Laser-Glass, Pump-Laser-Diodes, and Amplifier for the POLARIS Laser

16 downloads 0 Views 17MB Size Report
The POLARIS laser design, a diode-pumped Yb:glass amplifier system, will offer novel prospects for applications in high-field plasma physics, laser-pumped ...
Glass and the Photonics Revolution—First International Workshop 2002

Laser-Glass, Pump-Laser-Diodes, and Amplifier for the POLARIS Laser Thomas Töpfer, Joachim Hein, Wolfram Wintzer*, Doris Ehrt*, and Roland Sauerbrey Institut für Optik und Quantenelektronik, Friedrich-Schiller-Universität, MaxWien-Platz 1, 07743 Jena, Germany *Otto-Schott-Institut für Glaschemie, Friedrich-Schiller-Universität, Fraunhoferstraße 6, 07743 Jena, Germany

Keywords: petawatt-class amplifiers, POLARIS, laser-design, fluoride, phosphate-laser glasses, laser-diode pumping, Yb:glass Increased performance of high-power laser-diodes, decline in laser-diodepower cost, development of new losing materials, progress on key optical components. Novel pulse compression techniques enable the design of efficient petawatt class amplifiers based on diode-pumped chirped pulse amplification technology. The POLARIS laser design, a diode-pumped Yb:glass amplifier system, will offer novel prospects for applications in high-field plasma physics, laser-pumped x-ray sources, and particle accelerators. Introduction The advancement of available peak power of single laser diode bars meeting economic lifetime requirements and the availability of laser diode pump power at an affordable price have sparked numerous ideas of building ever larger laser amplifiers based on CPA technology towards the petawatt level.[1,2,3] This has triggered new research in high-field plasma physics,[4,5] laser-pumped x-ray sources,[6] laser-fusion, [7] particle accelerators[8] and generators[9] as well as astro-physics.[10] Structure and properties of ytterbium-doped fluoride phosphate glasses Applying chirped pulse amplification to novel ytterbium-doped fluoride phosphate laser glasses allows design of efficient ultrahigh power amplifiers. Laser diode pumping technology requires perfect matching of pumpsource to the choice of material. Fluoride phosphate glasses exhibit a wide transmission range; low linear and nonlinear refractive index; low hydroxyl-content; tailorable spectroscopic properties by varying the phosphate content; tolerable material strength and fracture toughness; and reasonable high damage threshold as desirable for laser host materials. The material can be melted on large scale with high performance optics quality. Possible doping levels with rare earth ions exceeding IO 2 Vcm 3 , Glass Sei. Technol. 75 C1 (2002)

223

Glass and the Photonics Revolution—First International Workshop 2002 broad absorption bands, long upper state fluorescence lifetimes, and large pump saturation fluence make these glasses attractive for laser diode pumping. The small quantum defect partly balances the repetition rate limits due to poor heat conductivity. Different glass compositions were melted in large batches (1-5 kg), containing between 95 and 60 mol% fluorides (AlF 3 , MF 2 with M=Mg, Ca, Sr, Ba) and corresponding 5 to 40 mol% M(PO 3 ) 2 each. Fluorides without water and metaphosphates of optical quality were used as raw materials. Yb 3+ -ions were introduced as YbF 3 (>99,9%), 5 to 20 χ IO 2 0 ions/cm3. In some cases, oxides, Ga 2 O 3 , Nb 2 O 5 and Ta 2 O 5 , in concentrations of 1 to 5 mol%, were added, or phosphates were substituted by sulfates, or fluorides by chlorides, to 10 mol%, enhancing the asymmetry of Y b 3 + sites leading to better laser performance. An enormous advantage of these alkali-free FP melts in comparison with phosphate glass systems[ll,12] are their extreme low solubility of platinum. Neither Pt-ions nor Pt-inclusions could be detected in the glass samples. Their extremely low OH content, normally, 20 mol%, diphosphates can crystallize. About 10 mol% SrF 2 were replaced by SrCl 2 in FP 10 and FP 20 glass samples. A partial substitution of F sites by Cl should be possible.[17] Some of Cl elements could be on interstitial sites, too. But chlorides have increased the evaporation and crystallization. The analytical values showed more than 50% of the introduced Cl were lost. Adding 1 to 5 mol% Ga 2 O 3 or Ta 2 O 5 has increased the crystallization tendency drastically. Precipitations of AlPO 4 by adding Ga 2 O 3 , and TaPO5 by adding Ta 2 O 5 were found in melts forming glass-ceramic samples. Deep temperature absorption and emission spectra were recorded to investigate Stark splitting of the upper and lower laser levels. Samples were placed into a Cryostat cooled to 14K. For recording absorption samples were illuminated by a bismuth lamp and transmittance recorded by a spectrograph. Fluorescence traces were acquired by exciting samples with a chopped Ti:sapphire laser set to 9 IOnm and acquiring spectra with the same monochromator. Figure 3 shows the normalized absorption and fluorescence plots as well as the fits to pseudo Voigt functions[18] for the Glass Sei. Technol. 75 C1 (2002)

Glass and the Photonics Revolution—First International Workshop 2002

Figure 4 Absorption cross section for FP20 glass (dotted line) and Emission cross sections calculation for a FP20 glass with the reciprocity method (thin line) and the Füchtbauer-Ladenburg equation (thick line).

Figure 5 Wavelength dependent gain-parameter for FP20 glass depending on the excitation level β calculated with the reciprocity-method. respective transitions of FP20 glass.[19] Spectra were recorded for FP5, FP10, FP20, and FP20N samples. The energy splitting of the respective levels varies only within the error bars of the peakfits with compositional changes. Absorption cross section was calculated from room temperature ground state absorption spectra since no excited state absorption is present. [20] The cross section increases with growing phosphate content. Addition of sulfate and niobium oxide also leads to an increase of cross section. Glass Sei. Technol. 75 C1 (2002)

227

Glass and the Photonics Revolution—First International Workshop 2002

Figure 6 Fluorescence lifetime for FP glasses with varying phosphate content (left) and constant doping level, 8 IO 2 0 Yb 3 + /cm 3 , and for glasses with several doping levels of an FP20 glass (right). Emission cross sections were calculated from recorded room temperature fluorescence traces applying the Fuechtbauer-Ladenburg formula, as well as from the room temperature absorption traces using the reciprocity method. [21] The respective results are shown in Figure 10 for a FP 20 glass sample. It should be noted that the Fuechtbauer-Ladenburg trace is strongly obscured due to residual absorption and hence differs strongly from the trace obtained by the reciprocity method. To measure thefluorescence lifetime, samples of Imm thickness were excited at 940nm by afiber coupled laser diode delivering 50ps pulses. Radiation was focused into the sample by two achromatic lenses. The emitted fluorescence light was imaged into a fiber coupled to a monochromator with a CsO PMT detector. Signal response was read into a digital scope terminated with 150k0hm resistance to ensure unobscured time resolution while keeping S/N ratio at an adequate level. Fitting detectors response to a monoexponential decay, one finds the data points (Figure 5). Fluorescence lifetime drops from 2.0 to 1.1ms with increasing phosphate content (left graph). Varying Yb 3 + dopand level between two and 12 10 2 °/cm 3 revealed no concentration quenching of lifetime, as can be seen in the right graph of Figure 5. Radiation trapping could be excluded due to the experimental setup. Glass Sei. Technol. 75 C1 (2002)

Glass and the Photonics Revolution—First International Workshop 2002 Laser performance of several glasses was investigated by exciting Imm thick samples at 940nm inside a hemispherical laser resonator (ОС 2 %, ROC 150 mm) with an OPC fiber coupled laser diode (3 Watts, ΙΟΟμτη fiber, NA 0.22). Figure 7 shows the slope efficiency over the threshold intensity for glasses with varying phosphate content. Yb-doped glasses exhibit in general very high slope efficiencies but lack of low threshold intensities due to the residual absorption at the laser wavelength. Adding Nb-oxide and sulfate to the glass composition leads to better performance. However, these glasses are also more difficult to fabricate in high quality. Therefore, FP15 or FP20 is the best glass choice for the POLARIS laser. Plates of up to 80mm diameter and 15mm thickness, free of striae, bubbles, and inclusions were inspected with an interferometer. Homogeneity of the refractive index was measured with an interference microscope to record density fluctuations and was found to be less than Δη = 2 χ IO"6.[20] Laser-diodes as pump sources Laser diodes have proven to be advantageous pump sources for solid state lasers considering its high electrical to optical efficiency, reliability, and capability to mass production that will eventually reduce significantly the unit cost. Modeling the employment in high power amplifiers requires determination of output characteristics like energy, spectral behavior, far

Figure 7 Evaluation of laser performance of several Yb-doped fluoride phosphate glasses. Glass Sei. Technol. 75 C1 (2002)

229

Glass and the Photonics Revolution—First International Workshop 2002 field, and degradation. Laser diode bars designed for long-pulse pumping meanwhile deliver more than 200W peak power at almost 50 percent electrical to optical efficiency. Progress in semiconductor technology, heat sink design, and diode packaging ensures safe pulsed operation at driving currents of up to 220A. Operation at extremely high current causes strong heating in the active layer of the semiconductor, leading to a red-shift of emitted frequency and to a drop in output power due to decrease in efficiency at higher temperatures. Both effects can be described to first order as linear dependent from temperature. To determine spectral shift due to thermal load, the diode radiation was collimated and coupled into a fiber connected to a monochromator with a PMT as detector. The diode was driven by 3ms pulses at 150A drive current. PMT response was measured time and frequency resolved as shown in Figure 8 (left). The wavelength value of the area point of gravity represented by the black line in Figure 8 (left), was calculated at different times to calculate wavelength shift in Figure 8 (left). Center wavelength deviation and spectral red shift due to heating amount to less than 15nm being uncritical for wide absorption bands being typical for Yb-doped glasses. The model is based on a one-dimensional linear heat flow into a semi-infinite solid with radiation at the surface into the medium. [22] Laser diodes drift in wavelength under thermal load with (WdT=0.35 nm/K for the samples employed herein as was verified by experiment. Hence, we can concludetemperature rise of the diodefrom wavelength measurements during the pulse. Time-resolved analysis of laser diode output spectra delivers detailed information on temperature evolution of the junction and heat removal. The average line

Figure 8 Energy vs. drive current for 2.0ms pulses of a single laser diode bar (left) and shift in emission wavelength of a 25 bar stack at 200A (right) Glass Sei. Technol. 75 C1 (2002)

Glass and the Photonics Revolution—First International Workshop 2002

Figure 9 Gain and bandwidth depending on seed-wavelength for the first regenerative amplifier. in Figure 9 can be fit to an error functional] in order to obtain evidence for quality of heat sink design and mounting technology. Amplifier simulation and results Availability of large aperture, high damage threshold pockels cells[23] and development of low loss high performance thin film polarizers and mirrors permit utilization of regenerative amplifiers favorable for high energy extraction out of low gain materials. Material saturation fluence is in excess of most materials damage thresholds and drives the requirement for long stretched pulses and high damage resistance.[3] Amplifier material and laser-diode data discussed herein determine the design of the POLARIS amplifiers. Simulation of gain, bandwidth, and wavelength-drift follows the model in Reference 24. Figures 10 and И present the calculated gain and bandwidth evolution for the first regenerative amplifier of the system. Both parameters depend strongly on the seed wavelength of the amplifier due to the inhomogeneous gain parameter. (See Figure 5) Experiments carried out with the first regenerative amplifier support the calculations above as shown in Figure 10. Center wavelength blue-shift as well as slight decrease in bandwidth could be observed. So, far damage threshold of the amplifier glass limits the extractable energy and the amplifier is not operated near saturation. Nevertheless, amplifier operation was successfully demonstrated with extremely low fluctuations. Glass Sei. Technol. 75 C1 (2002)

231

Glass and the Photonics Revolution—First International Workshop 2002

Figure 10 Experimental data for the first regenerative amplifier. Lower graphs show the seed spectrum for the respective wavelength the upper graph represent the amplified signal. Summary Preparation of fluoride phosphate glasses of high optical quality on a medium scale was successfully demonstrated. While varying the phosphate to fluoride content of the glasses, their spectroscopic properties can be tailored in a certain range. Fluoride phosphate glasses exhibit low linear and nonlinear refractive indices. Their wide absorption bands and relatively long fluorescence lifetimes are advantageous for direct laser diode pumping. The fluorescence bandwidth supports the generation and amplification of ultra short pulses. Since the glass matrix does not incorporate platinum and the OH content is relatively low, FP glasses are an attractive high power amplifier material. Commercially available laser-diode bars to be employed as pump sources in the POLARIS have proven to be advantageous pump sources, meeting the required output energy, spectral characteristic, beam propagation, and lifetime. Operation of the first regenerative was demonstrated, so far limited by the damage threshold of the glasses employed. References 1. Perry, M. D., and G. Mourou, "Terawatt to Petawatt Subpicosecond Lasers," Science 264, 917-923 (1994). 2. Perry, M. D., D. Pennington, В. C. Stuart, G. Tietbohl, J. A. Britten, C. Brown, S. Herman, B. Golick, M. Kartz, J. Miller, H. T. Powell, M. Vergino Glass Sei. Technol. 75 C1 (2002)

Glass and the Photonics Revolution—First International Workshop 2002 and V. Yanowsky, "Petawatt Laser Pulses," Opt. Lett. 24, 160-162 (1999). 3. Töpfer, Т., J. Hein, G. Quednau, Μ. Hellwing, J. Philipps, Η. Walther, W. Theobald, R. Sauerbrey, W. Wintzer, D. Ehrt, K. Stollberg, D. Wolff, D. Habs, M. D. Perry, "Scaling laser-diode pumped solid-state amplifiers to the petawatt level," CM004, CLEO (2001). 4. Karsch, S., D. Habs, Т. Schatz, U. Schramm, P. G. Thirolf, J. Meyer-terVehn, A. Pukhov, "Particle physics with petawatt class lasers," Laser and Particle Beams 17, 565-570 (1999). 5. Witte, К., С. Gahn, J. Meyer-ter-Vehn, G. Pretzier, A. Pukhov, G. Tsakiris, "Physics of ultra-intense laser-plasma interaction," Plasma Physics and Controlled Fusion 41: B221-B230 Suppl. 12B (1999). 6. Li, Y. L., H. Schillinger, С. Ziener, R. Sauerbrey, "Reinvestigation of the Duguay soft X-ray laser: a new parameter space for high power femtosecond laser pumped systems," (vol 144, pg 118, 1997), Opt. Commun. 175, 477-477 (2000). 7. Willi, O., L. Barringer, A. Bell, M. Borghesi, J. Davies, R. Gaillard, A. Iwase, A. MacKinnon, G. Malka, C. Meyer, S. Nuruzzaman, R. Taylor, C. Vickers, D. Hoarty, P. Gobby, R. Johnson, R. G. Watt, N. Blanchot, B. Canaud, H. Croso, B. Meyer, J. L. Miquel, C. Reverdin, A. Pukhov, J. Meyer-ter-Vehn "Inertial confinement fusion and fast ignitor studies," Nuclear Fusion 40, 537-545 Sp. Iss. 3 (2000). 8. Snavely, R. A., M. H. Key, S. P. Hatchett, Т. E. Cowan, M. Roth, T. W. Phillips, M. A. Stoyer, E. A. Henry, T. C. Sangster, M. S. Singh, S. C. Wilks, A. MacKinnon, A. Offenberger, D. M. Pennington, K. Yasuike, A. B. Langdon, B. F. Lasinski, J. Johnson, M. D. Perry, E. M. Campbell, "Intense high-energy proton beams from petawatt-laser irradiation of solids," Phys. Rev. Lett. 85, 2945-2948 (2000). 9. Gahn, C., G. D. Tsakiris, G. Pretzier, К. J. Witte, С. Delfin, С. G. Wahlstrom, D. Habs, "Generating positrons with femtosecond -laser pulses," Appl. Phys. Lett. 77, 2662-2664 (2000). 10. Budil, K S., D. Μ. Gold, К. G. Estabrook, В. A. Remington, J. Kane, P. M. Bell, D. M. Pennington, C. Brown, S. P. Hatchett, J. A. Koch, Μ. H. Key, M. D. Perry, "Development of a radiative-hydrodynamics testbed using the Petawatt Laser facility," Astrophysical Journal Supplement Series 127, 261-265 (2000). 11. Campbell, J. H., Τ. I. Suratwala, "Nd-doped phosphate glasses for high-energy /high peak power lasers," J. Non-Cryst. Sol 263&264, 318341 (2000). 12. Campbell, J.H., et al., "Continuous melting of phosphate laser glasses," J. Non-Cryst. Sol 263&264, 342-357 (2000). 13. Ehrt, D., R. Atzrodt, W. Vogel, "Struktureigenschaftsbeziehungen Glass Sei. Technol. 75 C1 (2002)

233

Glass and the Photonics Revolution—First International Workshop 2002 optischer Gläser mit anomaler Teildispersion," Proc. 2nd Int. Otto Schott Colloquium, Jena July 1982, Wiss. Z. Friedrich-Schiller-Universität Jena, Math.-Naturwiss. R.,32, 509-526, 2/3 (1983). 14. Ehrt, D., "Structure and properties of fluoride phosphate glasses," SPIE 1761, 213-221 (1992). 15. Mix, E., E. Heumann, G. Huber, D. Ehrt, W. Seeber, OSA Proc. 24 (1995). 16. Töpfer, Т., J. Hein, J. Philipps, D. Ehrt, R. Sauerbrey, "Tailoring the nonlinear refractive index of fluoride-phosphate glasses for laser applications," Appl. Phys. B 71, 203-206 (2000). 17. Ebeling, P., D. Ehrt, Μ. Friedrich, "Radiation-induced color centers in anion doped phosphate glasses," Phosphorus Research Bulletin 10, 484489 (1999). 18. Sanchez-Bajo, F., et al., "The use of the pseudo-Voigt function in the variance method of x-ray line-broadening analysis," J. Appl. Crystallography 30, 427-430 (1997). 19. Buchwald, Μ. I., et al., "Determination of spectral linewidth by Voigt profiles in Yb 3+ -doped fluorozirconate glasses," Phys. Review B 57, 76737678 (1998). 20. Ehrt, D., and T. Töpfer, "Preparation, structure and properties of Y b 3 + FP laser glass," Proc. SPIE 4103 (2000). 21. McCumber, D. E., "Einstein relations connecting broadband emission and absorption spectra," Physical Review 136, A954-A957 (1964). 22. Carslaw, H.S., and J. C. Jaeger, "Conduction of Heat in Solids," Oxford University Press Inc., New York (1959). 23. Rhodes, Μ. A., S. Fochs, P. Biltoft, "Plasma electrode pockels cell for the National Ignition Facility," Fusion Technology 34, 1113-1116 Part 2 (1998). 24. Frantz, L. M., J. S. Nodvik, "Theory of pulse propagation in a laser amplifier," Journ. of Appl. Phys. 34, 2346 (1963).

Glass Sei. Technol. 75 C1 (2002)