A KTiOAsO4 Raman laser is realized within a diode side-pumped acousto-optically Q-switched Nd:YAG laser. Efficient nanosecond first-Stokes generations at 1091.4 nm are obtained with three 30-mm-long KTA crystals. Under an incident diode power of 60.9 W and a pulse repetition rate of 4 kHz, a first-Stokes power of 4.55 W is obtained, corresponding to a diode-to-Stokes conversion efficiency of 7.5%. The single pulse energy is up to 1.14 mJ and the peak power is 18.0 kW.
© 2009 Optical Society of America
In recent years, more and more attention has been paid to stimulated Raman scattering (SRS) for its potential applications in generating yellow [1–2], near-infrared , eye-safe [4–6] and mid-infrared waves , in realizing mode-locking lasers  and in generating coherent anti-Stokes Raman scattering (CARS) [9–10], etc. Near recently, another interesting idea about SRS was reported in , where a cascade KTiOPO4 (KTP) Raman laser was realized for potential use in the terahertz (THz) generation with the nonlinear optical difference frequency method. Compared with KTP (Stokes shift being 270 cm-1), KTiOAsO4 (KTA) has a smaller Stokes shift (234 cm-1) . And we have realized efficient first- and second-Stokes KTA Raman lasers in  and , respectively. Especially in , 1.38 W of first-Stokes (1091.4 nm) and 1.17 W of fundamental (1064.2 nm) radiation powers were obtained simultaneously. This shows it’s a possible idea to use the KTA Raman radiations as laser sources for THz generations. For practical applications, the laser sources should present high pulse energies and high peak powers .
In this paper, we report on a diode side-pumped KTA Raman laser, which generates higher pulse energy and higher peak power than those reported in . A diode side-pumped acousto-optically (AO) Q-switched Nd:YAG laser emitting at 1064.2 nm is adopted as the pumping source. Three 30-mm-long x-cut KTA crystals are used in our experiments. Intracavity first-Stokes KTA Raman lasers were realized with one, two and three crystals, respectively. And the strongest Raman shift of KTA (234 cm-1) is used. When three KTA crystals are used, a first-Stokes power of 4.55 W is obtained with an incident diode power of 60.9 W and a pulse repetition rate (PRR) of 4 kHz. The pulse energy is 1.14 mJ and peak power is 18.0 kW. The conversion efficiency from diode power to the first-Stokes power (diode-to-Stokes conversion efficiency) is 7.5%. The wavelength of the first-Stokes line is measured to be 1091.4 nm.
2. Experimental setup
The configuration of the KTiOAsO4 Raman laser is shown in Fig. 1. The coatings of the cavity mirrors were designed for the conversion at the first-Stokes line in an intracavity Raman conversion configuration. The rear mirror (RM) was a 3000 mm radius-of-curvature concave mirror and was coated for high-reflection (HR) at 1060–1100 nm (R>99.8%). The output coupler (OC) was a plane-plane mirror and was coated to be highly reflective at 1064 nm (R>99.7%) and partially reflective (PR) at 1091.4 nm (R=91.7%). The band-pass filter (BPF) could block off lasers with wavelengths from 300 to 1200 nm except a 5-nm band around 1091 nm.
The laser head (Northrop Grumman, USA) consisted of a Nd:YAG rod (0.6 at. %, ∅3 mm×63 mm), a cooling sleeve, a diffusive optical pump cavity and three diode array modules. The 46-mm-long AO Q-switch (Gooch and Housego) had anti-reflection (AR) coatings on both faces at 1064 nm (T>99.8%) and was driven at 27.12 MHz center frequency with the rf power of 50 W. Three x-cut KTA crystals with AR coatings on each face at 1060–1100 nm (T>99.8%) were used in our experiments. And the three KTA crystals were all of the size of 5×5×30 mm3. Intracavity first-Stokes KTA Raman lasers were realized with one, two and three crystals, respectively. When two or three KTA crystals were used, their z-axes were kept parallel with each other in order to take full advantage of the X(ZZ)X̄ Raman configurations. The Nd: YAG laser head and the Q-switch were water cooled with the water temperature of 20 °C. The KTA crystals were wrapped with indium foil and mounted in water-cooled copper blocks. The water temperature was also maintained at 20 °C. The Raman laser shared the same cavity with the fundamental laser and the cavity lengths were 176 mm, 208 mm and 240 mm for the cases with one, two and three KTA crystals, respectively. The laser powers were measured by an EPM 2000 power meter (Coherent Inc.). The spectral information was monitored by a wide-range optical spectrum analyzer (AQ 6315 A, Yokogawa). The pulse temporal behavior was recorded by a digital phosphor oscilloscope (TDS 5052B, Tektronix) and a fast p-i-n photodiode.
3. Results and discussions
In the result curves in this paper, the diode powers were obtained from the current-power curve supplied by the manufacturer. Under a diode power of 60.9 W (corresponding to a current of 14.0 A), the output Raman power was studied versus pulse repetition rates. The experimental results with one, two and three KTA crystals are shown in Fig. 2. Unlike that reported with diode end-pumped scheme in , where the most efficient Raman conversion had been observed at a PRR of 25 kHz, it was found in this experiment that efficient Raman conversions occurred at much lower PRRs, i.e. 4~5 kHz.
For Q-switched lasers, more population inversions at lower repetition rates are consumed by spontaneous emissions but not stimulated emissions. As a result, the lower repetition rates can only lead to higher pulse energy, and higher average power is obtained at higher PRRs. And for the Q-switched lasers with intracavity Raman conversions, SRS processes can only occur after the oscillating threshold condition being satisfied. So, there is an optimal repetition rate for the average power of an intracavity Raman laser. If the repetition rate is higher than the optimal one, Raman conversion will be weaken and even cannot occur. In this paper, the diode side-pumped scheme had much longer laser cavity than that of the diode end-pumped scheme, the mode sizes and losses were larger, and fundamental-wave pulse durations were longer. As a result, higher pulse energy and hence higher intensity were available only at lower PRRs compared to the diode end-pumped scheme in . So, the efficient Raman conversions occurred at these lower PRRs, i.e. 4–5 kHz, where the intensity and Raman gain were high enough to counteract the losses.
As shown in Fig. 2, for the case with three KTA crystals, if the PRR was fixed at 15 kHz, no Raman conversion was observed when the diode power reached up to 60.9 W. For the case with one KTA crystal, the laser cavity was 64mm shorter than that with three KTA crystals, fundamental-wave pulse durations were shorter and losses were smaller. As a result, SRS process could still occur at PRRs higher than 15 kHz with the case of one KTA crystal. When the pump power was higher than 39.4 W, Raman conversions could occurred at higher PRRs in one-KTA case than in three-KTA case, just as the trends shown in Fig. 2. And when the pump power was lower than 39.4 W, Raman conversions could occur at higher PRRs in two-KTA case. If Raman gain was high enough to counteract the losses, Raman conversions could occur. Raman gain was proportional to the crystal length and larger losses resulted from more thermal load. Thermal load of Nd:YAG rod changed with the pump power and that of KTA crystals changed with both pump power and crystal length. Estimation of the Raman conversion threshold is given in the theoretical analysis below.
Average output Raman power with one, two and three KTA crystals were studied at 4 kHz. And the results are shown in Fig. 3. For the case of one KTA crystal being used, due to the relatively low Raman gain, the oscillating threshold was as high as 33.5 W. In order to increase the intracavity Raman gain, we placed two KTA crystals in the cavity. As is shown, the oscillating threshold fell evidently to be less than 25 W. The highest output power was obtained with three KTA crystals. When the pump power was 60.9 W, we obtained the first-Stokes power of up to 4.55 W with a pulse width of 63.3 ns. The single pulse energy could be obtained to be 1.14 mJ and the peak power was 18.0 kW. When the pump power was higher than 60.9 W, the output Stokes power went saturated, so we didn’t give the results with higher pump power.
In the diode end-pumped intracavity KTA Raman laser in , the highest output average power of 1.38 W had been obtained. The output power of 4.55 W in this paper was much higher than that. Especially, the single pulse energy of 1.15 mJ was twenty times higher than the value of 0.055 mJ and the peak power of 18.0 kW was much higher than 8.5 kW reported in . The diode end-pumped laser scheme was more compact and it was more efficient under low pump power, it induced more serious thermal lens effect under high pump power, however. So it was difficult to obtain high output average power limited by the thermal induced cavity instability. Higher output average power would be hopeful from this diode side-pumped scheme with higher quality KTA crystals and more reasonable cavity design.
Numerical analyses based on space-dependent rate equations are performed for theoretical studying of the Raman conversion thresholds. The pump light distribution is considered to be uniform inside the Nd:YAG rod. The initial population inversion density (IPID) in the cross-section of the rod, n(r, 0), is given by n(r, 0) = n(0,0)Θ(wP - r) , with w(0,0) being the IPID on laser axis and Θ the Heaviside step function. wP is the average radius of the pump photon distribution. The intracavity fundamental and Stokes photons are assumed to be of Gaussian spatial distributions [17–18], i.e. ϕi(r,t) = ϕi(r,t)exp(-2r 2/w 2) where i=F or S for the fundamental and Stokes waves, respectively. We can obtain the following equation for the population inversion density [16–17],
where F(t) is a function defined for convenience and is given by F(t) = ∫0 t ϕF(0,t)dt. σ is the stimulated emission cross section, γ represents the inversion reduction factor of the laser crystal and c the light speed in vacuum.
where llc is the “effective” length of the laser crystal. “Effective” means that in the laser head not the whole laser rod was pumped by the diode arrays, and only in the length of llc the rod was pumped. tr is the round trip time, τF and τS are the intracavity lifetimes of the fundamental and first-Stokes photons, respectively. G is a factor related to the Raman gain and is given by G=2gh νSclRC with g being the Raman gain coefficient, lRC being the length of the Raman crystal, h being the Plank constant and νS the frequency of the first Stokes beam. kFS=wF/wS is the beam size ratio of fundamental laser to the first Stokes and kSP the spontaneous Raman scattering factor.
By neglecting the final inversion density, which was proved reasonable in solving the rate equations, the initial population inversion density could be obtained by 
where τu was the upper level lifetime of Nd and f stood for the PRR. R(0) was the pumping rate given by [20–21],
where Pin was the incident diode power and h νP was the diode photon energy. ηP was a factor related to the pumping efficiency including the absorption efficiency and the transfer efficiency [20–21]. As Fujikawa et al. did in , ηP was determined to be 0.95 with the diode power below 30 W.
In the calculations, wP was 1.25 mm, g was 2.0 cm/GW , τu was 250 μs for the 0.6 at. % Nd-doped YAG , round-trip losses induced by one KTA crystal were determined to be 0.015 and round-trip losses induced by surface reflections were 0.008. The thermal induced round-trip losses were determined to be 0.036. The focal length of the thermal lens in Nd:YAG rod was measured to be 1400 mm and that in KTA crystal was determined to be 2000 mm. The calculated Raman conversion thresholds for the cases with one, two and three KTA crystals were 28.8 W, 24.8 W and 25.0 W, respectively. This was in good agreement with the experimental results in Fig. 3. The oscillating threshold with three KTA crystals was higher than that with two crystals. The configuration with three KTA crystals greatly improved the Raman gain, but it induced larger absorption losses. We didn’t try experimental configurations with more than three crystals.
For the case of three KTA crystals being used, we studied the output Raman power with respect to diode power at different PRRs. The results are shown in Fig. 4. It was found the oscillating threshold increased dramatically with the PRR rising. At 4 kHz, the threshold was less than 25 W and the value was higher than 33 W at 6 kHz.
Optical spectra of the fundamental and Raman laser radiations are shown in Fig. 5. This figure corresponded to the case with a pump power of 60.9 W and a PRR of 4 kHz. A resolution of 0.2 nm of the spectrum analyzer was selected and every point was obtained on average by five measurements. The wavelength of the first-Stokes Raman laser was measured to be 1091.4 nm. And the linewidths (FWHM) of the fundamental and Raman laser radiations were determined to be 0.48 nm and 0.40 nm, respectively. In our experiments, no higher-order Stokes lines were observed.
As reported in , the first, second, third and fourth Stokes waves were obtained from the cascade KTP Raman laser with the output power of 0.05 W, 0.61 W, 0.25 W and 0.11 W, respectively. The wavelengths were 1096 nm, 1129 nm, 1166 nm and 1204 nm, respectively, which were in agreement with the Stokes shift of 270 cm-1. We used KTA crystals as Raman medium in this paper and obtained an output power as high as 4.55 W. This power was much higher than all the KTP and KTA Raman lasers reported before. In addition, KTA has a smaller Stokes shift of 234 cm-1 and the wavelengths of the fundamental and first-Stokes waves in Fig. 5 showed good agreement with this. Through difference-frequency generation (DFG), two laser beams with smaller frequency difference will lead to longer THz wavelengths. So, the Raman laser source emitting two laser radiations with smaller Stokes shift would be of benefit to the THz-wave generations. And it is possible to generate fundamental and first-Stokes radiations simultaneously through KTA Raman lasers, as shown in . This kind of laser source has potential applications as the DFG THz-wave source.
Efficient stimulated Raman scattering have been realized in KTiOAsO4 (KTA) crystals within a diode side-pumped acousto-optically (AO) Q-switched Nd:YAG laser. The Raman oscillation shared the same cavity with the fundamental wave. When three x-cut KTA crystals were used as the Raman media, a Raman power of 4.55 W was obtained under a diode power of 60.9 W and a pulse repetition rate (PRR) of 4 kHz. The diode-to-Stokes conversion efficiency was 7.7%. The single pulse energy was up to 1.14 mJ and the peak power was 18.0 kW. These results show that KTA Raman lasers have potential applications as laser sources for THz generations. Theoretical analyses have been performed to study the Raman conversion thresholds. Future study will focus on generating high-peak-power fundamental and first-Stokes radiations simultaneously.
This work was supported by the Science and Technology Development Program of Shandong Province (No. 2007GG10001026), the National Natural Science Foundation of China (No. 60677027), and the Research Fund for the Doctoral Program of Higher Education of China (No. 20060422025).
References and links
1. J. Lee, H. M. Pask, P. Dekker, and J. A. Piper, “High efficiency, multi-Watt CW yellow emission from an intracavity-doubled self-Raman laser using Nd:GdVO4,” Opt. Express 16, 21958–21963 (2008), http://www.opticsinfobase.org/abstract.cfm?URI=oe-16-26-21958. [CrossRef] [PubMed]
2. S. T. Li, X. Y. Zhang, Q. P. Wang, X. L. Zhang, Z. H. Cong, H. J. Zhang, and J. Y. Wang, “Diode-side-pumped intracavity frequency-doubled Nd:YAG/BaWO4 Raman laser generating average output power of 3.14 W at 590 nm,” Opt. Lett. 32, 2951–2953 (2007). [CrossRef] [PubMed]
4. J. H. Huang, J. P. Lin, R. B. Su, J. H. Li, H. Zheng, C. H. Xu, F. Shi, Z. Z. Lin, J. Zhuang, W. R. Zeng, and W. X. Lin, “Short pulse eye-safe laser with a stimulated Raman scattering self-conversion based on a Nd:KGW crystal,” Opt. Lett. 32, 1096–1098 (2007). [CrossRef]
5. N. Zong, Q. J. Cui, Q. L. Ma, X. F. Zhang, Y. F. Lu, C. M. Li, D. F. Cui, Z. Y. Xu, H. J. Zhang, and J. Y. Wang, “High average power 1.5 μm eye-safe Raman shifting in BaWO4 crystals,” Appl. Opt. 48, 7–10 (2009). [CrossRef]
6. J. T. Murray, W. L. Austin, and R. C. Powell, “Intracavity Raman conversion and Raman beam cleanup,” Opt. Mater. 11, 353–371 (1999). [CrossRef]
7. T. T. Basiev, M. E. Doroshenko, L. I. Ivleva, V. V. Osiko, V. V. Badikov, and D. V. Badikov, #x201C;Some new approaches for development of mid-IR laser sources,” Proc. of SPIE 6998, 69980P, (2008). [CrossRef]
8. D. J. Spence and R. P. Mildren, “Mode locking using stimulated Raman scattering,” Opt. Express 15, 8170–8175 (2007), http://www.opticsinfobase.org/abstract.cfm?URI=oe-15-13-8170. [CrossRef] [PubMed]
9. N. Vermeulen, C. Debaes, P. Muys, and H. Thienpont, “Mitigating Heat Dissipation in Raman Lasers Using Coherent Anti-Stokes Raman Scattering,” Phys. Rev. Lett. 99, 093903 (2007). [CrossRef] [PubMed]
10. R. P. Mildren, D. W. Coutts, and D. J. Spence, “All-solid-state parametric Raman anti-Stokes laser at 508 nm,” Opt. Express 17, 810–818 (2009), http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-17-2-810. [CrossRef] [PubMed]
11. Y. T. Chang, Y. P. Huang, K. W Su, and Y. F. Chen, “Diode-pumped multi-frequency Q-switched laser with intracavity cascade Raman emission,” Opt. Express 16, 8286–8291 (2008), http://www.opticsinfobase.org/abstract.cfm?URI=oe-16-11-8286. [CrossRef] [PubMed]
12. G. H. Watson, “Polarized Raman spectra of KTiOAsO4 and isomorphic nonlinear-optical crystals,” J. Raman Spectrosc. 22, 705–713 (1991). [CrossRef]
13. Z. J. Liu, Q. P. Wang, X. Y. Zhang, Z. J. Liu, J. chang, H. Wang, S. S. Zhang, S. Z. Fan, W. J. Sun, G. F. Jin, X. T. Tao, S. J. Zhang, and H. J. Zhang, “A KTiOAsO4 Raman laser,” Appl. Phys. B 94, 585–588 (2009). [CrossRef]
14. Z. J. Liu, Q. P. Wang, X. Y. Zhang, S. S. Zhang, J. chang, H. Wang, S. Z. Fan, W. J. Sun, X. T. Tao, S. J. Zhang, and H. J. Zhang, “1120 nm second-Stokes generation in KTiOAsO4,” Laser Phys. Lett. 6, 121–124 (2009). [CrossRef]
15. K. Suizu, K. Miyamoto, T. Yamashita, and H. Ito, “High-power terahertz-wave generation using DAST crystal and detection using mid-infrared powermeter,” Opt. Lett. 32, 2885–2887 (2007). [CrossRef] [PubMed]
16. S. T. Li, X. Y. Zhang, Q. P. Wang, P. Li, J. Chang, X. L. Zhang, and Z. H. Cong, “Modeling of Q-switched lasers with top-hat pump beam distribution,” Appl. Phys. B 88, 221–226 (2007). [CrossRef]
17. X. Y. Zhang, S. Z. Zhao, Q. P. Wang, B. Ozygus, and H. Weber, “Modeling of diode-pumped actively Q-switched lasers,” IEEE J. Quantum. Electron. 35, 1912–1918 (1999). [CrossRef]
18. S. H. Ding, X. Y. Zhang, Q. P. Wang, J. Chang, S. M. Wang, and Y. R. Liu, “Modeling of actively Q-switched intracavity Raman lasers,” IEEE J. Quantum Electron. 43, 722–729 (2007). [CrossRef]
19. W. Koechner, Solid-State Laser Engineering (Springer, Berlin, Heidelberg, 1996)
20. S. Fujikawa, T. Kojima, and K. Yasui, “High-power and high-efficiency operation of a CW-diode-side-pumped Nd:YAG rod laser,” IEEE J. Sel. Top. Quantum Electron. 3, 40–44 (1997). [CrossRef]
21. O. A. Louchev, Y. Urata, N. Saito, and S. Wada, “Computational model for operation of 2 μm co-doped Tm,Ho solid state lasers,” Opt. Express 15, 11903–11912 (2007), http://www.opticsinfobase.org/abstract.cfm?URI=oe-15-19-11903. [CrossRef] [PubMed]
22. D. C. Brown, “Heat, fluorescence, and stimulated-emission power densities and fractions in Nd:YAG,” IEEE J. Quantum Electron. 34, 560–572 (1998). [CrossRef]