Satellite Pulse Generation in Diode-Pumped Passively ... - IEEE Xplore

IEEE JOURNAL OF QUANTUM ELECTRONICS, VOL. 42, NO. 7, JULY 2006

625

Satellite Pulse Generation in Diode-Pumped Passively Q-Switched Nd:GdVO4 Lasers S. P. Ng, D. Y. Tang, L. J. Qian, and L. J. Qin

Abstract—We have both experimentally and numerically investigated the satellite pulse generation in diode-pumped passively Q-switched Nd:GdVO4 lasers with Cr4+ :YAG saturable absorber. We show experimentally that satellite pulse generation is a general effect of the lasers under strong pumping, and its formation is always associated with the appearance of strong Q-switched pulses. Satellite pulse generation of the lasers is further numerically simulated based on a coupled rate-equation model. Our numerical simulations show that the phenomenon is caused by the finite lifetime of the lower laser energy level of the lasers. Index Terms—Coupled-rate equations, diode-pumped lasers, Nd:GdVO4 lasers, passive Q-switching, satellite pulses.

I. INTRODUCTION IODE-PUMPED passively Q-switched solid-state lasers, due to their simplicity, low cost, and ease of operation, have found widespread applications in micro-machining, medical surgery, material processing, and scientific researches and have been intensively investigated [1]–[7]. Conventionally, the neodymium-doped YAG and YVO crystals, feasible for direct laser diode pumping, are widely used as the laser gain medium to construct compact nanosecond pulse lasers. Passive Q-switching of these lasers using various saturable absorbers has been successfully demonstrated. Recently, the newly developed Nd:GdVO crystal has also attracted great attention due to its feasibility of direct laser diode pumping, large stimulated emission cross section, and high thermal conductivity [8]–[11]. As the Nd:GdVO crystal has a thermal conductivity of 11.4 W/mK -crystal direction [12], which is comparable to along the that of the Nd:YAG crystal, it has good potential to be used even for high-power diode-pumped solid-state (DPSS) lasers. Among the various passive Q-switching materials used for the near-infrared Nd -doped lasers, the Cr :YAG crystal possesses the properties of good photochemical stability, large absorption cross section, low saturation intensity, high thermal conductivity, high damage threshold, and ease of handling. Passive Q-switching of DPSS lasers with the crystal has been extensively investigated [1]–[3], [5]–[11]. Recent experimental studies on the diode-pumped passively Q-switched Nd:GdVO

D

Manuscript received December 5, 2005; revised March 7, 2006. This work was supported by the National Key Laboratory of Advanced Material and Devices, Fudan University, China. S. P. Ng and D. Y. Tang are with the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798 (e-mail: [email protected]; [email protected]). L. J. Qian is with the Department of Optical Science and Engineering, Fudan University, Shanghai 200433, China (e-mail: [email protected]). L. J. Qin is with the School of Environment and Material Engineering, Yantai University, Shandong 264005, China (e-mail: [email protected]). Digital Object Identifier 10.1109/JQE.2006.875866

laser with the Cr :YAG crystal have demonstrated Q-switched pulses with several tens kilowatts of peak power [9], [10]. However, in the high-power operation of the passively Q-switched Nd:GdVO lasers with the Cr :YAG as saturable absorber, it was frequently observed that the Q-switched pulses could exhibit multiple spikes: in addition to the normal Q-switched pulse, there is one or, sometimes, even more, so-called satellite pulses appearing within one Q-switching process. The appearance of the satellite pulses degrades the laser Q-switching performance and limits the maximum achievable Q-switched pulse energy and peak power. We note that a similar phenomenon was also observed in the early days of the actively Q-switched lasers, and this was found to be caused by the too slow switching speed of the active Q-switcher [13]. Recently, Bartschke et al. reported a phenomenon of double-pulse generation in a diode-pumped passively Q-switched and self-frequency-doubled Nd:YAB laser [14]. They attributed it to the coexistence of two longitudinal modes in the laser. As for the passively Q-switched lasers, the Q-switching process is determined by the mutual interaction of light with the gain medium and the saturable absorber, although a too slow absorber recovery time could affect the Q-switched pulse property and Q-switching dynamics; the exact nature of this influence still needs to be investigated. Furthermore, apart from the mechanism described in [14], whether or not other properties of the laser could also cause this effect is still an open question. We have conducted both experimental and numerical investigations on the diode-pumped Nd:GdVO lasers passively Q-switched with Cr :YAG saturable absorber. We experimentally observed satellite pulse generation in the lasers and investigated its features. Based on a coupled rate-equation model, we further numerically simulated the phenomenon of the lasers. We found numerically that the satellite pulse generation is caused by the finite lifetime of the lower laser energy level of the Nd:GdVO crystal. This research result not only complements our understanding to the multipulse generation in the actively Q-switched lasers, it also shows that, in the passively Q-switched lasers, the gain and saturable absorber play an equal role on determining the Q-switched pulse property. II. EXPERIMENTAL SETUP AND OBSERVATIONS Our experimental setup used to study the passive Q-switching features of the lasers is schematically shown in Fig. 1. The lasers have a simple plano-concave cavity configuration. A 15-W fiber-coupled laser diode array (OPC-B015-FCPS) with center wavelength of 808 nm at 25 C was used as the pump source. The fiber bundle of the pump laser has a diameter of 1.16 mm with a numerical aperture of 0.22. The beam from the

0018-9197/$20.00 © 2006 IEEE

626

Fig. 1. Schematic of the laser setup. L: coupling lens; LD: laser diode.


M : concave mirror; M : output coupler;

fiber end was collimated and focused by an optical re-imaging unit (OPC-ORU-03) to a spot size of approximately 200 m is a concave mirror in radius onto the Nd:GdVO crystal. with radius of curvature of 500 mm. Its flat facet enables high transmission at the pump wavelength of 808 nm, whereas the curved surface is antireflection (AR)-coated at 808 nm and highly reflective (HR) at the lasing wavelength of 1.06 m. is a flat mirror that serves as an output coupler of the cavity. In with the reflectivity (R) of either 70% or our experiments, 60% at 1.06 m was used. To study effects of gain medium, a-cut Nd:GdVO crystals with doping concentration of 0.52 at.%, 1.14 at.%, and 1.61 at.% were used in our experiments, respectively. All of the Nd:GdVO crystals have the same cross section of 3.5 3.5 but different thicknesses of 6, 4, and 3.5 mm, respectively. Both ends of the Nd:GdVO crystals are AR-coated at 1.06 m and 808 nm. In order to efficiently dissipate the heat that is generated, the Nd:GdVO crystals were wrapped with a thin layer of indium foil and mounted within a water-cooled copper holder. The water temperature was maintained at about 20 C during the experiment to avoid crystal fracture. Two Cr :YAG saturable absorbers, one with (dimension of mm ) initial transmission (dimension of mm ), and the other with oriented with their normal along the crystal axis were used. To reduce insertion loss, the Cr :YAG crystals are AR-coated at 808 and 1064 nm on both surfaces and used without extra cooling in the experiments. For the Nd:GdVO crystals with lower doping concentrations (0.52 at.% and 1.14 at.%), a cavity length of 60 mm was selected, while, for the heavily doped crystal (1.61 at.%), cavity length was reduced to 30 mm to avoid the thermal lens-effect-induced cavity instability. The cavity lengths were selected based on the achievable average output power of the passively Q-switched operation and the cavity stability consideration. The Cr :YAG saturable absorber was inserted into the laser cavity at a position as close as possibe to the output coupler where the laser beam spot size is at the minimum. The Q-switched laser output was detected with a high-speed InGaAs photon detector (New Focus 1611-AC) and measured by a 200-MHz digital storage oscilloscope (Tektronix TDS 360). Among the Nd:GdVO crystals with different doping levels, the one with the 1.14 at.% doping concentration was found to give the highest average Q-switched output power, while the one with the 1.61 at.% doping emitted pulses with the largest single pulse energy and narrowest pulse width. Details regarding the Q-switched operation of the lasers were reported previously [9]. During our experiments, it was noticed that, under strong

Fig. 2. Observed Q-switched pulse profiles. Nd:GdVO crystal doping concentration: 1.14 at.%, incident pump power 14 W. (a) Initial Cr :YAG transmission T and output coupler reflectivity R . (b) Initial Cr :YAG and output coupler reflectivity R . transmission T

= 60%

= 80%

= 60%

= 70%

pumping, within one Q-switching process the lasers could emit multiple pulses. Apart from a main Q-switched pulse, weak pulses (satellite pulses) can also appear on the tail of the main Q-switched pulse. Fig. 2 shows, for example, typical situations of Q-switched pulse profiles of the lasers. A similar experimental phenomenon was also noticed by other authors [10]. Depending on the concrete experimental conditions, the separation between the main pulse and the satellite pulses varies, and, following the Q-switched pulse strength jittering, the strength of the satellite pulses also fluctuates. When the main laser pulse amplitude decreased, the satellite pulse amplitude became slightly higher. It seems that their total energy remain constants. This feature of the satellite pulse generation differs clearly from that of the actively Q-switched laser under the slow switching cases [13]. With the same Nd:GdVO crystal and similar cavity configuration but less strong saturable absorber in the cavity, the satellite pulse generation could also be suppressed, as shown in Fig. 2(b). Although the average Q-switched laser output power is high with a weak saturable

NG et al.: SATELLITE PULSE GENERATION IN DIODE-PUMPED PASSIVELY Q-SWITCHED Nd:GdVO LASERS

627

Fig. 3. Simplified energy levels of the Nd ions and Cr ions in the hosts. N and N : population densities in the lower and upper laser levels. N and N N : population densities of Cr ions in the ground and excited levels. is the photon density and indicates the relevant ion relaxations.

0

absorber, the energy of each Q-switched pulse becomes smaller, and the pulse width also becomes broader. We have also studied the dependence of the satellite pulse generation on the cavity loss. To this end, we either used an output coupler of too large transmission for our laser or deliberately slightly misaligned the cavity by detuning the output coupler. We found that, in these cases, the satellite pulse generation could also be suppressed due to the increased cavity loss. However, at the same time, the average Q-switched output power and the Q-switched pulse repetition rate also become lower. Through our experiments, it seems that the generation of the satellite pulse is inevitable for achieving large pulse-energy Q-switched pulses or, in other words, it is related to the stored energy in the gain medium. III. NUMERICAL MODEL Passive Q-switching of solid-state lasers was theoretically extensively investigated [13], [15]–[19]. For passively Q-switched lasers, the Q-switching process is fully determined by the mutual interaction of laser light with the gain medium and the saturable absorber, and light interaction with the rare-earth ions doped in solid-state hosts can be well described by the laser rate equations [13]. As the Nd:GdVO laser is a typical fourlevel system, and the Cr :YAG has an energy-level diagram as shown in Fig. 3, whose ground state absorption involves the passive Q-switching, and excited state absorption introduces a nonsaturable loss, a set of coupled rate equations of the following form is used to describe the passive Q-switching of the Nd:GdVO laser with Cr :YAG crystal:

(1) (2) (3) (4) where is the photon density, is the cavity output mirror reflectivity, is the cavity dissipative loss, , , , and are the length of the laser gain medium, length of the saturable absorber,

the light velocity in vacuum, and the cavity round-trip transit and are ion population densities in time, respectively, is the stimulated emission the upper and lower laser levels, is the cross section, is the pump rate in unit cm s , population density in the ground level of the absorber, is the recovery rate of the absorber, is the saturable absorption cross is the Cr ion doping density, and are the section, decay rates of the laser upper and lower energy levels to the is the decay rate of the upper ground state, respectively, and level to the lower level. In deriving the rate equations, we have ignored the Boltzmann occupation factors and assumed that the ions are directly pumped to the upper laser level. The main difference between the rate equations given above and those used by other authors [16]–[18] is that we have kept the decay term of the lower laser level. As a matter of fact, the level has also a finite lifetime. It was shown previously by Fan that the finite lower laser level lifetime of the Nd:YAG crystal could degrade energy extraction efficiency of the Q-switched pulses [20]. We have numerically solved the coupled rate equations by using the fifth-order Runge–Kutta method. A practical problem for numerically determining the passive Q-switching dynamics of the lasers is that some of the material parameters are not accurately known, e.g., the various energy level decay rates. Although some of the laser parameters have been measured, there is unfortunately a large discrepancy between the reported values, e.g., the stimulated emission cross section of cm the Nd:GdVO crystals has been reported as [21] or cm [22], and the absorption cross section of Cr :YAG has also been reported with cm and the values of cm [17], [23]–[31]. As a compromise, in our numerical simulations, we have taken values of the parameters in the range of the reported ones. When quantitative comparison with the experimental results is needed, we then used the values that give the closest result to the experimental observations. Therefore, for the results reported , and are cm , here, the values used for , cm , and cm , respectively. The determination of some of the laser parameters needs to be explained. For the passively Q-switched lasers, cavity loss modulation is a result of light interaction with the saturable absorber, which is different than the actively Q-switched laser whose cavity loss is externally modulated. For the sake of a visual comparison between the instantaneous laser gain and loss

628


in a passive Q-switching process, we have defined a cavity loss factor as

(5) and compared it with the laser population inversion in our description. According to the Judd–Ofelt theory on electric-dipole transitions of rare-earth ions [32], [33], the radiative lifetime of a given upper level is determined by the total transition probability of spontaneous electric-dipole transition between the excited state and the lower lying terminal levels. Namely, , where is the transition probability from an upper level to a lower lying level . Therefore, once the corresponding fluorescence transition branch ratio is known, the laser sponcan be estimated through the upper taneous transition rate level fluorescent lifetime as

(6) Based on the experimentally measured branch ratio of and fluorescent lifetime of s [34], we calas 4731.33 s and as 3359.74 s . Thus far, the culated of Nd:GdVO laser is value of lower energy level lifetime unknown. As a rough estimation, we have taken it in the range of several tens of nanoseconds as that for the Nd:YAG lasers [35]. Later, through comparing the simulated pulse profile, pulse width, and repetition rate under certain pump strengths with those of the experimental results, we will further estimate its value as about 20 ns. As a simplification, we assumed that the pump beam is uniform. Therefore, under the end-pumping configuration, the effective pump rate is calculated by

(7) where is the input pump power, is the pump beam rais the frequency of the pump dius in the laser crystal, and light. is the absorption coefficient of the laser gain medium which was determined experimentally for different doping conof the centrations of laser crystals. The active ion density Nd:GdVO under the doping concentration of 0.52 at.%, 1.14 , , at.%, and 1.61 at.% are calculated to be cm , respectively. The Cr ion density and in the Cr :YAG crystals is also calculated based on their initial transmission and the thickness of the crystals. They are cm and cm for the and 60% Cr :YAG crystals, respectively. IV. RESULTS AND DISCUSSION We first simulated the Q-switched pulse profiles corresponding to the experimental conditions of obtaining Fig. 2(a) and (b). The numerical results are shown in Fig. 4. They are

Fig. 4. Simulated Q-switched pulse profiles corresponding to the experimental conditions of Fig. 2. (a) Parameters used corresponding to Fig. 2(a). (b) Parameters used corresponding to Fig. 2(b).

qualitatively in agreement with those of the experimental observations. In particular, Q-switched pulses with and without satellite pulse were well reproduced. We note that the main parameter difference in obtaining Fig. 4(a) and (b) is the saturable absorber strength. It is easy to see that, with a weak saturable absorber, a Q-switched pulse with weak pulse energy and broad pulse duration is obtained. The Q-switched pulse is also characterized as having a long decay tail and no satellite pulse following the main Q-switched pulse, while, with a stronger saturable absorber, the Q-switched pulse with a narrower pulse width and higher peak power can be achieved. Instead of having a long trailing tail, there is a weak satellite pulse following the main pulse. At first sight, it seems that the strong saturable absorption could be responsible for the satellite pulse generation. To verify this, we then calculated the passively Q-switched pulse profiles under exactly the same parameters as those of Fig. 4(a) but ignored the finite lifetime of the lower laser level. No satellite pulse was obtained in this case. In fact, we have simulated with a wide range of laser parameters, such as different gain medium doping concentrations, saturable absorber strengths, and cavity output losses, but in neither case was the satellite pulse generated. This result suggests clearly


Fig. 5. Evolution of various parameters during a Q-switching process. N : population inversion. Loss (depicted with short dashed line): loss factor of the cavity as defined in the text. Parameters used in calculation: Nd:GdVO doping concentration : at.%, output coupler reflectivity R , and Cr :YAG initial transmission of (a) T and (b) T .

= 1 14

= 80%

= 70% = 60%

that it is the finite lifetime of the lower level that causes the generation of the satellite pulse. To understand the influence of finite lifetime of the lower laser level on the passive Q-switching process and the satellite pulse generation, we further depicted the evolution of the various laser parameters during one Q-switching process in Fig. 5. These parameters are population densities of the active ions in , different energy levels, population inversion photon density, cavity loss factor, and Cr ions in the ground state. In calculating the figures, we have used the laser parameters as mentioned above along with the following: cavity reand pump rate cm s . flectivity Depending only on the saturable absorber strength or , the Q-switched pulse exhibits either a satellite pulse [see Fig. 5(b)] or no satellite pulse [see Fig. 5(a)]. When a weak saturable absorber is used [see Fig. 5(a)] to the point where the saturable absorber is bleached, only a small amount of active ions can be accumulated in the upper laser level. Consequently, a Q-switched pulse with low pulse energy and broad pulse width is generated. Despite the fact that the slow

629

transition of active ions from the lower laser level to the ground state results in an accumulation of the ions in the level, which reduces the effective population inversion, the influence of this population inversion reduction on the Q-switching process is apparently weak. It only causes that the Q-switched pulse to have a relatively longer falling tail. Although it takes more time, the Q-switched pulse still can reasonably draw away all of the stored energy in the upper level. After the Q-switching, effective gain of the laser is below cavity loss. Therefore, no satellite pulse could be formed. In the case that a strong saturable absorber is used, as shown and can be accumulated in in Fig. 5(b), higher values of the gain medium. The Q-switching quickly generates a significantly strong pulse. Due to the finite decay rate to the ground state, a large number of ions accumulate in the lower laser level and cause a sudden significant drop of the effective population inversion and laser gain. This process also contributes to a quick turn-off of the Q-switching process, leading to the formation of a very narrow Q-switched pulse. However, as the effective gain decrease is due to the population increase in the lower laser level, even after the Q-switched pulse, there is still quite a high population in the upper laser level. After the population in the lower laser level is relaxed to the ground state, population inversion rebuilds and eventually causes the gain in the laser to become larger than the loss again. A satellite pulse is then formed and emitted as shown. The numerical simulation clearly shows that the satellite pulse formation in the lasers is purely a result of the finite lifetime of the lower laser level. Furthermore, the temporal separation of the satellite pulse to the main pulse is closely related to the lower laser level lifetime. We note that Fan had numerically investigated the effect of finite lower level lifetime on Q-switched lasers [20] and shown that it can reduce the energy extraction efficiency of Q-switched lasers. This effect is also known as “bottleneck” effect. Based on our simulations, it is seen that not only the energy extraction efficiency but also the Q-switched pulse profile is degraded by the finite lifetime of the level. Recently, Degnan et al. have also investigated the effects of thermalization among Stark sublevels in the upper and lower multiplets as well as the multiplets’ relaxation on the Q-switched pulse properties [36]. They have shown that both the thermalization and lower multiplet relaxation affect the Q-switched pulse energy and temporal waveform. Our result is supportive of their conclusions. For the purposes of comparison, we have also shown in Fig. 6 the laser parameter evolution in the case where the lower laser level lifetime is ignored. In this case, as there is no population accumulation in the lower laser level, population in the upper level is fully dumped out by the Q-switched pulse. Therefore, a narrow high-peak-power Q-switched pulse is generated. We note that rate equations of this form were frequently used to simulate the passively Q-switched laser properties e.g., the Q-switched pulse profile, pulse energy, and repetition rate previously [16]–[18]. It is seen that, due to the assumption of ideal laser energy extraction, the Q-switched pulses predicted by the model have much stronger pulse energy and narrower pulse width. Although our simulations have well explained the physical origin of the satellite pulse generation in the passively

630


Fig. 7. Change of the mode size ratio of laser beam in the gain medium and the saturable absorber with the pump power.

Fig. 6. Evolution of various parameters in a Q-switching process where the lower level lifetime is ignored.

Q-switched Nd:GdVO lasers, it is to point out that obvious discrepancies on the pulse duration and time interval between the main pulse and the satellite pulse exist between the simulated results and the experimental observations. To explain these discrepancies, we note that, in a real laser system, other effects that exist but were not explicitly considered in the rate-equation model could also affect the detailed Q-switching properties of the lasers. One such effect is the thermal lens formation in the lasers. Under strong pumping, this effect unavoidably appears and changes the cavity mode size. The strength of thermal lens effect varies with the pump strength and, consequently, makes the laser beam size in the gain medium and the saturable absorber change with pumping level. As a passive Q-switching process is determined by the mutual interaction of laser light with the gain and saturable absorber, this effect would undoubtedly affect the Q-switched pulse profiles. In addition, our calculation on the pump rate is also an over-simplified estimation. In fact, one should also consider the quantum efficiency, the mode matching between pump beam and cavity mode, and the laser extraction efficiency [35]. To make a quantitative comparison between the simulation and experiment, effects caused by all of these factors need to be considered. To show the impact of these effects, we have also modified the rate equation (4) in our model by adding a term describing the ratio between the beam size in the gain medium and in the saturable absorber as

(8) With this term, we then considered the influence of the thermal lens effect on the Q-switched process. To this end, we experimentally measured the thermal lens effect in our laser, and, based on the measured thermal focal lengths and the cavity law, we estimated the change of with the pump power. The result is shown in Fig. 7. With the increase of pump power, the ratio of increases, indicating that the effect of saturation absorption of the absorber becomes stronger.

Fig. 8. Comparison between the simulated and experimental single Q-switched pulse profiles. Solid line: experimental data; dashed–dotted line: simulated result.

Pump rate of a passively Q-switched laser strongly affects the Q-switched pulse repetition rate. We have also determined the effective pump rate of our lasers based on the empirical formula

(9) where is a parameter determined experimentally based on the Findlay–Clay analysis [35]. For our laser with a doping con, centration of 1.14 at.%, initial saturable absorption , the measured paand output coupler reflectivity rameter is about 0.12 at the Q-switched lasing threshold. Taking these modifications into consideration, we have recalculated the Q-switched pulse profiles and pulse repetition rate. Fig. 8 shows a comparison of the calculated pulse profile with that of the experimentally measured one. A good quantitative agreement with the experimental observations were also obtained. From value has a strong our simulations, we found that the influence on the Q-switched pulse width, while the pump rate mainly affects the pulse repetition rate. Since the time interval between the satellite pulse to the main Q-switched pulse is essentially determined by the lifetime of the lower laser level,


based on the good agreement between the simulated and experimentally measured pulse profiles, we further estimated that the lower laser level lifetime of the Nd:GdVO crystal is approximately ns. V. CONCLUSION We have both experimentally and numerically investigated the satellite pulse generation in diode-pumped passively Q-switched Nd:GdVO lasers with Cr :YAG crystal. We have shown experimentally that the satellite pulse generation is a general effect of the lasers under strong pumping. Based on a coupled rate-equation model, which explicitly takes into account the effect of the finite lower laser level lifetime, we have successfully simulated the satellite pulse generation. Our numerical simulations clearly show that the phenomenon of the lasers is caused by the bottleneck effect of the lasers. Due to the finite decay rate of the lower laser energy level, in the case of a strong Q-switched pulse generation, lots of ions could temporally accumulate in the lower laser level, which reduces the laser population inversion and accelerates the termination of the Q-switching process. Consequently, even after the main Q-switching process, there are still a large number of ions remaining in the upper laser energy level, which can further build up population inversion in the laser after the relaxation of lower laser level ions to the ground state and generate another weak Q-switched pulse. Our result shows that not only could a slow Q-switch cause multiple Q-switched pulses, but a slow ion relaxation from the lower laser level to the ground state can also generate the same effect. REFERENCES [1] Y. Shimony, Z. Burshtein, A. Ben-Amar Baranga, Y. Kalisky, and M. Strauss, “Repetitive Q-switching of a CW Nd:YAG laser using Cr :YAG saturable absorber,” IEEE J. Quantum Electron., vol. 32, no. 2, pp. 305–310, Feb. 1996. [2] J. Song, C. Li, N. S. Kim, and K. Ueda, “Passively Q-switched diodepumped continuous-wave Nd:YAG-Cr : YAG laser with high peak power and high pulse energy,” Appl. Opt., vol. 39, pp. 4954–4957, 2000. [3] A. Agnesi, S. Dell’Acqua, C. Morello, G. Piccinno, G. C. Reali, and Z. Y. Sun, “Diode-pumped neodymium lasers repetitively Q-switched by Cr :YAG solid-state saturable absorbers,” IEEE J. Quantum Electron., vol. 33, no. 1, pp. 45–52, Jan. 1997. [4] Y. K. Kuo, M. F. Huang, and M. Birnbaum, “Tunable Cr : YSO Q-switched Cr:LiCAF laser,” IEEE J. Quantum Electron., vol. 31, no. 4, pp. 657–663, Apr. 1995. [5] A. Agnesi and S. Dell’acqua, “High-peak-power diode-pumped passively Q-switched Nd:YVO laser,” Appl. Phys. B, vol. 76, pp. 351–354, 2003. [6] Y. F. Chen and Y. P. Lan, “Comparison between c-cut and a-cut Nd:YVO lasers passively Q-switched with a Cr :YAG saturable absorber,” Appl. Phys. B, vol. 74, pp. 415–418, 2002. [7] J. H. Gu, F. Zhou, W. J. Xie, S. C. Tam, and Y. L. Lam, “Passive Q-switching of a diode-pumped Nd:YAG laser with a GaAs output coupler,” Opt. Commun., vol. 165, pp. 245–249, 1999. [8] C. Li, J. Song, D. Shen, N. S. Kim, J. Lu, and K. Ueda, “Diode-pumped passively Q-switched Nd:GdVO lasers operating at 1.06 m wavelength,” Appl. Phys. B, vol. 70, pp. 471–474, 2000. [9] S. P. Ng, D. Y. Tang, L. J. Qin, and X. L. Meng, “High power passively Q-switched Nd:GdVO lasers,” Opt. Commun., vol. 229, pp. 331–336, 2004. [10] J. H. Liu, B. Ozygus, S. H. Yang, J. Erhard, U. Seelig, A. Ding, H. Weber, X. L. Meng, L. Zhu, L. J. Qin, C. L. Du, X. G. Xu, and Z. S. Shao, “Efficient passive Q-switching operation of a diode-pumped Nd:GdVO laser with a Cr :YAG saturable absorber,” J. Opt. Soc. Amer. B, vol. 20, pp. 652–661, 2003. [11] J. Liu, J. M. Yang, and J. L. He, “Diode-pumped passively Q-switched c-cut Nd:GdVO laser,” Opt. Commun., vol. 219, pp. 317–321, 2003.

631

[12] L. J. Qin, X. L. Meng, H. Y. Shen, L. Zhu, B. C. Xu, L. X. Huang, H. R. Xia, and P. Zhao, “Thermal conductivity and refractive indices of Nd:GdVO crystals,” Cryst. Res. Technol., vol. 38, pp. 793–797, 2003. [13] A. E. Siegman, Lasers. Mill Valley, CA: Univ. Science, 1986, ch. 26. [14] J. Bartschke, K. J. Boller, R. Wallenstein, I. V. Klimov, V. B. Tsvetkov, and I. A. Shcherbakov, “Diode-pumped passively Q-switched self-frequency-doubling Nd:YAB laser,” J. Opt. Soc. Amer. B, vol. 14, pp. 3452–3456, 1997. [15] A. Szabo and R. A. Stein, “Theory of laser giant pulsing by a saturable absorber,” J. Apply. Phys., vol. 36, pp. 1562–1566, 1965. [16] J. J. Degnan, “Theory of the optimally coupled Q-switched laser,” IEEE J. Quantum Electron., vol. 25, no. 2, pp. 214–220, Feb. 1989. [17] X. Y. Zhang, S. Z. Zhao, Q. P. Wang, Q. D. Zhang, L. K. Sun, and S. J. Zhang, “Optimization of Cr -doped saturable absorber Q-switched lasers,” IEEE J. Quantum Electron., vol. 33, no. 12, pp. 2286–2294, Dec. 1997. [18] J. G. Liu, D. Y. Shen, S. C. Tam, and Y. L. Lam, “Modeling pulse shape of Q-switched lasers,” IEEE J. Quantum Electron., vol. 37, no. 6, pp. 888–896, Jun. 2001. [19] G. J. Spuhler, R. Paschotta, R. Fluck, B. Braun, M. Moser, G. Zhang, E. Gini, and U. Keller, “Experimentally confirmed design guidelines for passively Q-switched microchip lasers using semiconductor saturable absorbers,” J. Opt. Soc. Amer. B, vol. 16, pp. 376–388, 1999. [20] T. Y. Fan, “Effect of finite lower level lifetime on Q-switched lasers,” IEEE J. Quantum Electron., vol. 24, no. 12, pp. 2345–2349, Dec.. 1988. [21] T. Jensen, V. G. Ostroumov, J.-P. Meyn, G. Huber, A. I. Zagumennyi, and I. A. Shcherbakov, “Spectroscopic characterization and laser performance of diode-laser-pumped Nd:GdVO ,” Appl. Phys. B, vol. 58, pp. 373–379, 1994. [22] P. K. Mukhopadhyay, A. Nautiyal, P. K. Gupta, K. Ranganathan, J. George, S. K. Sharma, and T. P. S. Nathan, “Experimental determination of the thermo-optic coefficient (dn=dT ) and the effective stimulated emission cross-section ( ) of an a-axis cut 1.-at.% doped Nd:GdVO crystal at 1.06 m wavelength,” Appl. Phys. B, vol. 77, pp. 81–87, 2003. [23] Y. Shimony, Z. Burshtein, and Y. Kalisky, “ Cr :YAG as passive Q-switch and Brewster plate in a Nd:YAG laser,” IEEE J. Quantum Electron., vol. 31, no. 10, pp. 1738–1741, Oct. 1995. [24] Z. Burshtein, P. Blau, Y. Kalisky, Y. Shimony, and M. R. Kokta, “Excited-state absorption studies of Cr ions in several garnet host crystals,” IEEE J. Quantum Electron., vol. 34, no. 2, pp. 292–299, Feb. 1998. [25] H. Eilers, K. R. Hoffman, W. M. Dennis, S. M. Jacobsen, and W. M. Yen, “Saturation of 1.06 m absorption in Cr, Ca:Y Al O crystals,” Appl. Phys. Lett., vol. 61, pp. 2958–2960, 1992. [26] X. Zhang, A. Brenier, J. Wang, and H. Zhang, “Absorption cross-sections of Cr :YAG at 946 and 914 nm,” Opt. Mater., vol. 26, pp. 293–296, 2004. [27] K. Spariosu, W. Chen, R. Stultz, M. Birnbaum, and A. V. Shestakov, “Dual Q switching and laser action at 1.06 and 1.44 m in a Nd : YAG-Cr : YAG oscillator at 300 K,” Opt. Lett., vol. 18, pp. 814–816, 1993. [28] A. Sennaroglu, C. R. Pollock, and H. Nathel, “Efficient continuouswave chromium-doped YAG laser,” J. Opt. Soc. Amer. B, vol. 12, pp. 930–937, 1995. [29] G. Xiao, J. H. Lim, S. Yang, E. Van Stryland, M. Bass, and L. Weichman, “Z-scan measurement of the ground and excited state absorption cross sections of Cr in yttrium aluminum garnet,” IEEE J. Quantum Electron, vol. 35, no. 7, pp. 1086–1091, Jul. 1999. [30] S. Kuck, K. Petermann, and G. Huber, “Spectroscopic investigation of the Cr center in YAG,” OSA Pro. Advanced Solid State Lasers, vol. 10, pp. 92–94, 1991. [31] A. V. Shestakov, N. I. Borodin, V. A. Zhitnyuk, A. G. Ohrimtchyuk, and V. P. Gaponstev, “Tunbale Cr:YAG laser,” in Proc. Conf. Lasers ElectroOpt., 1991, vol. 10, Tech. Dig. Ser., Opt. Soc. Amer., pp. 592–593. [32] B. R. Judd, “Optical absorption intensities of rare-earth ions,” Phys. Rev., vol. 127, pp. 750–761, 1962. [33] G. S. Ofelt, “Intensities of crystal spectra of rare-earth ions,” J. Chem. Phys., vol. 37, pp. 511–520, 1962. [34] H. D. Jiang, H. J. Zhang, J. Y. Wang, H. R. Xia, X. B. Hu, B. Teng, and C. Q. Zhang, “Optical and laser properties of Nd:GdVO crystal,” Opt. Commun., vol. 198, pp. 447–452, 2001. [35] W. Koechner, Solid-State Laser Engineering, 5th ed. Berlin, Germany: Springer-Verlag, 1999, ch. 3. [36] J. J. Gegnan, D. B. Coyle, and R. B. Kay, “Effects of thermalization on Q-switched laser properties,” IEEE J. Quantum Electron., vol. 34, no. 5, pp. 887–899, May 1998.

632


S. P. Ng was born in Malaysia on November 26, 1978. She received the B.Sc. degree in industrial physics from Universiti Teknologi Malaysia in 2001. She is currently working toward the Ph.D. degree in electrical and electronic engineering at the Photonics Research Center, Nanyang Technological University, Singapore. Her research interests include development of high-power diode-pumped solid-state lasers, passive Q-switching with solid-state saturable absorbers, and ultrashort pulse generation.

D. Y. Tang received the B.Sc. degree in physics from Wuhan University, Wuhan, China, in 1983, the M.Sc. degree in laser physics from the Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Science, Shanghai, China, in 1986, and the Ph.D. degree in physics from Hannover University, Hannover, Germany, in 1993. From 1993 to 1994, he was a scientific employee with the Physikalisch-Technische Budesanstalt, Braunschweig, Germany. From 1994 to 1997, he was a University Postdoctoral Research Fellow and, from 1997 to 1999, he was an Australian Research Council (ARC) Postdoctoral Research Fellow with the University of Queensland, Australia. He is currently an Associate Professor with the School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. Dr. Tang is a member of the Optical Society of America and the International Society for Optical Engineering.

L. J. Qian was born in 1965. He received the Ph.D. degree in optics from Chinese Academy of Science, Beijing, China, in 1989, for his work on femtosecond dye lasers. From 1991 to 1993, he was a Postdoctoral Fellow with Politecnico de Milano, Milan, Italy, where he was engaged in the research of high-power solidstate lasers and their intracavity frequency doubling. In 1993, he joined the Shanghai Institute of Optics and Fine Mechanics, Shanghai, China, as a Senior Researcher. Since 2001, he has been a Professor with the Department of Optical Science and Engineering, Fudan University, Shanghai. He spent the academic years of 1997–1999 on leave with the Department of Applied Physics, Cornell University, Ithaca, NY. He has published over 100 papers in international indexed journals and conferences. His research interests include ultrafast science and technology, and quadratic nonlinear optics.

L. J. Qin received the B.Sc. degree in chemistry from Shandong University, Shandong, China, in 1997, and the Ph.D. degree in synthetic crystals from the National Laboratory of Crystal Materials, Shandong, in 2003. He is currently an Associate Professor with the School of Environment and Material Engineering, Yantai University, China.