Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Membrane saturable absorber mirror (MESAM) in a red-emitting VECSEL for the generation of stable ultrashort pulses

Open Access Open Access

Abstract

We present a new saturable absorber device principle which has the potential for broad spectral range applications. An active region membrane is separated from the substrate and placed on a dielectric end mirror. By combining the absorbing membrane with the dielectric mirror to one device we get a membrane saturable absorber mirror (MESAM) which is similar to the well-known semiconductor saturable absorber mirror (SESAM) without the restriction of the stop-band reflectivity of the distributed Bragg reflector (DBR). Stable mode-locking with the MESAM was achieved in a red-emitting VECSEL at a pump power of 4.25 W with a pulse duration of 3.06 ps at 812 MHz repetition rate. We compare the performance and pulses of both SESAM and MESAM in a z-shaped VECSEL cavity.

© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

Optically pumped semiconductor disk lasers, also known as vertical external-cavity surface-emitting lasers (VECSELs), provide several advantageous properties like a near-diffraction-limited beam profile, wavelength tunability and the flexibility to integrate optical components inside the cavity, for example frequency-doubling crystals for higher harmonics generation or etalons for single mode operation. However, the quantum defect arising from the difference of the pump wavelength and the laser wavelength incorporates heat in the active region, which appears as the limiting factor to the maximum output power of VECSELs [1].

To overcome these limitations, a new laser concept called the membrane external-cavity surface-emitting laser (MECSEL) has been demonstrated in 2016 [2]. For the MECSEL, the active region is separated from the substrate and squeezed between two transparent heatspreaders. This provides two beneficial features in comparison to the aforementioned VECSEL: First, the heat generated in the active region can be efficiently removed through the heatspreaders, which have a thermal conductivity that is up to two orders of magnitudes higher than semiconductor materials [3]. Second, by abandoning the DBR, the growth-related issues allowing only for specific spectral bandwidths do not impose a restriction of the accessible wavelength range anymore and the laser gain regions can be further exploited, which has already been shown by filling the spectral gap in the near-infrared spectral range between 815 and 835 nm [4] and by fabricating MECSELs around 1700 nm [5,6], where the low refractive index contrast of the suitable DBR layer materials opposes a challenge on the DBR growth.

For various applications as in medicine and spectroscopy pulsed laser sources are desired. A simple yet powerful device for achieving mode locking in lasers is the semiconductor saturable absorber mirror (SESAM) which is made of a monolithic chip containing the absorber region on top of a DBR. However, the standard SESAM’s performance also exhibits losses arising from temperature rise and curvature change with increasing pump powers. This can be significantly reduced by removing excess semiconductor material, i.e. the substrate [7]. This is not only helping to remove the thermal load within the structures, but also reducing the effect of thermal lensing, which is both necessary for applications in high-power class lasers.

Absorber structures with built-in strain within the active region, as is necessary for SESAMs based on InGaAs quantum wells to reach laser wavelengths between 1.3 µm and 1.5 µm, may suffer from higher nonsaturable losses within the structure due to the formation of defects. However, by including a lattice-mismatching layer between the DBR and the active region the crystalline quality of the latter can be improved, resulting in lower absorption recovery times and thus the generation of shorter pulses [8]. Other methods to achieve this include low-temperature growth [9], intentional doping of structures [10] or post-growth treatment using proton or ion bombardment [11,12], though at the cost of more complex device fabrication, reliability and the aforementioned nonsaturable losses. Fully discarding the DBR in the semiconductor saturable absorber structure hence results in even less limitations within the epitaxial growth process as the benefits of built-in strain can be exploited even further using different layer compositions for the metamorphic growth. This allows for a larger design range of the absorber region which is less dependent on the DBR crystal lattice parameters regarding defect formation and emission wavelength and offers better tunability of the pulsed laser parameters.

Our idea is to release the active region from the substrate and thereby fabricate a semiconductor saturable absorber membrane with improved functionality compared to the SESAM. Abandoning the DBR results in a more flexible applicability of an absorber inside a laser cavity. Aside from the accessible spectral range for mode-locking of lasers, the saturable absorber membrane can be used in a larger variety of cavity geometries. Specifically interesting is the case of a free-standing saturable absorber membrane placed inside a ring resonator as spatial hole burning effects are eliminated and the beam inside the cavity does not suffer from astigmatism.

In order to compare the mode locking performance of the two devices, we simulate the SESAM working principle by combining the saturable absorber membrane with an HR dielectric mirror to form a membrane saturable absorber mirror (MESAM). The MESAM is supposed to function similarly compared to a standard SESAM with the aforementioned increased wavelength flexibility due to absent DBR and the associated growth limitations. Furthermore, the MESAM approach still allows for separated designing of mirror and absorber region.

In this paper we explain the conceptual design and the fabrication steps of the MESAM with an AlGaInP-based semiconductor heterostructure using GaInP quantum wells as the absorbing material for operation in the red spectral range. We compare the dynamic properties of both SESAM and MESAM used in a z-shaped VECSEL cavity by analyzing the temporal trace and radiofrequency spectra as well as the autocorrelation of the mode-locked laser.

2. Absorber structures, fabrication and cavity design

Our samples are fabricated by metal-organic vapor-phase epitaxy (MOVPE) in an AIXTRON 3$\times$2" close-coupled showerhead reactor at a pressure of $100\,\text {mbar}$ and a temperature of $640\,^\circ \text {C}$, using trimethylgallium, trimethylaluminum, trimethylindium, arsine and phosphine as source gases. The samples were grown on (100)-GaAs substrates with a 6$^\circ$ miscut towards the [111]$_\text {A}$-direction.

The SESAM comprises a single 5 nm thick Ga$_{0.40}$In$_{0.60}$P/(Al$_{0.33}$Ga$_{0.66}$)$_{0.49}$In$_{0.51}$P quantum well in a resonant design on top of a 55$\lambda /4$-pair Al$_{0.50}$GaAs/Al$_{0.95}$GaAs DBR, already used in [13].

For the saturable absorber membrane structure, no DBR but an AlAs sacrificial layer is grown between active region and substrate. The active region contains two GaInP quantum wells embedded between Al$_{0.33}$GaInP barriers. As we here consider a membrane in the resonator, instead of a monolithic device with an integrated mirror structure as it is the case in a SESAM, the distance between the membrane and the dielectric mirror of the MESAM can hardly be determined as there might be a remaining liquid film with unknown thickness between those two. The standing wave electric field pattern’s maxima may actually not match the quantum well position so the absorption may not be exploited. In order to have at least one actively absorbing quantum well, the distance between the two quantum wells is chosen as a quarter of the anticipated laser wavelength. Therefore, if one quantum well is in the antinode of the standing wave pattern, the other one is positioned in the node and vice versa (see Fig. 1). The membrane was released by the same wet-chemical process steps that have been used before in [2], consisting of a two-part process of GaAs wafer etching with a NH$_4$OH:H$_2$O$_2$ solution and AlAs sacrificial layer removal with HF to obtain the isolated absorber membrane. An SEM image of the membrane after these processing steps can be seen in Fig. 2(a). AFM measurements were used to determine the root mean square (RMS) roughness of the processed surface to be about 0.5 nm, which is a very good match to the epi-ready GaAs wafer.

 figure: Fig. 1.

Fig. 1. Refractive index profile and intensity of the simulated electric field of the saturable absorber membrane structure. In the case shown, the electric field is maximal at the left quantum well so that absorption can take place and minimal at the right quantum well, where no electric field intensity is absorbed.

Download Full Size | PDF

 figure: Fig. 2.

Fig. 2. (a) Tilt view SEM image of the free-standing membrane after wet-chemical processing. The membrane thickness of approximately 150 nm is indicated by the white bar. (b) Photograph of the MESAM in a mirror mount.

Download Full Size | PDF

In the last step, the less than 150 nm thin membrane is combined with a flat dielectric mirror with 99.98% reflectivity by using the method of liquid capillary bonding [14] to form a MESAM. An image of the MESAM with numerous membrane pieces is shown in Fig. 2(b).

The gain structure consists of five packages of four GaInP quantum wells as a resonant-periodic gain structure with a nearly-antiresonant coupling at a wavelength of 665 nm at 20$^{\circ }$ incidence.

We use these structures in a z-shaped cavity design similar to earlier reports [13,15] and employ a reflectivity output-coupling mirror with R$_{\mathrm {OC}}=99$%. Since the MESAM is mounted with an HR mirror we have a separate mount for the temperature-controlled SESAM. A schematic drawing of the z-shaped cavity is shown in Fig. 3.

 figure: Fig. 3.

Fig. 3. Schematic drawing of the used VECSEL with a z-shaped cavity design. The VECSEL gain chip is used as a folding mirror with a wedged anti-reflection coated diamond heatspreader on top while pumped with normal incidence. The SESAM or the MESAM is placed at the same position as a cavity end mirror. The outcoupling mirror is placed as the other end mirror.

Download Full Size | PDF

3. Mode locking performance

The SESAM mode-locked VECSEL performance is optimized in terms of maximum average output power with a short pulse length for fundamental mode-locked emission. Here, it proved beneficial to heat the SESAM to a temperature of 70$^{\circ }$C.

The temporal signal for the SESAM and MESAM mode-locked VECSEL is shown in Fig. 4 with the corresponding FFT spectra while the insets depict a larger span. The SESAM mode-locked VECSEL emits $P_{3.00\,\mathrm {W,\,SESAM}} = 26.8\,\mathrm {mW}$ of average power for which in Fig. 4 a stable fundamental mode-locked pulse train is visible with almost complete absence of fluctuations on the large span in the inset. The MESAM mode-locked VECSEL emission with an average power of $P_{4.25\,\mathrm {W,\,MESAM}} = 6.76\,\mathrm {mW}$ in Fig. 4(c) also shows a single pulse per round-trip while the inset has only minor envelope fluctuations. The temporal trace at an increased pump power of $P_{\mathrm {pump}} = 4.75\,$W in Fig. 4(e) also shows a single pulse per round-trip while the inset now shows on-off dynamics. This is accompanied by a drop in average power to $P_{4.75\,\mathrm {W,\,MESAM}} = 5.52\,\mathrm {mW}$. The temporal trace at a further increased pump power of $P_{\mathrm {pump}} = 5.25\,$W in Fig. 4(g) shows shorter on-periods of the periodic modulations in the inset and simultaneously the average power drops to $P_{5.25\,\mathrm {W,\,MESAM}} = 2.03\,\mathrm {mW}$. Also an increased on-off-frequency is visible in the inset.

 figure: Fig. 4.

Fig. 4. The temporal trace (left column) and the corresponding radio frequency spectra (right column) of the SESAM and MESAM mode-locked VECSEL. The insets contain a larger span. Please note that the span of insets e) and g) is four times the span in a) and c).

Download Full Size | PDF

The FFT spectrum of the SESAM mode-locked VECSEL in Fig. 4(b) has a narrow peak with an amplitude of more than 60 dB above the noise level at the repetition rate of 823 MHz. The higher harmonics in the inset have fluctuations of the peak amplitude below 4.9 dB within the 3  dB-bandwidth of the photodiode. For the MESAM at a pump power of $P_{\mathrm {pump}} = 4.25\,$W the FFT spectrum in Fig. 4(d) has a resolution-limited peak at 812 MHz with an amplitude of more than 60 dB above the noise floor. In the inset, the fluctuations of the peak amplitude of the higher harmonics are limited to below 4.7 dB. The FFT spectrum in Fig. 4(f) at a higher pump power $P_{\mathrm {pump}} = 4.75\,$W shows a lower amplitude of 50 dB for the peak at the fundamental frequency. Furthermore, the shape of the fundamental peak now develops side bands while the inset still only shows fluctuations of up to 4.6 dB for the higher harmonics. Using the MESAM with the VECSEL pumped at $P_{\mathrm {pump}} = 5.25\,$W, the FFT spectrum in Fig. 4(h) shows a peak on top of a broad pedestal while the inset has fluctuations of the peak amplitude of 2.7 dB.

A summary of the noncollinear autocorrelations corresponding to these temporal dynamics is compiled in Fig. 5(a). For the SESAM a sech$^2$-shaped temporal intensity distribution is fit to the autocorrelation to obtain good agreement and a temporal pulse width of $\tau _{\mathrm {p,SESAM }}= 4.30\,$ps. The MESAM at a pump level of $P_{\mathrm {pump}} = 4.25\,$W also provides sech$^2$-pulses with a duration of $\tau _{\mathrm {p,MESAM }}= 3.06\,$ps. However, when using the MESAM at higher pump powers of $P_{\mathrm {pump}} = 4.75\,$W and $P_{\mathrm {pump}} = 5.25\,$W, the autocorrelation suffers from noise and develops wings.

 figure: Fig. 5.

Fig. 5. a) Noncollinear autocorrelation of the SESAM and MESAM mode-locked VECSEL corresponding to the temporal trace. The stable autocorrelations of the SESAM and the MESAM at $P_{\mathrm {pump}} = 4.25\,$W are fit assuming a sech$^2$-intensity distribution which are shown as lines. b) The optical spectrum of the SESAM and MESAM mode-locked VECSEL. For the MESAM emission at a pump power of $P_{\mathrm {pump}} = 4.75\,$W a small shoulder appears as indicated by the arrow.

Download Full Size | PDF

The optical spectra are summarized in Fig. 5(b). When using the SESAM, the emission is centered at $\lambda _{\mathrm {3.00\,W, SESAM}} = 661.5\,$nm and has a spectral width of $\Delta \lambda _{\mathrm {3.00\,W, S., FWHM}} = 1.54\,$nm featuring an asymmetric spectral shape. Using the MESAM at a pump power of $P_{\mathrm {pump}} = 4.25\,$W, the emission is centered at $\lambda _{\mathrm {4.25\,W, MESAM}} = 660.7\,$nm with a width of $\Delta \lambda _{\mathrm {4.25\,W, M., FWHM}} = 1.32\,$nm and the spectral shape is almost symmetric. With the MESAM, an increase in pump power to $P_{\mathrm {pump}} = 4.75\,$W leads to a slight shoulder as indicated in Fig. 5(b) around 662 nm while the center wavelength remains unchanged at $\lambda _{\mathrm {4.75\,W, MESAM}} = 660.7\,$nm. At the highest pump power of $P_{\mathrm {pump}} = 5.25\,$W the center wavelength of the optical spectrum shifts to $\lambda _{\mathrm {5.25\,W, MESAM}} = 662\,$nm.

4. Discussion

Both the SESAM and the MESAM are capable of stable fundamental mode locking as indicated by the stable temporal trace in Fig. 4(a) and 4(c) and by the large amplitude of the first harmonics in Fig. 4(b) and Fig. 4(d). This is further supported by their smooth autocorrelations in Fig. 5(a) which also have a very good agreement with the fits according to a sech$^2$ temporal intensity distribution.

In general, temperature control of the SESAM allows fine-tuning of the modulation depth [16] and can be used in conjunction with higher pumping to leverage the increased gain while avoiding a second pulse. Without temperature tuning, in our case typically strong pumping leads to a second pulse above a certain intracavity power. This limit can be altered in case of elevated SESAM temperatures.

The reduced average output power of the MESAM compared to the SESAM mode-locked VECSEL of $P_{4.25\,\mathrm {W,\,MESAM}} = 6.76\,\mathrm {mW}$ compared to $P_{3.00\,\mathrm {W,\,SESAM}} = 26.8\,\mathrm {mW}$ is a first indication of decreased efficiency when using the MESAM. In principle, the high reflectivity of the dielectric mirror (R=99.98%) is vastly superior to semiconductor DBRs with a designed reflectivity of 99.95%. However, the processing and thereby created additional interface could contribute significantly to increased nonsaturable losses. The 11 MHz discrepancy in repetition rate between MESAM and SESAM is due to the change from a mirror mount for the MESAM to a temperature-controlled mount for the SESAM and subsequent realignment. The pump-power dependence of the MESAM mode-locked VECSEL also shows an unexpected drop in average power for increased pumping power. This is related to the on-off behavior, where a shorter on-period as well as a higher repetition is evident when increasing the pump power. We suppose that the duration of the on-period is determined by the initial intracavity power levels. Due to higher nonsaturable losses with the MESAM, this on-off behavior occurs at lower intracavity power levels compared to the SESAM. The spectral redshift of the laser emission of the MESAM mode-locked VECSEL for increasing pump power could be related to a thermal redshift of the MESAM absorption edge, increasing the losses at the initial wavelength of $\lambda _{\mathrm {4.25\,W, MESAM}} = 660.7\,$nm and leading to the location of the new gain maximum at $\lambda _{\mathrm {5.25\,W, MESAM}} = 662\,$nm.

The deviation from a resolution-limited shape of the fundamental peak for the MESAM with the VECSEL pumped at $P_{\mathrm {pump}} = 4.75\,$W in Fig. 4(f) resembles the effect of zero-padding a time trace. Here, a modulation of the true spectral shape is obtained due to the change in frequency-resolution when artificially elongating the temporal signal with zeros during the process of zero-padding. Differently, for the MESAM with the VECSEL pumped at $P_{\mathrm {pump}} = 5.25\,$W, a broad pedestal is formed below the peak at fundamental frequency which is typically indicative of amplitude fluctuations [17]. This agrees well with the time trace where now a large amount of the signal shows amplitude fluctuations. We suppose that since the amplitude stabilizes in Fig. 4(e), the zero-padding effect is more pronounced and merely obscured by the amplitude fluctuations in Fig. 4(h).

This power instability can also be observed as increased noise in the autocorrelations of Fig. 5(a) for pump powers of $P_{\mathrm {pump}} = 4.75\,$W and $P_{\mathrm {pump}} = 5.25\,$W. Here, the fluctuating pulse peak power visible in the time-trace results in stronger fluctuations of the second-harmonic power recorded for the autocorrelation.

For the SESAM mode-locked VECSEL, the obtained pulse width from Fig. 5(a) and the spectral width of $\Delta \lambda _{\mathrm {3.00\,W, S., FWHM}} = 1.54\,$nm leads to a time-bandwidth product (TBP) which is 14 times the Fourier-limited value for sech$^2$-pulses. Similarly, when using the MESAM, a spectral bandwidth of $\Delta \lambda _{\mathrm {4.25\,W, M., FWHM}} = 1.32\,$nm leads to a TBP more than 8 times its Fourier-limited value. This is because a significantly shorter pulse is generated using a narrower optical spectrum in case of the MESAM compared to the SESAM. The overall large TBP of the SESAM and MESAM mode-locked VECSEL indicates the presence of strongly chirped pulses, independent of the absorber structure. Since the antiresonant design of the gain chip should contribute minimal dispersion, we currently expect the AR-coated wedged intracavity diamond to have a large contribution to the overall dispersion. This could be possible if a slight Fabry-Perot effect is still present and introduces increased dispersion which can become more significant than the sole material dispersion of diamond.

Nevertheless, the MESAM assembly in principal allows the combination of low nonsaturable losses and sufficient modulation for mode-locked emission. The MESAM only includes the necessary losses required for modulation, which is especially significant for high-power mode-locked thin-disk lasers, where the SESAM degradation is mainly attributed to the deterioration of the semiconductor DBR [18]. To improve the thermal behavior of semiconductor chips, combinations with metals have been employed in the past already, such as metal-DBR hybrid mirrors with a deposition of metal onto the DBR backside and soldering them to a metal heatsink for VECSEL chips [19,20] and similarly also for SESAMs [7]. In general, metals provide better cooling due to their higher values of thermal conductivity compared to semiconductors (by one order of magnitude) or dielectrics (by two orders of magnitude), while also being able to enhance the reflectivity. Simulations show that combining a thermally higher-conductive heatsink like diamond next to the active region outperforms those that have a DBR placed in between in terms of temperature rise within the active region [19,21]. Therefore, an optimized temperature control for the MESAM using a high quality intracavity diamond in combination with improved processing and bonding appears as an interesting package towards improved high-power mode locking. Also, a single-side bonded membrane on a heatspreader used without a dielectric mirror or a sandwich of a membrane placed between two heatspreaders as it is used in the MECSEL again imposes itself as a next step of improvement. This assembly allows for using absorber membranes in different cavity configurations as a separate component, for example inside a ring cavity, and to increase the threshold to catastrophic optical damage of the absorber membrane.

Another DBR-free approach of a MESAM-like device using an evaporated metallic mirror on the backside of a saturable absorber membrane has already been demonstrated for generation of ultrashort pulses in Ti:Sa lasers [22,23]. The drawback of this method is the dependency of the reflectivity on the material itself. As a result, while a metallic mirror might be useful for the infrared spectral range with wavelenghts above 1 $\mathrm{\mu}$m, the significantly worse reflectivity of metals in the visible spectral range and the thereby increased nonsaturable losses due to absorption will refrain the device from starting stable mode-locked operation.

Another option to tailor the reflectivity bandwidth of the absorber device is by the use of meta-mirrors such as high-contrast grating (HCG) structures which include quantum wells or quantum dots [24]. Compared to a DBR-based structure which usually requires many DBR pairs with a low refractive index contrast from the alternating layers at the expense of a low high-reflection bandwidth, a HCG-based saturable absorber can be designed to provide a larger bandwidth while also exhibiting a lower saturation fluence as the field is enhanced inside the grating structure. Furthermore, to reduce the complexity and increase stability of the device, the use of monolithic high-contrast gratings can be of interest [25]. An implementation of the grating structure at the rear side of the absorber region can lead to a significantly lowered amount of semiconductor material that is needed to provide a high-reflection bandwidth. The benefits of this opposed to the influence of the inhomogeneous surface of a grating have to be evaluated yet.

5. Conclusion

In this paper, we have demonstrated the stable fundamental mode-locking with a MESAM, whose working principle is applicable to a broader wavelength range than the widely used SESAM. The pulse durations (4.30 ps with SESAM, 3.06 ps with MESAM) as well as the FWHM of the optical spectra (14 times and 8 times its Fourier limit) of the mode-locked VECSEL both lie in a similar range when used with the same gain chip, while for the MESAM mode-locked VECSEL the values are slightly lower. The temporal trace of the MESAM mode-locked VECSEL is stable for lower pump powers but tends to on-off dynamic behaviour for higher pump powers.

As only the active region membrane is necessary for the device, this offers multiple new possibilities for absorber designs and improvement strategies. New wavelength regions can be accessed by not only exploiting the full absorbing bandwidth in combination with mirrors with a wider stopband than typical semiconductor DBRs, but also by growth on artificial substrates, i.e. with metamorphic buffers below the active region and, hence, other material properties that are not compatible with high-quality DBRs. For enhanced thermal management, a packaged combination of the sole membrane with an efficient heat sink is an interesting approach, where already one-sided heatsinking, e.g with a diamond, would be beneficial. Similarly to the MECSEL approach, the absorber membrane can be added as a separate component with double-sided placement of diamonds into the laser to reduce the temperature rise of the saturable absorber membrane during laser operation so that it’s also suitable for high-power laser applications.

Funding

Deutsche Forschungsgemeinschaft (JE 422/9-1, MI 500/24-2).

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. 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]  

2. H. Kahle, C. M. N. Mateo, U. Brauch, P. Tatar-Mathes, R. Bek, M. Jetter, T. Graf, and P. Michler, “Semiconductor membrane external-cavity surface-emitting laser (MECSEL),” Optica 3(12), 1506–1512 (2016). [CrossRef]  

3. A. J. Kemp, G. J. Valentine, J.-M. Hopkins, J. E. Hastie, S. A. Smith, S. Calvez, M. D. Dawson, and D. Burns, “Thermal management in vertical-external-cavity surface-emitting lasers: finite-element analysis of a heatspreader approach,” IEEE J. Quantum Electron. 41(2), 148–155 (2005). [CrossRef]  

4. H.-M. Phung, H. Kahle, J.-P. Penttinen, P. Rajala, S. Ranta, and M. Guina, “Power scaling and thermal lensing in 825 nm emitting membrane external-cavity surface-emitting lasers,” Opt. Lett. 45(2), 547–550 (2020). [CrossRef]  

5. B. Jeżewski, A. Broda, I. Sankowska, A. Kuźmicz, K. Gołaszewska-Malec, K. Czuba, and J. Muszalski, “Membrane external-cavity surface-emitting laser emitting at 1640 nm,” Opt. Lett. 45(2), 539–542 (2020). [CrossRef]  

6. A. Broda, B. Jezewski, M. Szymanski, and J. Muszalski, “High-power 1770 nm emission of a membrane external-cavity surface-emitting laser,” IEEE J. Quantum Electron. 57(1), 1–6 (2021). [CrossRef]  

7. A. Diebold, T. Zengerle, C. G. E. Alfieri, C. Schriber, F. Emaury, M. Mangold, M. Hoffmann, C. J. Saraceno, M. Golling, D. Follman, G. D. Cole, M. Aspelmeyer, T. Südmeyer, and U. Keller, “Optimized SESAMs for kilowatt-level ultrafast lasers,” Opt. Express 24(10), 10512–10526 (2016). [CrossRef]  

8. S. Suomalainen, A. Vainionpää, O. Tengvall, T. Hakulinen, S. Karirinne, M. Guina, O. G. Okhotnikov, T. G. Euser, and W. L. Vos, “Long-wavelength fast semiconductor saturable absorber mirrors using metamorphic growth on GaAs substrates,” Appl. Phys. Lett. 87(12), 121106 (2005). [CrossRef]  

9. S. Gupta, J. Whitaker, and G. Mourou, “Ultrafast carrier dynamics in III-V semiconductors grown by molecular-beam epitaxy at very low substrate temperatures,” IEEE J. Quantum Electron. 28(10), 2464–2472 (1992). [CrossRef]  

10. L. Qian, S. D. Benjamin, P. W. E. Smith, B. J. Robinson, and D. A. Thompson, “Subpicosecond carrier lifetime in beryllium-doped InGaAsP grown by He-plasma-assisted molecular beam epitaxy,” Appl. Phys. Lett. 71(11), 1513–1515 (1997). [CrossRef]  

11. J. T. Gopinath, E. R. Thoen, E. M. Koontz, M. E. Grein, L. A. Kolodziejski, E. P. Ippen, and J. P. Donnelly, “Recovery dynamics in proton-bombarded semiconductor saturable absorber mirrors,” Appl. Phys. Lett. 78(22), 3409–3411 (2001). [CrossRef]  

12. E. L. Delpon, J. L. Oudar, N. Bouché, R. Raj, A. Shen, N. Stelmakh, and J. M. Lourtioz, “Ultrafast excitonic saturable absorption in ion-implanted InGaAs/InAlAs multiple quantum wells,” Appl. Phys. Lett. 72(7), 759–761 (1998). [CrossRef]  

13. M. Großmann, R. Bek, M. Jetter, and P. Michler, “Stable fundamental and dual-pulse mode locking of red-emitting VECSELs,” Laser Phys. Lett. 17(6), 065001 (2020). [CrossRef]  

14. Z. L. Liau, “Semiconductor wafer bonding via liquid capillarity,” Appl. Phys. Lett. 77(5), 651–653 (2000). [CrossRef]  

15. S. Ranta, A. Härkönen, T. Leinonen, L. Orsila, J. Lyytikäinen, G. Steinmeyer, and M. Guina, “Mode-locked VECSEL emitting 5 ps pulses at 675 nm,” Opt. Lett. 38(13), 2289–2291 (2013). [CrossRef]  

16. P. Klopp, U. Griebner, M. Zorn, A. Klehr, A. Liero, M. Weyers, and G. Erbert, “Mode-locked InGaAs-AlGaAs disk laser generating sub-200-fs pulses, pulse picking and amplification by a tapered diode amplifier,” Opt. Express 17(13), 10820–10834 (2009). [CrossRef]  

17. D. Von der Linde, “Characterization of the noise in continuously operating mode-locked lasers,” Appl. Phys. B 39(4), 201–217 (1986). [CrossRef]  

18. C. Saraceno, C. Schriber, F. Emaury, O. Heckl, C. Baer, M. Hoffmann, K. Beil, C. Kränkel, M. Golling, T. Südmeyer, and U. Keller, “Cutting-edge high-power ultrafast thin disk oscillators,” Appl. Sci. 3(2), 355–395 (2013). [CrossRef]  

19. Z. Zhao, S. Bouchoule, L. Ferlazzo, A. Sirbu, A. Mereuta, E. Kapon, E. Galopin, J.-C. Harmand, J. Decobert, and J.-L. Oudar, “Cost-effective thermally-managed 1.55-µm VECSEL with hybrid mirror on copper substrate,” IEEE J. Quantum Electron. 48(5), 643–650 (2012). [CrossRef]  

20. K. Gbele, A. Laurain, J. Hader, W. Stolz, and J. V. Moloney, “Design and fabrication of hybrid metal semiconductor mirror for high-power VECSEL,” IEEE Photonics Technol. Lett. 28(7), 732–735 (2016). [CrossRef]  

21. V. Iakovlev, J. Walczak, M. Geębski, A. K. Sokoł, M. Wasiak, P. Gallo, A. Sirbu, R. P. Sarzała, M. Dems, T. Czyszanowski, and E. Kapon, “Double-diamond high-contrast-gratings vertical external cavity surface emitting laser,” J. Phys. D: Appl. Phys. 47(6), 065104 (2014). [CrossRef]  

22. R. Fluck, I. D. Jung, G. Zhang, F. X. Kärtner, and U. Keller, “Broadband saturable absorber for 10-fs pulse generation,” Opt. Lett. 21(10), 743–745 (1996). [CrossRef]  

23. I. D. Jung, F. X. Kärtner, N. Matuschek, D. H. Sutter, F. Morier-Genoud, G. Zhang, U. Keller, V. Scheuer, M. Tilsch, and T. Tschudi, “Self-starting 6.5-fs pulses from a Ti:sapphire laser,” Opt. Lett. 22(13), 1009–1011 (1997). [CrossRef]  

24. C. Chang-Hasnain, C. Chase, W. Yang, and F. Sedgwick, “High contrast grating based saturable absorber for mode-locked lasers and its applications in passively mode-locked vertical-cavity surface-emitting lasers,” PCT patent WO 2011/106554 A3 (2011).

25. M. Gebski, M. Dems, A. Szerling, M. Motyka, L. Marona, R. Kruszka, D. Urbańczyk, M. Walczakowski, N. Pałka, A. Wójcik-Jedlińska, Q. J. Wang, D. H. Zhang, M. Bugajski, M. Wasiak, and T. Czyszanowski, “Monolithic high-index contrast grating: a material independent high-reflectance VCSEL mirror,” Opt. Express 23(9), 11674–11686 (2015). [CrossRef]  

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (5)

Fig. 1.
Fig. 1. Refractive index profile and intensity of the simulated electric field of the saturable absorber membrane structure. In the case shown, the electric field is maximal at the left quantum well so that absorption can take place and minimal at the right quantum well, where no electric field intensity is absorbed.
Fig. 2.
Fig. 2. (a) Tilt view SEM image of the free-standing membrane after wet-chemical processing. The membrane thickness of approximately 150 nm is indicated by the white bar. (b) Photograph of the MESAM in a mirror mount.
Fig. 3.
Fig. 3. Schematic drawing of the used VECSEL with a z-shaped cavity design. The VECSEL gain chip is used as a folding mirror with a wedged anti-reflection coated diamond heatspreader on top while pumped with normal incidence. The SESAM or the MESAM is placed at the same position as a cavity end mirror. The outcoupling mirror is placed as the other end mirror.
Fig. 4.
Fig. 4. The temporal trace (left column) and the corresponding radio frequency spectra (right column) of the SESAM and MESAM mode-locked VECSEL. The insets contain a larger span. Please note that the span of insets e) and g) is four times the span in a) and c).
Fig. 5.
Fig. 5. a) Noncollinear autocorrelation of the SESAM and MESAM mode-locked VECSEL corresponding to the temporal trace. The stable autocorrelations of the SESAM and the MESAM at $P_{\mathrm {pump}} = 4.25\,$W are fit assuming a sech$^2$-intensity distribution which are shown as lines. b) The optical spectrum of the SESAM and MESAM mode-locked VECSEL. For the MESAM emission at a pump power of $P_{\mathrm {pump}} = 4.75\,$W a small shoulder appears as indicated by the arrow.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.