. V. Korostelin, V. I. Kozlovsky, Y. P. Podmar’kov, and Y. K. Skasyrsky, “High-energy thermoelectrically cooled Fe:ZnSe laser tunable over 3.75-4.82 μm,” Opt. Lett. 43(3), 623–626 (2018). [CrossRef] [PubMed]
7. K. N. Firsov, M. P. Frolov, E. M. Gavrishchuk, S. Y. Kazantsev, I. G. Kononov, Y. V. Korostelin, A. A. Maneshkin, S. D. Velikanov, I. M. Yutkin, N. A. Zaretsky, and E. A. Zotov, “Laser on single-crystal ZnSe:Fe2+ with high pulse radiation energy at room temperature,” Laser Phys. Lett. 13(1), 015002 (2016). [CrossRef]
8. S. D. Velikanov, N. A. Zaretskiy, E. A. Zotov, V. I. Kozlovsky, Y. V. Korostelin, O. N. Krokhin, A. A. Maneshkin, Y. P. Podmar’kov, S. A. Savinova, Y. K. Skasyrsky, M. P. Frolov, R. S. Chuvatkin, and I. M. Yutkin, “Investigation of Fe:ZnSe laser in pulsed and repetitively pulsed regimes,” Quantum Electron. 45(1), 1–7 (2015). [CrossRef]
9. V. A. Akimov, A. A. Voronov, V. I. Kozlovsky, Y. V. Korostelin, A. I. Landman, Y. P. Podmar’kov, and M. P. Frolov, “Efficient lasing in a Fe2+:ZnSe crystal at room temperature,” Quantum Electron. 36(4), 299–301 (2006). [CrossRef]
10. S. D. Velikanov, E. M. Gavrishchuk, N. G. Zakharov, V. I. Kozlovsky, Y. V. Korostelin, V. I. Lazarenko, A. A. Maneshkin, Y. P. Podmar’kov, E. V. Saltykov, Y. K. Skasyrsky, M. P. Frolov, V. S. Tsykin, R. S. Chuvatkin, and I. V. Yutkin, “Efficient operation of a room-temperature Fe2+:ZnSe laser pumped by a passively Q-switched Er: YAG laser,” Quantum Electron. 47(9), 831–834 (2017). [CrossRef]
11. A. A. Voronov, V. I. Kozlovsky, Y. V. Korostelin, A. I. Landman, Y. P. Podmar’kov, V. G. Polushkin, T. I. Ragimov, Y. K. Skasyrsky, M. Y. Filipchuk, and M. P. Frolov, “Passive Q-switching of the diode-pumped Er:YAG laser cavity with the Q-switch based on the Fe2+:ZnSe crystal,” Bull. Lebedev Phys. Inst. 37(6), 169–172 (2010). [CrossRef]
12. A. Dergachev, J. H. Flint, Y. Isyanova, B. Pati, E. V. Slobodtchikov, K. F. Wall, and P. F. Moulton, “Review of multipass slab laser system,” IEEE J. Sel. Top. Quantum Electron. 13(3), 647–660 (2007). [CrossRef]
14. M. V. Inochkin, V. V. Nazarov, D. Y. Sachkov, L. V. Khloponin, V. Y. Khamov, Y. V. Korostelin, A. I. Landman, Y. P. Podmar’kov, and M. P. Frolov, “A compact Er:YLF laser with a passive Fe2+:ZnSe shutter,” J. Opt. Technol. 79(6), 337–339 (2012). [CrossRef]
15. V. I. Kozlovsky, Y. V. Korostelin, Y. K. Skasyrsky, and M. P. Frolov, “Nanosecond room-temperature Fe: ZnSe laser pumped inside the resonator of a transversely diode-pumped Er: YLF laser,” Quantum Electron. 48(8), 686–690 (2018). [CrossRef]
16. T. T. Basiev, B. I. Denker, S. B. Mirov, V. V. Osiko, V. G. Park, A. M. Prokhorov, I. Kertesz, N. Kroo, and K. Ferencz, “Intracavity 1.05-1.15-μm radiation conversion in lasers based on Nd+ -.glass. Lithium fluoride crystal with F2- color centers and three-mirror resonator,” Adv. Sol. St. Lasers 5, 332–333 (1989).
17. W. de Rossi, F. E. Costa, N. D. Vieira Jr, N. U. Wetter, S. P. Morato, T. T. Basiev, V. A. Konyushkin, and S. B. Mirov, “Mode-locking operation of a pulsed Nd:YAG laser with F2 BF:LiF color-center crystal in a dual configuration,” Appl. Opt. 31(15), 2719–2721 (1992). [CrossRef] [PubMed]
18. V. I. Kozlovsky, V. A. Akimov, M. P. Frolov, Y. V. Korostelin, A. I. Landman, V. P. Martovitsky, V. V. Mislavskii, Y. P. Podmar’kov, Y. K. Skasyrsky, and A. A. Voronov, “Room-temperature tunable mid-infrared lasers on transition-metal doped II-VI compound crystals grown from vapour phase,” Phys. Status Solidi, B Basic Res. 247(6), 1553–1556 (2010). [CrossRef]
19. J. W. Evans, R. W. Stites, and T. R. Harris, “Increasing the performance of an Fe:ZnSe laser using a hot isostatic press,” Opt. Mater. Express 7(12), 4296–4303 (2017). [CrossRef]