We report on cavity-dumping of a semiconductor disk laser as a method to generate energetic wavelength-tunable nanosecond pulses with repetition rates ranging from 0.1 to 4MHz. Experimentally, emission of 24ns pulses with peak output power of 41W in a single beam output (and of 30 ns with peak power of 57W in a combined beam output) with wavelength tuning from 1045 to 1080nm was obtained. Numerical modeling is also introduced to provide more insight into the most important parameters controlling this mode of operation and to define optimization avenues.
©2010 Optical Society of America
In the past decade, semiconductor disk lasers (SDLs), also known as vertical external-cavity surface-emitting lasers (VECSELs), have proven to be attractive sources for the generation of high-brightness laser radiation [1–4]. The combination of a surface-emitting semiconductor gain element and a bulk external optical cavity has enabled SDLs to produce (multi)-Watt-level single-transverse-mode operation with fundamental emission wavelength ranging from the red to 2.8µm [1–3,5,6]. Furthermore, efficient intracavity nonlinear-frequency-conversion in these lasers has permitted output from the ultraviolet [1,7,8] to the mid-infrared . So far, these sources have primarily been operated in continuous-wave [1–3,10] or quasi-continuous, high-repetition-rate mode-locked regimes , capitalizing on the nanosecond upper state lifetime characteristic of the III-V semiconductor gain section. However, recently, there has been increased interest in investigating their potential as sources of energetic nanosecond pulses. To-date, this regime of operation has been approached by gain-switching [1,11–14] with either pulsed semiconductor or solid-state laser pumps.
Here, we introduce and demonstrate an alternative method, cavity-dumping, which exploits a CW-pumped gain section and an intracavity acousto-optic deflector to generate wavelength-tunable nanosecond pulses from an SDL. This approach benefits from ready access to the high-power intracavity fields inherently associated with SDLs to facilitate high-energy pulsed emission and is, in principle, applicable to SDLs at any wavelength. Furthermore, it capitalizes on a simple implementation of the modulation to generate pulses of controlled repetition rate as high as tens of mega-Hertz in contrast to direct pump modulation. Finally, it readily offers the ability to generate electrical trigger signals for applications requiring electrical/optical synchronization.
2. Laser description
The SDL cavity arrangement used in this initial demonstration around 1060nm is similar to that proposed in early papers on cavity-dumped lasers [15–17]. A four-mirror cavity was formed by an InGaAs/GaAs SDL gain/mirror structure  placed at the focus of a 150mm radius of curvature (ROC) mirror M1, and two curved mirrors with ROC of 205mm (Fig. 1 ). The semiconductor structure includes 10 strain-compensated 7-nm-thick In0.28Ga0.72As quantum wells (QWs), distributed over 10 anti-nodes of the optical field, and a 35.5-pair Al0.2Ga0.8As/AlAs distributed Bragg reflector. An acousto-optic modulator (AOM) was placed at the waist of the laser mode (mode radius 52μm) between mirrors M2 and M3 (see Fig. 1) with the output beam extraction being carried out as described in . The modulator had plane-plane parallel surfaces with anti-reflection (AR) coatings centered at 1060nm. A 2-mm-thick birefringent filter (BRF) was placed in a long cavity arm between mirrors M1 and M2, and provided laser wavelength tuning. The gain medium was pumped by a fiber-coupled laser diode array emitting at 808nm and delivering power of up to 25W. The pump spot diameter on the gain element was ~100 μm. The laser gain element was thermally managed using a 250-µm-thick type-IIa natural diamond heatspreader. This was bonded to the epilayer surface using liquid capillarity and the ensemble was mounted in a brass holder cooled to 6°C via chilled water circulation.
The crystal quartz AOM provided 60% single-pass diffraction efficiency of the linearly polarized beam (with its polarization plane perpendicular to the acoustic propagation direction) at 1060nm. The acoustic frequency of 210MHz inside the AOM was excited by a driver with an average RF power of 20W. The desired laser pulse repetition rate was controlled by the external frequency generator which triggered the driver. The frequency generator delivered square pulses with a 31 to 60ns on-time pulse duration (rise/fall time of 18ns) within the frequency range from 100kHz up to 4MHz. The difference in applied pulse duration for the various repetition rates is explained below.
Both diffracted beams were measured to have the same output power and pulse duration. The experimental results below will first be presented for only one of the extracted beams (Output #1 (see Fig. 1)). Both outputs could be combined into a single beam by retro-reflection of Output #2 back into the resonator. It was possible because the diffracted beam after reflection from the mirror M2 is parallel to the initial laser mode between the mirrors M1 and M2 and therefore is confined within the laser resonator (for a detailed description of the propagation traces of the diffracted and initial beams see ) The results of such combined output are also presented at the end of the experimental section.
3. Experimental results
The dependence of the average output power of the diffracted beam on the incident pump power (on the diamond/semiconductor structure) is plotted in Fig. 2 (a) . The characteristic rollover in this power transfer curve is observed at ~20W of pump power. Such behavior is typical for SDLs  and is attributed to induced thermal effects in the gain material at high pump powers. In the remainder of this work, the pump power was kept constant at 20W to ensure the highest output power from the laser. The output power of the same laser in cw mode of operation reaches 3.5W at 20W of pump power with output coupling of 10%.
The frequency response of this cavity dumping scheme was studied using a 60ns-duration RF pulse within the 100-500kHz range and monotonically decreasing the pulse duration (down to 31ns) going from 500kHz up to 4MHz. As shown in Fig. 2 (b), the energy of the extracted laser pulse at the incident pump power of 20W decreased from 1 down to 0.17μJ as the AOM signal frequency increased from 100kHz up to 4MHz with a cut-off frequency (defined as the frequency for which the energy is halved) of ~1.2MHz. As expected, the pulses occurred at repetition rate corresponding to the RF driver signal. They were measured to be of duration 24 ± 3ns full width at half maximum (Fig. 2 (b), inset), independent of the pulse repetition rate, a fact which will be explained below. The corresponding peak power of the pulse therefore had the same dependence upon the RF signal frequency and reached its maximum value of ~41W at ~200kHz and below (see Fig. 2 (b), right-hand vertical scale). The similarity between the frequency responses obtained for pump powers of 11 and 20W (Fig. 2 (b)) suggests that the cavity build-up time is only weakly dependent on pump power.
Figure 3 (a) shows the evolution of the intracavity field (measured with a fast silicon detector placed behind the mirror M1) over a 500ns time window around the pulse emission time for AOM modulation frequencies of 200kHz (RF pulse duration 60ns) and 1.2MHz (RF pulse duration 44ns), respectively. It can be observed that, at high repetition rates, the intracavity field is neither restored to its maximum value nor fully depleted, explaining the drop in pulse energy. We also note that the field initial build up phase is apparently faster at high repetition rates, most probably as a consequence of the higher value of the intracavity field remaining after laser pulse extraction.
The dependence of the diffracted pulse energy at high frequencies (namely, 1.2MHz) on the RF pulse duration is shown in Fig. 3 (b). As illustrated, the improved peak power performance associated with shorter RF pulses results from a better management of the intracavity field temporal evolution. Indeed, the reduced cavity opening time promotes an incomplete depletion of the intracavity field by the end of the pulse which subsequently leads to higher field intensities being reached at the end of the period. We note that, in practice, the combination of a reduced RF pulse duration and the finite (18ns) switching times of the AOM driver also decreases the effective AOM diffraction efficiency (from 60% down to 52%) which, in turn, further enhances the above-described effects. This optimization of the output energy was the prime determinant of the selected laser driving conditions.
Wavelength tunability of the cavity-dumped emission was recorded to range from ~1045 up to ~1080 nm (see Fig. 4 (a) , circles) and, in comparison with the tunability of the laser without AOM (Fig. 4, squares), was limited only by the characteristics of the AR coatings of the AOM. This demonstrates that the all-important tuning characteristic is retained in this form of operation of an SDL, offering tunable nanosecond output around any central wavelength at which SDLs can be demonstrated to operate. The fine structure in the output pulse spectrum (Fig. 4 (b)) is typical for the SDL with the heatspreader on top of the gain material and is associated with the etalon-modes of the diamond heatspreader mentioned above.
Beam profile (see Fig. 2 (a), inset) assessment performed using the knife-edge scanning technique revealed an M2 factor of ~1.7 × 1.4 at the incident pump power of 20W and of 1.5 × 1.4 at the incident pump power of 11W (which corresponds to a pulse energy of 0.4μJ, see Fig. 2 (b)). This could be further improved by adjusting the cavity length in order to achieve a ratio between the intracavity mode and the pump spot at the semiconductor chip slightly larger than unity. This should not come at the expense of a significant energy penalty if the trend follows the behavior observed during continuous-wave operation .
The performance of the laser when the two output beams (Outputs #1 and #2) are combined is shown in Fig. 5 . The pulse energy rises to ~1.7μJ (Fig. 5 (a)) whilst the FWHM pulse duration increases to 30 ns (Fig. 5 (b), solid line). The resulting maximum peak power is therefore ~57W (Fig. 5 (a), right-hand vertical scale). Both the tunability range and M2 factor of the laser with the combined beam output remain the same as for the single beam output. The M2 factor did not degrade as the combined beams are precisely superimposed with each other resulting in a single output beam [15,16]. Contrary to single output beam operation, the combined output pulse shape exhibits oscillations that are mainly noticeable at the pulse falling edge. The oscillation frequency was measured to be ~420MHz which corresponds to twice the frequency of the acoustic wave generated in the AOM. This is consistent with the fact that a fraction of the reflected beam is fed back into the cavity and this fraction undergoes two frequency shifts when being reflected in and out of the cavity .
4. Numerical simulation and analysis
In order to understand and predict the performance of SDLs operated in the cavity-dumping regime, the dynamic behavior of the laser was simulated using semiconductor rate equations for the carrier and photon densities (denoted N and S respectively) [19,20]. In this model, we assume that the pump-induced carrier generation (in the barriers) is uniform and their capture by the QWs happens on much faster timescales than the cavity dumping events. This allows us to reduce the complexity of the model to the following two coupled equations:
The carrier dependent gain can be expressed as:
The photon decay time, τph, is taken to be:
Finally, the deflected output power is given by:
The parameters used in the calculations are listed in Table 1 .
These coupled equations were solved using a fourth-order Runge-Kutta method  using as initial parameters the values obtained for continuous-wave operation (i.e. for TAOM = 1). The calculations were performed over several periods until the normalized variation of the peak intracavity field was less than 0.1%.
Typical simulation results are shown in Fig. 6 for RF frequencies of 600kHz and 2MHz, respectively. As expected, the opening of the cavity induces a dramatic reduction of the intracavity field until total depletion and leads to the generation of the optical pulse. Due to the reduced intracavity field the gain from the quantum wells is no longer depleted, hence the carrier density in the quantum-wells rises. Once the cavity is closed, the intracavity field re-builds and the carrier density returns to the level associated with continuous-wave operation. As the RF signal frequency is increased, the intracavity field no longer has sufficient time to be fully re-built (i.e. the minimum carrier density is no longer equal to its CW value and the maximum intracavity field reduces) and consequently the extracted pulse energy reduces. The pulse shape is controlled in the first instance by the opening speed of the AOM (hence the measured rising edge of 18ns) and subsequently by the decay of the intracavity field i.e. the cavity photon lifetime. As a result, modulators with higher switching speed and/or higher extraction efficiencies will enable the generation of shorter cavity-dumped pulses with identical pulse energy (equal to the energy level stored in the cavity).
To further our understanding of cavity-dumped SDLs, we extended our numerical investigations to study the influence of the cavity length on achievable performance. As expected and illustrated in Fig. 7 (red, green, blue curves), the cavity build up time increases (or equivalently the cut-off frequency decreases) linearly with cavity length. For practical cavity lengths (Lc<4m), the pulse energy is also found to increase linearly with cavity length (see inset of Fig. 7) as a result of an approximate square root increase in peak power and pulse duration (Ppeak = 41, 55, 66W and τpulse = 20, 29, 35ns for 1.2, 2.4 and 3.6m respectively). We note that, ultimately, the maximum energy achievable will be constrained by the available pump power and/or the carrier density reaching its saturation level. Finally, comparison of the simulated frequency responses for constant RF pulse duration and monotonically reduced RF pulse duration (red and black curves in Fig. 7) illustrates that the reduction of the RF pulse duration with increasing frequency leads to more energetic pulses as observed and explained earlier.
In conclusion, we report what we believe to be the first demonstration of a cavity-dumped SDL for the generation of wavelength-tunable, micro-Joule, nanosecond pulses. This form of operation takes full advantage of the high intracavity powers and broad wavelength tunability available in SDLs and the rapid and stable recovery of the intracavity field due to the short carrier lifetime gain medium. Furthermore, it is in principle applicable at any fundamental emission wavelength at which SDLs have been demonstrated, currently 650nm – 2.8μm. Stable pulse trains with repetition rates varying from 100kHz to 4MHz and a pulse peak power of 57W and beam M2 factor of 1.7 × 1.4 were obtained at 1075nm. Wavelength tunable pulsed operation was also achieved from 1045 to 1080nm. A numerical model allowing qualitative simulation of the performance of SDLs operating in the cavity dumping regime was also successfully introduced as an analysis and predictive tool for further optimization of this mode of operation.
This research was supported by the Engineering and Physical Sciences Research Council (EPSRC) programme “Ultrafast Modular Lasers”.
References and links
1. S. Calvez, J. E. Hastie, M. Guina, O. G. Okhotnikov, and M. D. Dawson, “Semiconductor disk lasers for the generation of visible and ultraviolet radiation,” Laser Photonics Rev. 3(5), 407–434 (2009). [CrossRef]
2. N. Schulz, J.-M. Hopkins, M. Rattunde, D. Burns, and J. Wagner, “High-brightness long-wavelength semiconductor disk lasers,” Laser Photonics Rev. 2(3), 160–181 (2008). [CrossRef]
3. A. C. Tropper and S. Hoogland, “Extended cavity surface-emitting semiconductor lasers,” Prog. Quantum Electron. 30(1), 1–43 (2006). [CrossRef]
4. U. Keller and A. C. Tropper, “Passively modelocked surface-emitting semiconductor lasers,” Phys. Rep. 429(2), 67–120 (2006). [CrossRef]
5. J. E. Hastie, S. Calvez, M. D. Dawson, T. Leinonen, A. Laakso, J. Lyytikäinen, and M. Pessa, “High power CW red VECSEL with linearly polarized TEM00 output beam,” Opt. Express 13(1), 77–81 (2005). [CrossRef] [PubMed]
6. B. Rösener, M. Rattunde, R. Moser, C. Manz, K. Köhler, and J. Wagner, “GaSb-based optically pumped semiconductor disk lasers emitting at a wavelength of 2.8 μm”, in Photonics West – LASE 2010, paper 7578–32 (2010).
7. J. E. Hastie, L. G. Morton, A. J. Kemp, M. D. Dawson, A. B. Krysa, and J. S. Roberts, “Tunable ultraviolet output from an intracavity frequency-doubled red vertical-external-cavity surface-emitting laser,” Appl. Phys. Lett. 89(6), 061114 (2006). [CrossRef]
8. J. Chilla, “Recent Advances in Optically Pumped Semiconductor Lasers,” in Proc. of the Conference on Photonic Applications Systems Technologies, San Jose, Paper PTuD3 (2008).
9. D. J. M. Stothard, J.-M. Hopkins, D. Burns, and M. H. Dunn, “Stable, continuous-wave, intracavity, optical parametric oscillator pumped by a semiconductor disk laser (VECSEL),” Opt. Express 17(13), 10648–10658 (2009). [CrossRef] [PubMed]
10. G. Baili, F. Bretenaker, M. Alouini, L. Morvan, D. Dolfi, and I. Sagnes, “Experimental Investigation and Analytical Modeling of Excess Intensity Noise in Semiconductor Class-A Lasers,” J. Lightwave Technol. 26(8), 952–961 (2008). [CrossRef]
11. N. Hempler, J.-M. Hopkins, A. J. Kemp, N. Schulz, M. Rattunde, J. Wagner, M. D. Dawson, and D. Burns, “Pulsed pumping of semiconductor disk lasers,” Opt. Express 15(6), 3247–3256 (2007). [CrossRef] [PubMed]
12. J.M. Yarborough, Y.-Y. Lai, Y. Kaneda, J. Hader, J.V. Moloney, T.J. Rotter, G. Balakrishnan, C. Hains, D. Huffaker, S.W. Koch, R. Bedford, “Record pulsed power demonstration of a 2 µm GaSb-based optically pumped semiconductor laser grown lattice-mismatched on an AlAs/GaAs Bragg mirror and substrate”, Applied Physics Letters 95, 081112–081112–3 (2009).
13. H. L. Chang, S. C. Huang, Y.-F. Chen, K. W. Su, Y. F. Chen, and K. F. Huang, “Efficient high-peak-power AlGaInAs eye-safe wavelength disk laser with optical in-well pumping,” Opt. Express 17(14), 11409–11414 (2009). [CrossRef] [PubMed]
14. S. C. Huang, H. L. Chang, K. W. Su, A. Li, S. C. Liu, Y. F. Chen, and K. F. Huang, “AlGaInAs/InP eye-safe laser pumped by a Q-switched Nd:GdVO4 laser,” Appl. Phys. B 94(3), 483–487 (2009). [CrossRef]
15. D. Maydan, “Fast modulator for extraction of internal laser power,” J. Appl. Phys. 41(4), 1552–1559 (1970). [CrossRef]
16. D. Maydan, “Q-Switching and Cavity Dumping of Nd:YAlG Lasers,” J. Appl. Phys. 42(3), 1031–1034 (1971). [CrossRef]
18. A. J. Maclean, R. B. Birch, P. W. Roth, A. J. Kemp, and D. Burns, “Limits on efficiency and power scaling in semiconductor disk lasers with diamond heatspreaders,” J. Opt. Soc. Am. B 26(12), 2228–2236 (2009). [CrossRef]
19. L. A. Coldren, and S. W. Corzine, Diode Lasers and photonic integrated circuits (John Wiley and sons, Inc, 1995, ISBN 0–471–117875–3), chap. 2.
20. M. Kuznetsov, F. Hakimi, R. Sprague, and A. Mooradian, “Design and characteristics of high-power (>0.5-W CW) diode-pumped vertical-extenal-cavity surface-emitting semiconductor lasers with circular TEM00 beams,” IEEE J. Sel. Top. Quantum Electron. 5(3), 561–573 (1999). [CrossRef]
21. J. C. Butcher, Numerical methods for ordinary differential equations (John Wiley and sons, Inc, 2003, ISBN 0–471–96758–0), chap. 2.