ISSN 1054660X, Laser Physics, 2010, Vol. 20, No. 4, pp. 761–765.

SOLID STATE AND LIQUID LASERS

© Pleiades Publishing, Ltd., 2010. Original Russian Text © Astro, Ltd., 2010.

DiodePumped Simultaneously QSwitched and ModeLocked Nd:GdVO4/LBO Red Laser1 Z.Y. Lia, b, B.T. Zhangb, J.F. Yangb, J.L. Heb, *, H.T. Huangb, C.H. Zuob, J.L. Xub, X.Q. Yangb, and S. Zhaob a

b

School of Sciences, Linyi Normal University. Lin’yi, 276005 China State Key Laboratory of Crystal Materials, Institute of Crystal Materials. Shandong University, Ji’nan, 250100 China *email: [email protected] Received October 5, 2009; in final form, October 15, 2009; published online March 5, 2010

Abstract—A diodeendpumped simultaneously Qswitched and modelocked intracavity frequency dou bled Nd:GdVO4/LBO red laser with an acoustooptic Qswitch was realized. The maximum red laser output power of 250 mW was obtained at the incident pump power of 8.3 W and the repetition rate of 10 kHz. At 5 kHz, the maximum modelocking modulation depth of about 80% was achieved with the Qswitched pulse width of 440 ns. The red modelocked pulse inside the Qswitched pulse had a repetition rate of 115 MHz, its average pulse width was estimated to be about 350 ps. DOI: 10.1134/S1054660X10070170 1

INTRODUCTION

Efficient red laser sources with high repetition rate, high peak power and moderate average output power have attracted much attention for their wide applica tions in medical treatment, laser color display and so on. This kind of laser sources can be obtained by fre quencydoubling the simultaneously Qswitched and modelocked (QML) lasers operating at the funda mental wavelength of 1.3 μm. QML lasers can be obtained by the passive method with the saturable absorbers, the active method with acoustooptic (AO) or electrooptic (EO) modulator and selfKerrlens effect of the laser crystals [1–8]. In the passively QML operation, the temporal and energetic jitter is severe. What’s more, the pulse repetition rate can not be easily controlled. But the actively QML and self QML with active modulators can lead to stable pulse train output with high peak power and high modulation depth. Compared with the actively QML, the self QML with an active modulator is more compact. So far, many gain mediums have been successfully used in the QML operating at 1.06 μm, such as Nd:YLF [3], Nd:YAG [5], Nd:YVO4 [4, 6], and Nd:GdVO4 [9–11], etc. There also have been many reports about the 1.3 μm QML lasers [12–14]. Among the laser crystals men tioned above, due to the higher absorption coefficient for diode pumping and a larger emission cross section, Nd:GdVO4 has been widely used in the diodepumped solidstate lasers (DPSSLs) [15–23]. However, to out best knowledge, there has not been any reports about red laser generation by frequencydoubling of the self QML 1.3 μm Nd:GdVO4 lasers with active modula tors. 1 The article is published in the original.

In this paper, simultaneously Qswitched and modelocked intracavity frequency doubled Nd:GdVO4/LBO red laser with an AO Qswitch was experimentally demonstrated. The maximum average output power of 250 mW was obtained at the incident pump power of 8.3 W and the repetition rate of 10 kHz. When the repetition rate was 5 kHz, the max imum modelocking modulation depth of about 80% with the Qswitched pulse width of 440 ns was achieved with the Qswitched pulse energy of 42.1 μJ. The modelocked pulse inside the Qswitched pulse had a repetition rate of 115 MHz, and its average pulse width was estimated to be about 350 ps. EXPERIMENTAL SETUP The experimental arrangement was shown sche matically in Fig. 1. A fourmirror ztype cavity was designed to obtain good mode matching with the pump beam and provide tight focusing in the nonlin ear crystal. The input mirror M1 was a flat mirror with high transmission (HT) at 808 nm and high reflection (HR) at 1342 nm. Two spherical mirrors M2 and M3, with radii of the curvature of 500 and 100 mm, were used in our experiment. M2 was HR coated at 1342 nm. M3 with HR coated at 1342 nm and HT coated at 671 nm was used as the output coupler. The flat mirror M4 was HR coated at 1342 and 671 nm. All the mirrors used in the experiment were HTcoated at 1064 nm to suppress the oscillation of 1064 nm in the resonator. The lengths between M1 and M2, M2 and M3 were 36 and 90 cm, respectively. Due to the ther mal lens effect in the gain medium could lead to the variation of oscillating mode size and instability of the oscillation, the distance between the mirror M3 and M4 was designed to range from 5 to 4 cm, respectively,

761

762

LI et al. M1 Laser crystal

LD

A O

M2

Focusing optics

LBO M3

M4

Fig. 1. The experimental arrangement for the Qswitched modelocked Nd:GdVO4/LBO red lasers with an AO Qswitch.

corresponding to the variation of pump power between threshold and maximum. The pump source was a commercial fibercoupled laserdiode working at the maximum absorption wavelength (808 nm) of the Nd3+ ions. The fiber bundle was with a diameter of 400 μm and a numerical aperture of 0.22. Its radiation was coupled into the laser crystal by a focusing optical sys tem consisting of two identical achromatic lenses with focal length of ~25 mm. The beam spot radius gener ated within the crystal was estimated to be 200 μm. A 0.25at % Nd3+, 3 × 3 × 8 mm3 (8 mm corresponding to the light passing direction) Nd:GdVO4 crystal was employed as the laser medium. We used a Nd:GdVO4 crystal with a low doping concentration to avoid ther mally induced fractures. It was antireflection (AR) coated for 1342 and 808 nm on the two end faces. The laser crystal was wrapped with indium foil and held in an aluminum block cooled by water at a temperature of 18°C. An AO Qswitch (NEOS 330275031) modulator with high diffraction loss at 1.3 μm was placed close to the laser medium, and its repetition rate could be tuned continuously form 1 to 50 kHz. The ratiofrequency (RF) and the power of the modu lator were 27 MHz and 100 W. The type I (θ = 85.9° and ϕ = 0°) phasematching LBO crystal with the dimensions of 3 × 3 × l5 mm (15 mm corresponding to the light passing direction) was used for the second harmonic generation (SHG). To minimize the inter nal losses caused by the Fresnel reflection, the typeI LBO crystal was antireflection coated at 1342 and 671 nm on both end faces. It was wrapped with indium foil and held in an aluminum block cooled by water at a temperature of 25°C. The temporal shape of the pulse signal was recorded by a Tektronix TDS7104 digital oscilloscope (1 GHz bandwidth, 5 G/s sam pling rate) and a photo detector with a rising time of about 400 ps (new focus, model 1611). The average output power was measured in a laser power meter (Fieldmax II, Coherent).

RESULTS AND DISCUSSIONS Firstly, in order to distinguish the modelocking mechanism in our experiment with the nonlinear mir ror modelocking, we remove the LBO crystal from the cavity and replace the flat mirror M4 with trans mission of 3% at 1342 nm. We find that the laser can also operate in the QML regime. Therefore, it is not the nonlinear modelocking mechanism that plays a role in our experiment [24]. The traditional active model for mode locking that described in [7] is that the principle of operation of AO Qswitched modelocked laser consists in enforcing in the laser resonator the mode locking on the frequency equal to RF of the AO modulator. However, the RF of the AO Qswitch used in out experiment is 27 MHz, which is different from modelocked pulse repetition rate of 115 MHz. Therefore, the phenomenon might not be explained by the traditional model. However, it could be explained by Kerrlens modelocking effect reported in [25]. The magnitude of the nonlinear refractive index for crystal is directly proportional to the strength of the selffocusing effect that determines the capability for efficient Kerrlens mode locking (KLM). Like YVO4, GdVO4 crystal also possesses large thirdorder nonlinearity and is possible to be a promising host crystal for efficient selfstarting KLM operation. In our experiment, the AO Qswitch is responsible for the Qswitching operation, and the Kerrlens effect of the Nd:GdVO4 crystal accounts for the modelocking operation. Figure 2 showed the average output power of the QML Nd:GdVO4/LBO red laser versus the incident pump power. The threshold was found to be about 1.2 W for 5 kHz repetition rate. The average output power increased with the augment of the pump power. The maximum output power of 250 mW was obtained at the incident pump power of 8.3 W ant the repetition rate of 10 kHz. The Qswitched pulse energy of the SHG red laser could be obtained by considering the average output power and the repetition rate. The maximum pulse energy was obtained to be 42.1 μJ at the repetition rate of 5 kHz, which was shown in LASER PHYSICS

Vol. 20

No. 4

2010

DIODEPUMPED SIMULTANEOUSLY QSWITCHED Pulse energy, mJ 50

Average output power, mW 250

5 kHz

10 kHz 5 kHz

10 kHz

200

40

150

30

100

20

50 0

763

10 1

2

3

4

5 6 7 8 9 Incident pump power, W

0

1

2

3

4

5 6 7 8 9 Incident pump power, W

Fig. 2. The dependence of the average output power of the red Qswitched pulse on the incident pump power.

Fig. 3. The dependence of the Qswitched pulse energy of the red Qswitched pulse on the incident pump power.

Fig. 3. The duration of the Qswitched envelope deceased as the incident pump power increased. When the repetition rate was 5 kHz, the shortest duration of 440 ns was obtained at the incident pump power of 8.3 W with the maximum modulation depth of about 80%. The corresponding temporal shape of the actively Qswitched pulse envelope was shown in Fig. 4. Figure 5 showed the expanded pulse oscillo scope traces of modelocked pulse train. From Fig. 5, it could be seen that the modelocked pulses within the Qswitched envelope were separated by about 8.7 ns, which matched exactly with the cavity

roundtrip time, corresponding to the repetition rate of 115 MHz. Using the formula [26]

200 mV/div

50Ω BW:1.0G

t me =

2

2

2

t re + t pro + t osc ,

where tme was the measured rise time of the mode locked pulse. It was about 600 ps obtained from Fig. 5 with a close observation. tre, tpro, and tosc were the real rise time of the pulse, the probe and the oscilloscope employed. The rise time of oscilloscope was deter mined by tosc × ΔfBW = 0.35–0.40, where ΔfBW was the bandwidth of the oscilloscope. The oscilloscope used

260 mV

200 ns 5.0G/s

200 ps/pt

Fig. 4. Oscilloscope traces of a typical Qswitched and modelocked red pulse envelope at the incident pump power of 8.3 W. LASER PHYSICS

Vol. 20

No. 4

2010

(1)

764

LI et al.

200 mV/div

50Ω BW:1.0G

644 mV

10.0ns 10.0G/s

100 ps/pt

Fig. 5. Expanded pulse oscilloscope traces of modelocked pulse train.

in our experiment was with a bandwidth of 1 GHz and the rise time of the probe was 400 ps. By using Eq. (1), the real rise time of the modelocked pulse was about 280 ps. According to the definition of the rise time and considering the symmetric shape of the modelocked pulse, we could assume the width of the pulse was approximately 1.25 times more than the rise time of the pulse. Then the duration of the modelocked red pulse was estimated to be about 350 ps. CONCLUSIONS We have demonstrated a Qswitched and mode locked intracavity frequency doubled Nd:GdVO4/LBO red laser with an AO Qswitch. The maximum red laser output power of 250 mW was obtained at the incident pump power of 8.3 W and the repetition rate of 10 kHz. When the Qswitch repeti tion rate was 5 kHz, the maximum modulation depth of about 80% and the pulse energy of 42.1 μJ were obtained at the incident pump power of 8.3 W. The red modelocked pulse inside the Qswitched pulse had a repetition rate of 115 MHz, and its average pulse width was estimated to be about 350 ps. The QML phenom enon in our experiment could be explained by Kerr lens modelocking effect. The AO Qswitch was responsible for the Qswitching operation, and the Kerrlens effect of the Nd:GdVO4 crystal accounts for the modelocking operation. ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation of China (grant nos. 60878012 and 50721002), the Natural Science Foundation of

Shandong Province (Y2007A22) and Program for Taishan Scholars. REFERENCES 1. F. Krausz, C. Spielmann, T. Brabec, E. Wintner, A. J. Schmidt, and A. J. Generat, Opt. Lett. 17, 204 (1992). 2. W. Jiaxian and Z. Wenzhen, Opt. Laser Technol. 32, 277 (2000). 3. S.D. Pan, J.L. He, Y.E. Hou, Y.X. Fan, H.T. Wang, Y.G. Wang, and X.Y. Ma, IEEE J. Quantum Elec tron. 42, 1097 (2006). 4. Y.F. Chen and S. W. Tsai, IEEE J. Quantum Electron. 37, 580 (2001). 5. T. Tomie, T. Kasai, and M. Yano, Jpn. J. Appl. Phys. 22, L441 (1983). 6. J. Jabczynski, W. Zendzian, and J. Kwiatkowski, Opt. Express 14, 2184 (2006). 7. J. K. Jabczynski, W. Zendzian, and J. Kwiatkowski, Proc. SPIE 6599, 65990B (2007). 8. H. Huang, J. He, X. Fan, Y. zhong, H. Chai, and Y. Wang, Opt. Laser Technol. 40, 828 (2008). 9. M. Li, S. Zhao, K. Yang, G. Li, D. Li, and J. An, Laser Phys. Lett. 5, 722 (2008). 10. J.H. Lin, K.H. Lin, H.H. Hsu, and W.F. Hsieh, Laser Phys. Lett. 5, 276 (2008). 11. L. Li, J. Liu, M. Liu, S. Liu, F. Chen, W. Wang, and Y. Wang, Laser Phys. Lett. 6, 113 (2009). 12. B.T. Zhang, J.L. He, H.T. Huang, C.H. Zuo, K.J. Yang, X.L. Dong, J.L. Xu, and S. Zhao, Laser Phys. Lett. 6, 22 (2009). 13. H.T. Huang, B.T. Zhang, J.L. He, J.F. Yang, J.L. Xu, X.Q. Yang, and S. Zhao, Laser Phys. Lett. 6, 775 (2009). LASER PHYSICS

Vol. 20

No. 4

2010

DIODEPUMPED SIMULTANEOUSLY QSWITCHED 14. Kejian Yang, Shengzhi Zhao, Jingliang He, Baitao Zhang, Chunhua Zuo, Guiqin Li, Dechun Li, and Ming Li, Opt. Express 16, 20176 (2008). 15. F. Chen, J. Liu, W. W. Wang, X. W. Fan, and L. J. Qin, Laser Phys. Lett. 6, 441 (2009). 16. S. Zhao, G. Li, D. Li, K. Yang, Y. Li, M. Li, T. Li, K. Cheng, G. Zhang, and H. Ge, Laser Phys. Lett. 7, 29 (2010). 17. X. D. Li, X. Yu, F. Chen, R. P. Yan, J. H. Yu, and D. Y. Chen, Laser Phys. Lett. 6, 631 (2009). 18. X. Yu, F. Chen, J. Gao, X. D. Li, R. P. Yan, K. Zhang, J. H. Yu, and Z. H. Zhang, Laser Phys. Lett. 6, 34 (2009). 19. X. Yu, K. Zhang, J. Gao, F. Chen, X. D. Li, R. P. Yan, J. H. Yu, and Y. Z. Wang, Laser Phys. Lett. 5, 797 (2008).

LASER PHYSICS

Vol. 20

No. 4

2010

765

20. J. S. Ma, Y. F. Li, Y. M. Sun, H. J. Qi, R. J. Lan, and X. Y. Hou, Laser Phys. Lett. 5, 593 (2008) 21. J. Gao, X. Yu, F. Chen, X. D. Li, R. P. Yan, Z. Zhang, J. H. Yu, and Y. Z. Wang, Laser Phys. Lett. 5, 577 (2008). 22. M. Liao, R. Lan, Z. Wang, H. Zhang, J. Wang, X. Hou, and X. Xu, Laser Phys. Lett. 5, 503 (2008). 23. P. K. Datta, S. Mukhopadhyay, S. K. Das, L. Tartara, A. Agnesi, and V. Degiorgio, Opt. Express 12, 4041 (2004). 24. H. C. Liang, R. C. C. Chen, Y. J. Huang, K. W. Su, and Y. F. Chen, Opt. Express 16, 21149 (2008). 25. K. J. Yang, S. Z. Zhao, G. Q. Li, M. Li, D. C. Li, J. Wang, and J. An, IEEE J. Quantum Electron. 42, 683 (2006).