Total-reflection active-mirror amplifier for high pulse energy and high average power by using a composite ceramic

Highlights ► By combining the “Thin–Disk” concept with a composite ceramic, a total-reflection active-mirror (TRAM) amplifier is functionally designed for simultaneous operation in both high pulse energy and high average power. ► A chirped-pulse regenerative amplifier with a cryogenic TRAM has been successfully demonstrated, showing 3.5 mJ pulse energy at 100 Hz. ► The corresponding energy fluence was high at 1.5 J/cm 2 for 0.46 short pulses. The M 2 -factor was below 1.1 and no significant beam distortion was observed. Abstract A total-reflection active-mirror (TRAM) amplifier is functionally designed by using a composite ceramic for high pulse energy and high average power. A chirped-pulse regenerative amplifier with a cryogenic TRAM has been successfully demonstrated as a feasibility study. A 3.5 mJ pulse energy is obtained at a repetition rate of 100 Hz. The corresponding energy fluence on the TRAM is as high as 1.5 J/cm 2 for 0.46 ns short pulses. The M 2 -factor is below 1.1 and no significant beam distortion is observed. Keywords Ceramic lasers Diode-pumped lasers Solid-state lasers 1 Introduction High power lasers in the next generation, which have both high pulse energy and high average power simultaneously, are strongly desired for the advanced scientific and industrial applications such like laser fusion energy [1] , debris remove, laser lightening, neutron source and so on. There are inevitable key issues in the development of such high power lasers, especially optically induced damages and thermal effects of active materials. Designs of amplifier structures and researches of laser materials are important ways against these key issues. One of the most promising laser amplifier structures to reduce thermal effects is the “Thin-Disk” laser [2] . The thinness less than 200-μm of the laser disk considerably reduces its temperature rise. As a heat flow in the disk is along the disk thickness as well as the amplifying laser propagation, the radial temperature gradient of the disk is ideally zero, which leads to no thermal lens effect. Using the “Thin-Disk”, many cw and quasi-cw lasers at high average power have been developed [3–5] . Recently, using the excellent power scalability, a high power laser system up to 16 kW is commercially obtainable. Only a few high pulse energy operations, however, have been demonstrated [6,7] . A reduction of the optically induced damage threshold due to the interference of the laser pulse on the input/output surface is significant, especially, in use of an ytterbium-doped material as a laser material. Because the saturation fluence U sat of the materials is generally comparable to or higher than the typical damage threshold for nano-second pulses, for example, U sat ∼ 10 J/cm 2 for an Yb:YAG crystal. It is, therefore, difficult to extract the storage energy efficiently without optical damages. On the view of the laser material research, a laser ceramic with a high optical quality has been developed in 1995 [8] . Some advantages of the ceramics such as size scaling and high ion concentration have been demonstrated with Nd- or Yb-doped ceramics though the host ceramics are limited to YAG and sesquioxide up to today [9,10] . Diode-pumped 10 cm-class Nd:YAG ceramic disks enable 67 kW heat capacity laser system [11] . An additional advantage is a composite ceramic, which increases the degrees of freedom in an amplifier structure design. A Nd:YAG/YAG cylindrical composite rod improved a spatial beam overlap between the pump area and the laser beam. Using eight amplifiers with the composite ceramic rods in series, a high average power of 5.24 kW has been demonstrated with a high optical efficiency of 17% [12] . Combining the “Thin-Disk” concept and the composite ceramic technology, total-reflection active-mirror (TRAM) has been proposed as a novel amplifier for both average power and high pulse energy. A Yb:YAG/YAG TRAM laser oscillator at liquid nitrogen temperature has been already demonstrated in cw operation to certify its high thermal strength [13] . The high output power of 273 W is obtained with a high optical efficiency of 65% and a high slope efficiency of 72%. In this letter, a cryogenic TRAM with a Yb:YAG/YAG composite ceramic was used in a chirped-pulse regenerative amplifier as a feasibility study for the simultaneous operation in high pulse energy and high average power. A high energy fluence of 1.5 J/cm 2 on the TRAM is obtained for 0.46 ns short pulses at a repetition rate of 100 Hz. The M 2 -factor is less than 1.1 and no significant beam distortion is observed. 2 Total-reflection active-mirror (TRAM) amplifier In an active-mirror amplifier such like the “Thin-Disk,” a local enhancement of the electric field due to the interference of the laser beam on the input/output surface reduces the optically induced damage threshold. The damage threshold in energy fluence is reduced to one forth, 2.5 J/cm 2 for 1 ns pulses if a typical surface damage threshold is 10 J/cm 2 [14] . Recently an ytterbium-doped material is often focused on as a high power laser material and its saturation fluence is considerably high. It is difficult to extract a stored energy efficiently without optical damages. The TRAM is functionally designed against the optical damages. It is a monolithic composite ceramic with a non-doped YAG prism and a thin Yb:YAG active layer, shown in Fig. 1 . The prism is used to avoid the interference of the laser beam by separating an input and an output surface. In addition, Brewster-input and -output increase an operatable pulse energy per a unit cross section of a laser beam because the effective energy fluence on the Brewster surface reduces. A pump beam is absorbed by the Yb:YAG thin layer. A linearly polarized seed pulse is propagating into the TRAM through a roof of the prism and reflected on the bottom surface of the active layer to go out of the TRAM. The “Thin-Disk” concept of superior thermal strength and power scaling is inherited. One of additional ideas of the TRAM is a total reflection, which is used on the back surface of the Yb:YAG layer instead of the high-reflection (HR) coating. A considerable temperature rise at the HR-coating due to its low thermal conductivity, therefore, is not imposed on the TRAM. Also, no coat reduces a cost. Then, the TRAM is used at low temperature to improve material coefficients according to high power lasers, emission cross section, thermal conductivity, thermo-optic coefficient and thermal expansion coefficient [15–17] . 3 Regenerative amplifier with a TRAM A cryogenic TRAM of an Yb:YAG/YAG composite ceramic is used in the regenerative amplifier, shown in Fig. 2 . The TRAM is made by Konoshima Chemical Co., Ltd. A cross section of the YAG prism is an isosceles triangle and a base angle of 31.2°. The Yb:YAG layer is attached to the base and its size is 30 × 30 × 0.4 mm. An ytterbium ion concentration is 20 at.%. Any surfaces are polished below a surface roughness of λ /4 without any kinds of coats. A cavity consists of one flat and two concave high reflectors. The cavity length is 1.7 m. The TRAM is fixed to a stainless steel holder of a liquid nitrogen cryostat and is set at the beam waist of the cavity mode. The holder has a 20 mm diameter hole and the active Yb:YAG layer is directly cooled by the liquid nitrogen through the hole. The cooled Yb: YAG shows excellent performances such like a wide tunability of emission cross section (reducing saturation fluence U sat = hν / σ em , 1.3 J/cm 2 at liquid nitrogen temperature), high thermal strength (large thermal conductivity, weak thermo-optic effect, and small thermal expansion) and an efficient laser operation without re-absorption. A pump source is a 600-μm-core fiber-coupled laser diode and the central emission wavelength is 940 nm with a 1.3 nm spectral width. The diode emission is focused on the active layer of the TRAM. The thickness of 0.4 mm and the high ion concentration of 20 at.% shows enough absorption of the pump power more than 80% for one bounce, and a complicated multi-pass pumping is not needed. The maximum pump intensity is 4.7 kW/cm 2 at the maximum emission power of 140 W. The temporal emission waveform is rectangular and the typical pulse duration is 700 μs. The repetition rate is 100 Hz. A seed pulse is spectrally chirped by using a combination of a volume bragg grating and a gold-coated grating pair from the mode locked fiber oscillator. The pulse is spectrally shaped to match to the gain spectrum of a cryogenic Yb:YAG during the pulse stretching. After a single-mode fiber amplifier, the seed pulse energy is 0.8 pJ and the center wavelength is 1029.4 nm with the spectral width of 0.34 nm (FWHM). The pulse duration is stretched to 0.65 ns (FWHM). A combination of a Pockels cell (Lasermetrics 5046E PFN), a half-wave plate and a thin film polarizer is used for trapping and extracting the laser pulse. The trapped seed pulse is regeneratively amplified in the cavity until the pulse energy reaches the maximum. The round trip number is 42 at minimum. The build-up of the trapped pulse in the cavity is temporally monitored with a PIN photo diode behind the end high reflector. The average power of the extracted pulses is observed with a power meter (Ophir PM10) after an optical isolator, which consists of a Faraday rotator, a half-wave plate and a pair of thin film polarizers. The near-field profile at 1 m from the thin-film polarizer in the cavity is measured by CCD camera (DMK21BF04, ARGO Inc). 4 Results and discussion The observed pulse energy increases linearly with the pump energy and the round trip number decreases, shown in Fig. 3 . The optical slope efficiency is 15%. The maximum pulse energy of 3.5 mJ is obtained at 100 Hz at the 44th round trip at 37 mJ pump energy (53 W, 700 μs). More pump power is not supplied to avoid an optical damage. The optical-to-optical efficiency is 9% and the optical gain is 4.4 × 10 9 . The spatial beam profile is Gaussian in an inset of Fig. 2 . The energy fluence on the Yb:YAG layer is as high as 1.5 J/cm 2 at the spatial Gaussian peak, which is comparable to the saturation fluence of 1.3 J/cm 2 for Yb:YAG at the liquid nitrogen temperature. The spectral width of the amplified pulse gain-narrows at 0.17 nm in Fig. 4 and the corresponding pulse duration is shortened to 0.46 ns in Fig. 5 . The chirped-pulse will be temporally compressed to 9.6 ps assuming the Fourier-transform-limited pulse. By fitting our numerical calculation to the observed build-up waveform of the pulse energy in the cavity, a small signal gain and a total cavity loss are estimated at G 0 = 2.05 at maximum and L = 16%, respectively, for one round trip. At the parasitic oscillation condition of g 0 l = 4, the pump spot size is enlarged at twice under the same small signal gain coefficient g 0 and the output pulse energy of 14 mJ will be obtainable at the pump power of 212 W. To increase the pulse energy further, a thicker active layer at a lower ion concentration will be used. The M 2 -factor of the amplified laser pulse is below 1.1 at any pump power and no significant beam distortion is observed. The temperature rise of the TRAM is estimated to be several kelvin. For higher repetition rate over 1 kHz, thermal effects of thermal lensing, thermal depolarization and re-absorption due to the quasi-three level are not ignorable. Replacing the static liquid nitrogen dewer to the flowing system, efficient hydrodynamic cooling is expected in the future. 5 Conclusion Total-reflection active-mirror (TRAM) amplifier is functionally designed by using a composite ceramic to realize both high pulse energy and high average power simultaneously. A chirped-pulse regenerative amplifier with a cryogenic TRAM has been successfully demonstrated as a feasibility study. A 3.5 mJ pulse energy is obtained at a repetition rate of 100 Hz. The corresponding energy fluence on the TRAM is as high as 1.5 J/cm 2 for the 0.46 ns short pulses. The M 2 -factor below 1.1 with a high slope efficiency of 15% shows no significant thermal problem. A good scalability of the TRAM will make the pulse energy increase and a hydrodynamic cooling with a liquid nitrogen flowing system will increase the repetition rate up to more than 1 kHz. Acknowledgement Some of these works were supported by Grants-in-Aid for Scientific Research of Ministry of Education, Culture, Sports, Science and Technology in Japan (No. 21360447). References [1] J.C. Chanteloup D. Albach A. Ludicanetti K. Ertel S. Banerjee P.D. Mason C. Hemandez-Gomez J.L. Collier J. Hein M. Wolf J. Körner B.J. Le Garrec J. Phys. Conf. Ser. 244 2010 012010 [2] A. Giesen H. Hügel A. Voss K. Wittig U. Brauch H. Opower Appl. Phys. B 58 1994 365 [3] A. Giesen J. Speiser IEEE J. Sel. Top. Quant. Electr. 13 2007 598 [4] M. Tsunekane T. Taira Appl. Phys. Lett. 90 2007 121101 [5] A.J. Kemp G.J. Valentine D. Burns Prog. Quant. Electr. 29 2005 297 [6] J. Tümmler R. Jung H.J. Stiel P.V. Nickles W. Sandner Opt. Lett. 34 2009 1378 [7] T. Metzger A. Schwarz C.Y. Teisset D. Sutter A. Killi R. Kienberger F. Krausz Opt. Lett. 34 2009 2123 [8] A. Ikesue T. Kinoshita J. Am. Ceram. Soc. 78 1995 1033 [9] H. Kühn S. Heinrich A. Kahn K. Petermann J.D.B. Bradley K. Wörhoff M. Pollnau G. Huber Opt. Lett. 34 2009 2718 [10] K. Takaichi, H. Yagi, S. Todor, M. Tokurakawa, A. Shirakawa, K. Ueda, S. Hosokawa, T. Yanagitani, J. Kawanaka, and A. Kaminskii, OSA Trends in Optics and Photonics, Advanced Solid-State Photonics (ASSP) 2005 (Paper 134). [11] T. Soules, OSA Technical Digest (CD), Frontiers in Optics (FIO) 2006 (Paper FWW2). [12] H. Fujita, K. Tsubakimoto, H. Yoshida, R. Bhushan, N. Miyanaga and M. Nakatsuka, in: The European Conference on Lasers and Electro-Optics (CLEO/Euope) 2009 (Paper CA5_2). [13] H. Furuse J. Kawanaka N. Miyanaga T. Saiki K. Imasaki M. Fujita K. Takeshita S. Ishii Y. Izawa Opt. Lett. 34 2009 3439 [14] Solid-State Laser Engineering Fourth Extensively Revised and Updated Edition, W. Koechner, Springer Series in Optical Sciences, 1996, ISBN 3-540-60237-2, (Chapter 11). [15] J. Kawanaka S. Tokita H. Nishioka M. Fujita K. Yamakawa K. Ueda Y. Izawa Laser Phys. 15 2005 1306 1312 [16] R.L. Aggarwal D.J. Ripin J.R. Ochoa T.Y. Fan J. Appl. Phys. 98 2005 103514 [17] D.C. Brown IEEE J. Sel. Top. Quant. Electr. 11 2005 587