Efficient second to ninth harmonic generation using megawatt peak power microchip laser R. Bhandari,1,* N. Tsuji,2 T. Suzuki,3 M. Nishifuji,2 and T. Taira1 1
Laser Research Center, Institute for Molecular Science, 38 Nishigonaka, Myodaiji, Okazaki 444-8585, Japan 2 Nippon Steel & Sumitomo Metal, 20-1 Shintomi, Futtsu, Chiba 293-0011, Japan 3 Nippon Steel Technoresearch Corporation, 20-1 Shintomi, Futtsu, Chiba 293-0011, Japan * [email protected]
Abstract: We report the design and use of a megawatt peak power Nd:YAG/Cr4+:YAG microchip laser for efficient second to ninth harmonic generation. We show that the sub-nanosecond pulse width region, between 100 ps and 1 ns, is ideally suited for efficient wavelength conversion. Using this feature, we report 85% second harmonic generation efficiency using lithium triborate (LBO), 60% fourth harmonic generation efficiency usingßbarium borate, and 44% IR to UV third harmonic generation efficiency using Type I and Type II LBO. Finally, we report the first demonstration of 118 nm VUV generation in xenon gas using a microchip laser. ©2013 Optical Society of America OCIS codes: (190.4360) Nonlinear optics, devices; (190.2620) Harmonic generation and mixing; (140.3610) Lasers, ultraviolet; (140.7240) UV, EUV, and X-ray lasers.
References and links 1. 2.
J. J. Zayhowski, “Microchip lasers,” Opt. Mater. 11(2-3), 255–267 (1999). N. Pavel, J. Saikawa, S. Kurimura, and T. Taira, “High average power diode end-pumped composite Nd:YAG laser passively Q-switched by Cr4+:YAG saturable absorber,” Jpn. J. Appl. Phys. 40(Part 1, No. 3A), 1253–1259 (2001). 3. M. Tsunekane, T. Inohara, A. Ando, N. Kido, K. Kanehara, and T. Taira, “High peak power, passively Qswitched microlaser for ignition of engines,” IEEE J. Quantum Electron. 46(2), 277–284 (2010). 4. R. Bhandari and T. Taira, “> 6 MW peak power at 532 nm from passively Q-switched Nd:YAG/Cr4+:YAG microchip laser,” Opt. Express 19(20), 19135–19141 (2011). 5. R. Bhandari, T. Taira, A. Miyamoto, Y. Furukawa, and T. Tago, “> 3 MW peak power at 266 nm using Nd:YAG/ Cr4+:YAG microchip laser and fluxless-BBO,” Opt. Mater. Express 2(7), 907–919 (2012). 6. J. J. Zayhowski, “Ultraviolet generation with passively Q-switched microchip lasers: errata,” Opt. Lett. 21(19), 1618 (1996). 7. N. P. Lockyer and J. C. Vickerman, “Single photon ionization mass spectrometry using laser-generated vacuum ultraviolet photons,” Laser Chem. 17(3), 139–159 (1997). 8. J. Yang, X. B. Wang, X. P. Xing, and L. S. Wang, “Photoelectron spectroscopy of anions at 118.2 nm: Observation of high electron binding energies in superhalogens MCl4- (M=Sc, Y, La),” J. Chem. Phys. 128(20), 201102 (2008). 9. T. Taira, “Domain-controlled laser ceramics toward giant micro-photonics [invited],” Opt. Mater. Express 1(5), 1040–1050 (2011). 10. G. C. Bjorklund, “Effects of focusing on third-order nonlinear processes in isotropic media,” IEEE J. Quantum Elect. 11, 287–296. 11. R. Bhandari and T. Taira, “Palm-top size megawatt peak power ultraviolet microlaser,” Opt. Eng. 52(7), 076102 (2013).
1. Introduction Passively Q-switched microchip lasers offer several attractive features, such as simple configuration, compact construction and rugged operation [1–3]. Their small cavity length of, typically, a few millimeters results in sub-nanosecond pulse width, providing a high peak power of several megawatts even at a moderate pulse energy of a few millijoules. This feature makes them very attractive for nonlinear wavelength conversion. We have used passively Q-switched Nd:YAG/Cr4+:YAG microchip lasers for efficient second harmonic generation [SHG] [4] and fourth harmonic generation [FHG] [5], with 85%
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28849
and 60% conversion efficiency, respectively. Here, we report the use of these lasers for harmonic generation till the ninth harmonic (118 nm) of Nd:YAG. So far, passively Qswitched lasers have been used for wavelength conversion till the fifth harmonic of Nd:YAG [6]. We believe that this is the first demonstration of ninth harmonic generation using a passively Q-switched microchip laser. At present, a frequency tripled, actively Q-switched, flash-lamp-pumped, Nd:YAG laser providing 20 to 50 mJ pulse energy with 5 to 20 nanosecond pulse width at 355 nm is normally used as a source to obtain vacuum ultraviolet (VUV) at 118 nm by third harmonic generation [THG] in xenon gas [7, 8]. The size of the laser makes the system bulky and difficult to use. The use of a frequency tripled passively Q-switched microchip laser is expected to be very useful for portable VUV systems, such as time-of-flight mass spectroscope for comprehensive analysis of complex organic gas mixtures in an industrial environment. 2. Advantage of sub-nanosecond pulse width We term the pulse width region from 100 ps to 1 ns as a ‘pulse gap’ region because it is not easily accessible by either actively Q-switched lasers, or mode-locked lasers [9]. This ‘pulsegap’ region, attainable by passively Q-switched lasers, has a unique advantage for wavelength conversion. Consider the wavelength conversion efficiency given by the equation
1/2 Pω A
η = tan h 2 κ L
(1)
where L is the length of the nonlinear crystal, Pω is the input laser power, A is the effective input beam area and κ is a constant that depends on the nonlinear crystal and the input wavelength. With nanosecond lasers, since the peak power is not very high, the laser beam is required to be focused in the nonlinear crystal to increase the laser intensity Pω/A. However, this effectively limits L due to the walk-off of the nonlinear crystal, and so the conversion efficiency gets limited. With femtosecond lasers, the laser beam is not required to be focused, as the peak power is already high. But, femtosecond lasers have a wide spectral width due to the Fourier limit. Consequently, L is normally limited to a few hundred micrometers to avoid the laser pulse from getting broadened due to group velocity mismatch. Therefore, the small L again limits the conversion efficiency. However, for a laser having a pulse width in the ‘pulse gap’ region (100 ps to 1 ns), the peak power is high and the spectrum width is still small. Hence, it is possible to use a very weakly focused, or a parallel, beam in a long nonlinear crystal, making both L and Pω/A large. This enables high wavelength conversion efficiency in the ‘pulse gap’ region. 3. Second and fourth harmonic generation We designed a microchip laser using a 4 mm-thick 1.1 at.% Nd:YAG crystal as the laser medium and a 30% initial transmission Cr4+:YAG crystal as the passive Q-switching element. A flat 50% output coupler was used to form an 11 mm-long laser cavity. The laser provided an output pulse train of 3 mJ, 365 ps, 100 Hz for 100 W, 300 μs width, 100 Hz, quasicontinuous-wave (QCW) pumping. This resulted in an output peak power of 8.2 MW. SHG and FHG experiments were carried out with this microchip laser, as reported by us earlier [4,5]. Here, we give the results for completeness. SHG was obtained by using a 10 mm-long Type I LiB3O5 (Lithium Triborate, LBO) crystal in the critical phase matching (CPM) regime. Under optimum conditions, described in [4], we obtained an output of 1.7 mJ, 265 ps, 100 Hz at 532 nm with conversion efficiency of 85%. For FHG, we used a fluxless-grown Type I β-BaB2O4 (β-barium borate, BBO) crystal in the CPM regime. We obtained an output of 840 μJ, 250 ps, 100 Hz at 266 nm. This resulted in
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28850
a peak power of 3.4 MW for an input peak power of 5.6 MW at 532 nm, giving a conversion efficiency of 60%. The experimental details are described in [5]. 4. Third harmonic generation The experimental set-up for THG is shown in Fig. 1. In this case, the cavity length of the microchip laser was 10 mm and the output at the fundamental wavelength was 3 mJ, 345 ps at 100 Hz, giving a peak power of 8.7 MW. The laser beam diameter was 0.8 mm approx. The M2 value of the fundamental beam was 3.4. The laser output stability was within 1% when measured over 8 hours. We used a 10 mm-long Type I LBO crystal in the CPM regime for SHG. We did not use any optics between the laser and the LBO crystal to focus the input beam, in order to make the set-up very compact. By angle-tuning the LBO crystal, we obtained an output of 1.15 mJ, 230 ps, 100 Hz at 532 nm. This gave a peak power of 5 MW and a conversion efficiency of 57%. This conversion efficiency was sufficient for our purpose, since, in the next step, we required the residual 1064 nm beam for sum-frequency generation (SFG) to achieve THG. Hence, here we did not aim for maximum SHG conversion, as was done in the previous section.
Fig. 1. Experimental set-up for third harmonic generation.
For THG, we used a 10 mm-long Type II LBO crystal in the CPM regime. Again, no optics was used between the two LBO crystals. It was necessary to slightly retune the first LBO crystal, in order to maximize the THG output. This was done by monitoring the residual 532 nm beam after the second LBO crystal, using a CCD camera. The CCD images, before and after retuning, are shown in Figs. 2(a) and 2(b), respectively. Before retuning, the 532 nm beam profile is quite scattered as shown in Fig. 2(a). We retuned the angle of the first LBO crystal to obtain a single peak beam profile as shown Fig. 2(b). By this retuning, we increase the overlap between the residual 1064 nm beam and the 532 nm beam in the second LBO crystal.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28851
Fig. 2. Optical profile of the residual 532 nm beam (a) before retuning and (b) after retuning of the first LBO crystal.
Consequently, we obtained an output of 0.65 mJ, 170 ps, 100 Hz at 355 nm. This gave a peak power of 3.8 MW and a conversion efficiency of 76% from 532 nm to 355 nm. The conversion efficiency from 1064 nm to 355 nm was 44%. The optical and temporal profiles of the 355 nm beam are shown in Fig. 3 and Fig. 4, respectively. The 355 nm laser beam diameter was 0.6 mm approx. and the M2 value was 2.5.
Fig. 3. Optical profile of 355 nm beam.
Fig. 4. Temporal profile of 355 nm pulse.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28852
5. Ninth harmonic generation Third-order nonlinear interaction in xenon gas was used to obtain tripling of the 355 nm laser beam generated as described in the previous section. A theoretical analysis of third-order nonlinear interaction of focused laser beams has been given by Bjorklund [10]. The power P3 generated at the third harmonic frequency is given by
P3 ∝ N 2 χ 2 P13 F (bΔk , b, f , L)
(2)
where N is the density of Xe, χ is the third-order susceptibility, P1 is the input power, and F is a geometrical factor which is a function of the gas cell length L, the distance between the entrance window of the gas cell and the focus spot f, the confocal parameter b = 2πω02 n / λ0 , and the phase mismatch Δk = k3 − 3k1 . Here, ω0 is the beam-waist radius, n is the refractive index of the nonlinear medium, λ0 is the vacuum wavelength, k1 and k3 are the wave vectors of the input and output beams, respectively. For efficient harmonic generation, F must be controlled so that constructive interference occurs over the interaction length. This is done by controlling the phase mismatch, Δk. In a single component gaseous medium, the phase mismatch varies with the number density N, and is optimized by varying the pressure of the gas [10]. Hence, for a given experimental setup, the generated ninth harmonic output depends on the gas pressure and the input third harmonic power. The experimental set-up for ninth harmonic generation is shown in Figs. 5(a) and 5(b). The 355 nm beam, obtained through THG, was focused into a xenon gas cell using a 200 mm focal length lens. At the input of the Xe gas cell, there was another lens of focal length 100 mm, whose position could be adjusted by a micrometer, to finely adjust the 355 nm spot in the gas cell. The output 118 nm was used to ionize benzene in a time-of-flight massspectroscope (TOFMS). The ionized particles were detected by a micro-channel plate (MCP) detector in the TOFMS.
Fig. 5. Experimental set-up for ninth harmonic generation: (a) schematic (b) photograph.
At first, the detected signal, which is proportional to the 118 nm generation, was measured while varying the xenon gas pressure to establish the best phase-matching conditions. The measurement results are shown in Fig. 6 for an input 355 nm pulse energy of 600 μJ. It is seen that maximum output signal is obtained for a xenon gas pressure of 45 to 50 Torr.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28853
Fig. 6. Ninth harmonic (118 nm) generation versus Xe gas pressure.
Next, the xenon gas pressure was maintained at 45 Torr, and the input 355 nm pulse energy was varied. The corresponding output signal is shown in Fig. 7. The output signal, which is proportional to the 118 nm generation, deviates from the ideal cube function of the 355 nm pulse energy, as required by Eq. (2), since the input beam is not an ideal Gaussian beam, resulting in some mode mismatch. For energy greater than 600 μJ, there is an onset of saturation. At this point, the optical intensity is estimated to be > 500 GW/cm2. We believe that the saturation is due to self-phase modulation at high intensity, which deteriorates the phase matching conditions. Figure 8 shows the mass-spectrum with benzene for different input pulse energies at 355 nm. The mass-spectroscope signal shows no fragmentation, confirming single-photon ionization.
Fig. 7. Ninth harmonic (118 nm) generation versus input 355 nm pulse energy.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28854
Fig. 8. Mass spectrum with benzene for different pulse energies at 355 nm.
6. Discussion
The ability to generate 118 nm VUV using a compact picosecond microchip laser is a significant development for both scientific and industrial fields. For instance, we found that some gases, such as dichlorophenol, which could not be detected using a flash-lamp pumped 12 mJ, 10 ns, 355 nm Nd:YAG laser as a source for xenon tripling, could be identified by using the 600μJ, 170 ps, 355 nm output obtained from the microchip laser. We intend to further study the interaction of picosecond 118 nm radiation with gases, and compare it with that using nanosecond 118 nm radiation. For industrial use, the biggest advantage of using the microchip laser is its compactness and ruggedness. The 355 nm laser system, described in this paper, can be easily realized in a palm-top size device, as we have recently done for a 266 nm system [11]. In this case, the 355 nm laser module can be directly mounted on the input port of the mass-spectroscope. This will eliminate time-consuming laser alignment procedures, besides making the complete system very compact and portable. 7. Conclusion
We have demonstrated efficient second to ninth harmonic generation with a compact Nd:YAG/Cr4+:YAG microchip laser. The operation in the sub-nanosecond pulse width region enabled us to obtain high conversion efficiencies: 85% SHG, 60% FHG, 44% 1064 nm to 355 nm THG. We have also obtained ninth harmonic generation with this compact microchip laser producing 118 nm VUV, which was successfully used for single photon ionization of benzene in a mass-spectroscope. The results presented here should be useful for many applications, such as, single photon ionization, UV laser induced breakdown spectroscopy, pulsed laser deposition and materials microprocessing. Acknowledgment
We acknowledge the support of SENTAN, JST (Japan Science and Technical Agency) for this work.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28855
Laser Research Center, Institute for Molecular Science, 38 Nishigonaka, Myodaiji, Okazaki 444-8585, Japan 2 Nippon Steel & Sumitomo Metal, 20-1 Shintomi, Futtsu, Chiba 293-0011, Japan 3 Nippon Steel Technoresearch Corporation, 20-1 Shintomi, Futtsu, Chiba 293-0011, Japan * [email protected]
Abstract: We report the design and use of a megawatt peak power Nd:YAG/Cr4+:YAG microchip laser for efficient second to ninth harmonic generation. We show that the sub-nanosecond pulse width region, between 100 ps and 1 ns, is ideally suited for efficient wavelength conversion. Using this feature, we report 85% second harmonic generation efficiency using lithium triborate (LBO), 60% fourth harmonic generation efficiency usingßbarium borate, and 44% IR to UV third harmonic generation efficiency using Type I and Type II LBO. Finally, we report the first demonstration of 118 nm VUV generation in xenon gas using a microchip laser. ©2013 Optical Society of America OCIS codes: (190.4360) Nonlinear optics, devices; (190.2620) Harmonic generation and mixing; (140.3610) Lasers, ultraviolet; (140.7240) UV, EUV, and X-ray lasers.
References and links 1. 2.
J. J. Zayhowski, “Microchip lasers,” Opt. Mater. 11(2-3), 255–267 (1999). N. Pavel, J. Saikawa, S. Kurimura, and T. Taira, “High average power diode end-pumped composite Nd:YAG laser passively Q-switched by Cr4+:YAG saturable absorber,” Jpn. J. Appl. Phys. 40(Part 1, No. 3A), 1253–1259 (2001). 3. M. Tsunekane, T. Inohara, A. Ando, N. Kido, K. Kanehara, and T. Taira, “High peak power, passively Qswitched microlaser for ignition of engines,” IEEE J. Quantum Electron. 46(2), 277–284 (2010). 4. R. Bhandari and T. Taira, “> 6 MW peak power at 532 nm from passively Q-switched Nd:YAG/Cr4+:YAG microchip laser,” Opt. Express 19(20), 19135–19141 (2011). 5. R. Bhandari, T. Taira, A. Miyamoto, Y. Furukawa, and T. Tago, “> 3 MW peak power at 266 nm using Nd:YAG/ Cr4+:YAG microchip laser and fluxless-BBO,” Opt. Mater. Express 2(7), 907–919 (2012). 6. J. J. Zayhowski, “Ultraviolet generation with passively Q-switched microchip lasers: errata,” Opt. Lett. 21(19), 1618 (1996). 7. N. P. Lockyer and J. C. Vickerman, “Single photon ionization mass spectrometry using laser-generated vacuum ultraviolet photons,” Laser Chem. 17(3), 139–159 (1997). 8. J. Yang, X. B. Wang, X. P. Xing, and L. S. Wang, “Photoelectron spectroscopy of anions at 118.2 nm: Observation of high electron binding energies in superhalogens MCl4- (M=Sc, Y, La),” J. Chem. Phys. 128(20), 201102 (2008). 9. T. Taira, “Domain-controlled laser ceramics toward giant micro-photonics [invited],” Opt. Mater. Express 1(5), 1040–1050 (2011). 10. G. C. Bjorklund, “Effects of focusing on third-order nonlinear processes in isotropic media,” IEEE J. Quantum Elect. 11, 287–296. 11. R. Bhandari and T. Taira, “Palm-top size megawatt peak power ultraviolet microlaser,” Opt. Eng. 52(7), 076102 (2013).
1. Introduction Passively Q-switched microchip lasers offer several attractive features, such as simple configuration, compact construction and rugged operation [1–3]. Their small cavity length of, typically, a few millimeters results in sub-nanosecond pulse width, providing a high peak power of several megawatts even at a moderate pulse energy of a few millijoules. This feature makes them very attractive for nonlinear wavelength conversion. We have used passively Q-switched Nd:YAG/Cr4+:YAG microchip lasers for efficient second harmonic generation [SHG] [4] and fourth harmonic generation [FHG] [5], with 85%
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28849
and 60% conversion efficiency, respectively. Here, we report the use of these lasers for harmonic generation till the ninth harmonic (118 nm) of Nd:YAG. So far, passively Qswitched lasers have been used for wavelength conversion till the fifth harmonic of Nd:YAG [6]. We believe that this is the first demonstration of ninth harmonic generation using a passively Q-switched microchip laser. At present, a frequency tripled, actively Q-switched, flash-lamp-pumped, Nd:YAG laser providing 20 to 50 mJ pulse energy with 5 to 20 nanosecond pulse width at 355 nm is normally used as a source to obtain vacuum ultraviolet (VUV) at 118 nm by third harmonic generation [THG] in xenon gas [7, 8]. The size of the laser makes the system bulky and difficult to use. The use of a frequency tripled passively Q-switched microchip laser is expected to be very useful for portable VUV systems, such as time-of-flight mass spectroscope for comprehensive analysis of complex organic gas mixtures in an industrial environment. 2. Advantage of sub-nanosecond pulse width We term the pulse width region from 100 ps to 1 ns as a ‘pulse gap’ region because it is not easily accessible by either actively Q-switched lasers, or mode-locked lasers [9]. This ‘pulsegap’ region, attainable by passively Q-switched lasers, has a unique advantage for wavelength conversion. Consider the wavelength conversion efficiency given by the equation
1/2 Pω A
η = tan h 2 κ L
(1)
where L is the length of the nonlinear crystal, Pω is the input laser power, A is the effective input beam area and κ is a constant that depends on the nonlinear crystal and the input wavelength. With nanosecond lasers, since the peak power is not very high, the laser beam is required to be focused in the nonlinear crystal to increase the laser intensity Pω/A. However, this effectively limits L due to the walk-off of the nonlinear crystal, and so the conversion efficiency gets limited. With femtosecond lasers, the laser beam is not required to be focused, as the peak power is already high. But, femtosecond lasers have a wide spectral width due to the Fourier limit. Consequently, L is normally limited to a few hundred micrometers to avoid the laser pulse from getting broadened due to group velocity mismatch. Therefore, the small L again limits the conversion efficiency. However, for a laser having a pulse width in the ‘pulse gap’ region (100 ps to 1 ns), the peak power is high and the spectrum width is still small. Hence, it is possible to use a very weakly focused, or a parallel, beam in a long nonlinear crystal, making both L and Pω/A large. This enables high wavelength conversion efficiency in the ‘pulse gap’ region. 3. Second and fourth harmonic generation We designed a microchip laser using a 4 mm-thick 1.1 at.% Nd:YAG crystal as the laser medium and a 30% initial transmission Cr4+:YAG crystal as the passive Q-switching element. A flat 50% output coupler was used to form an 11 mm-long laser cavity. The laser provided an output pulse train of 3 mJ, 365 ps, 100 Hz for 100 W, 300 μs width, 100 Hz, quasicontinuous-wave (QCW) pumping. This resulted in an output peak power of 8.2 MW. SHG and FHG experiments were carried out with this microchip laser, as reported by us earlier [4,5]. Here, we give the results for completeness. SHG was obtained by using a 10 mm-long Type I LiB3O5 (Lithium Triborate, LBO) crystal in the critical phase matching (CPM) regime. Under optimum conditions, described in [4], we obtained an output of 1.7 mJ, 265 ps, 100 Hz at 532 nm with conversion efficiency of 85%. For FHG, we used a fluxless-grown Type I β-BaB2O4 (β-barium borate, BBO) crystal in the CPM regime. We obtained an output of 840 μJ, 250 ps, 100 Hz at 266 nm. This resulted in
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28850
a peak power of 3.4 MW for an input peak power of 5.6 MW at 532 nm, giving a conversion efficiency of 60%. The experimental details are described in [5]. 4. Third harmonic generation The experimental set-up for THG is shown in Fig. 1. In this case, the cavity length of the microchip laser was 10 mm and the output at the fundamental wavelength was 3 mJ, 345 ps at 100 Hz, giving a peak power of 8.7 MW. The laser beam diameter was 0.8 mm approx. The M2 value of the fundamental beam was 3.4. The laser output stability was within 1% when measured over 8 hours. We used a 10 mm-long Type I LBO crystal in the CPM regime for SHG. We did not use any optics between the laser and the LBO crystal to focus the input beam, in order to make the set-up very compact. By angle-tuning the LBO crystal, we obtained an output of 1.15 mJ, 230 ps, 100 Hz at 532 nm. This gave a peak power of 5 MW and a conversion efficiency of 57%. This conversion efficiency was sufficient for our purpose, since, in the next step, we required the residual 1064 nm beam for sum-frequency generation (SFG) to achieve THG. Hence, here we did not aim for maximum SHG conversion, as was done in the previous section.
Fig. 1. Experimental set-up for third harmonic generation.
For THG, we used a 10 mm-long Type II LBO crystal in the CPM regime. Again, no optics was used between the two LBO crystals. It was necessary to slightly retune the first LBO crystal, in order to maximize the THG output. This was done by monitoring the residual 532 nm beam after the second LBO crystal, using a CCD camera. The CCD images, before and after retuning, are shown in Figs. 2(a) and 2(b), respectively. Before retuning, the 532 nm beam profile is quite scattered as shown in Fig. 2(a). We retuned the angle of the first LBO crystal to obtain a single peak beam profile as shown Fig. 2(b). By this retuning, we increase the overlap between the residual 1064 nm beam and the 532 nm beam in the second LBO crystal.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28851
Fig. 2. Optical profile of the residual 532 nm beam (a) before retuning and (b) after retuning of the first LBO crystal.
Consequently, we obtained an output of 0.65 mJ, 170 ps, 100 Hz at 355 nm. This gave a peak power of 3.8 MW and a conversion efficiency of 76% from 532 nm to 355 nm. The conversion efficiency from 1064 nm to 355 nm was 44%. The optical and temporal profiles of the 355 nm beam are shown in Fig. 3 and Fig. 4, respectively. The 355 nm laser beam diameter was 0.6 mm approx. and the M2 value was 2.5.
Fig. 3. Optical profile of 355 nm beam.
Fig. 4. Temporal profile of 355 nm pulse.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28852
5. Ninth harmonic generation Third-order nonlinear interaction in xenon gas was used to obtain tripling of the 355 nm laser beam generated as described in the previous section. A theoretical analysis of third-order nonlinear interaction of focused laser beams has been given by Bjorklund [10]. The power P3 generated at the third harmonic frequency is given by
P3 ∝ N 2 χ 2 P13 F (bΔk , b, f , L)
(2)
where N is the density of Xe, χ is the third-order susceptibility, P1 is the input power, and F is a geometrical factor which is a function of the gas cell length L, the distance between the entrance window of the gas cell and the focus spot f, the confocal parameter b = 2πω02 n / λ0 , and the phase mismatch Δk = k3 − 3k1 . Here, ω0 is the beam-waist radius, n is the refractive index of the nonlinear medium, λ0 is the vacuum wavelength, k1 and k3 are the wave vectors of the input and output beams, respectively. For efficient harmonic generation, F must be controlled so that constructive interference occurs over the interaction length. This is done by controlling the phase mismatch, Δk. In a single component gaseous medium, the phase mismatch varies with the number density N, and is optimized by varying the pressure of the gas [10]. Hence, for a given experimental setup, the generated ninth harmonic output depends on the gas pressure and the input third harmonic power. The experimental set-up for ninth harmonic generation is shown in Figs. 5(a) and 5(b). The 355 nm beam, obtained through THG, was focused into a xenon gas cell using a 200 mm focal length lens. At the input of the Xe gas cell, there was another lens of focal length 100 mm, whose position could be adjusted by a micrometer, to finely adjust the 355 nm spot in the gas cell. The output 118 nm was used to ionize benzene in a time-of-flight massspectroscope (TOFMS). The ionized particles were detected by a micro-channel plate (MCP) detector in the TOFMS.
Fig. 5. Experimental set-up for ninth harmonic generation: (a) schematic (b) photograph.
At first, the detected signal, which is proportional to the 118 nm generation, was measured while varying the xenon gas pressure to establish the best phase-matching conditions. The measurement results are shown in Fig. 6 for an input 355 nm pulse energy of 600 μJ. It is seen that maximum output signal is obtained for a xenon gas pressure of 45 to 50 Torr.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28853
Fig. 6. Ninth harmonic (118 nm) generation versus Xe gas pressure.
Next, the xenon gas pressure was maintained at 45 Torr, and the input 355 nm pulse energy was varied. The corresponding output signal is shown in Fig. 7. The output signal, which is proportional to the 118 nm generation, deviates from the ideal cube function of the 355 nm pulse energy, as required by Eq. (2), since the input beam is not an ideal Gaussian beam, resulting in some mode mismatch. For energy greater than 600 μJ, there is an onset of saturation. At this point, the optical intensity is estimated to be > 500 GW/cm2. We believe that the saturation is due to self-phase modulation at high intensity, which deteriorates the phase matching conditions. Figure 8 shows the mass-spectrum with benzene for different input pulse energies at 355 nm. The mass-spectroscope signal shows no fragmentation, confirming single-photon ionization.
Fig. 7. Ninth harmonic (118 nm) generation versus input 355 nm pulse energy.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28854
Fig. 8. Mass spectrum with benzene for different pulse energies at 355 nm.
6. Discussion
The ability to generate 118 nm VUV using a compact picosecond microchip laser is a significant development for both scientific and industrial fields. For instance, we found that some gases, such as dichlorophenol, which could not be detected using a flash-lamp pumped 12 mJ, 10 ns, 355 nm Nd:YAG laser as a source for xenon tripling, could be identified by using the 600μJ, 170 ps, 355 nm output obtained from the microchip laser. We intend to further study the interaction of picosecond 118 nm radiation with gases, and compare it with that using nanosecond 118 nm radiation. For industrial use, the biggest advantage of using the microchip laser is its compactness and ruggedness. The 355 nm laser system, described in this paper, can be easily realized in a palm-top size device, as we have recently done for a 266 nm system [11]. In this case, the 355 nm laser module can be directly mounted on the input port of the mass-spectroscope. This will eliminate time-consuming laser alignment procedures, besides making the complete system very compact and portable. 7. Conclusion
We have demonstrated efficient second to ninth harmonic generation with a compact Nd:YAG/Cr4+:YAG microchip laser. The operation in the sub-nanosecond pulse width region enabled us to obtain high conversion efficiencies: 85% SHG, 60% FHG, 44% 1064 nm to 355 nm THG. We have also obtained ninth harmonic generation with this compact microchip laser producing 118 nm VUV, which was successfully used for single photon ionization of benzene in a mass-spectroscope. The results presented here should be useful for many applications, such as, single photon ionization, UV laser induced breakdown spectroscopy, pulsed laser deposition and materials microprocessing. Acknowledgment
We acknowledge the support of SENTAN, JST (Japan Science and Technical Agency) for this work.
#197547 - $15.00 USD Received 17 Sep 2013; revised 17 Oct 2013; accepted 20 Oct 2013; published 15 Nov 2013 (C) 2013 OSA 18 November 2013 | Vol. 21, No. 23 | DOI:10.1364/OE.21.028849 | OPTICS EXPRESS 28855
