Optically pumped semiconductor disk lasers are an important class of solid state lasers. Despite all their advantages, however, they suffer from heat incorporation into the active region caused by the excess energy of the pump photons. To overcome the limits of common methods in thermal management, we realized a semiconductor membrane external-cavity surface-emitting laser (MECSEL) consisting of a diamond heat spreader sandwiched active region design without a monolithically integrated distributed Bragg reflector (DBR). This diamond-sandwich approach improves the heat dissipation out of the active region and makes generally low-heat conductive DBRs obsolete. In an AlGaInP-based system, we demonstrate 595 mW output power at a wavelength of 657 nm and heatsink temperature of 10°C. The MECSEL enables a variety of new material combinations for new laser wavelengths and further potential for power scaling.
© 2016 Optical Society of America
Optically pumped semiconductor vertical external-cavity surface-emitting lasers (VECSELs) nowadays  have reached the femtosecond-pulse regime  in mode-locked operation as well as the 100-watt level  in continuous wave output. There is strongly temperature-dependent performance due to the heating from the large quantum defect, limited carrier confinement, and the interplay of gain and cavity resonance with temperature . The latter is especially a challenge for the AlGaInP material system [5,6], in which the confinement is rather low and the laser structure is based on a thick distributed Bragg reflector (DBR). The thermal conductivity of this DBR is one order of magnitude inferior compared to well-conducting metals, which are often used as a backside heatsink, and two orders of magnitude worse compared to diamond, which is used as a common backside or intra-cavity (IC) heat spreader . The semiconductor structure itself, with a thickness of several micrometers (active region plus DBR) and the substrate, with a typical thickness of 350 μm, impede the heat flow out of the active region. Numerous strategies for thermal management such as heat spreader arrangement , substrate removing [9,10], flip-chip processes , or the insertion of compound mirrors [12,13] have led to continued improvements in the performance of VECSELs. Following this path further, the corollary would be finally abandoning each semiconductor part of the VECSEL not essentially needed to build up a whole laser. This can be achieved by growing the active region directly onto the substrate (see Subsection 2.A) without the DBR, removing the substrate (see Subsection 2.B), and finally embedding the released active region membrane in between two diamond heat spreaders (see Subsection 2.C) to create a compact gain device with superior cooling. Regarding a fixed temperature increase inside the active region, the simulations of Yang et al.  clearly show an up to one magnitude higher pump power and thus an up to one magnitude higher output power comparing the standard VECSEL heat spreader design with the MECSEL approach. Such a sandwiched active-region membrane configuration would additionally allow the growth of semiconductor structures that are otherwise impossible to grow due to limitations imposed by the need to lattice-match the DBR to the substrate or the active region to the DBR. This concept to improve the cooling in a VECSEL was theoretically studied and simulated by Iakovlev et al.  in 2014. A DBR-free VECSEL has been realized with a released active region bonded to one side of an IC heat spreader  recently.
In the present paper, we introduce the experimental proof of principle (see Fig. 1) of the semiconductor membrane external-cavity surface-emitting laser (MECSEL ) where an optically pumped, multi-quantum-well-containing semiconductor membrane is squeezed between two antireflection-coated diamond heat spreaders. In addition to the fundamental characterization of this new laser, a comparison with a conventional VECSEL comprising the identical active region is included.
A. Semiconductor Architecture
The investigated gain membrane as well as the conventional VECSEL were grown by metal-organic vapor-phase epitaxy (MOVPE) in an Aixtron 3 x 2″ showerhead reactor. Standard sources (trimethylgallium, trimethylindium, trimethylaluminum, arsine, and phosphine) were used. The deposition took place at a pressure of 100 mbar and a growth temperature of about 650°C on -GaAs substrates, misoriented by 6° toward the direction. The gain region (detailed scheme shown in Fig. 2) is built up of 20 compressively strained (GaInP) quantum wells (QWs) with thicknesses of 5 nm and distances of 4 nm arranged in five packages. Each package is placed at the antinode of the simulated electrical standing wave to depict the typical resonant periodic gain structure. Lattice-matched () and slightly compressively strained () layers are used as barrier and cladding material, respectively. One lattice-matched 12 nm thick (AlInP) electron-blocking layer followed by one 10 nm thick () capping layer on each side encloses the structure. The capping layer acts as protection layer to prevent oxidation as well as an etch stop layer for the selective etching of a 200 nm AlAs layer, which separates the active region from the substrate (for process details, see Subsection 2.B). In the case of the conventional VECSEL, a -thick DBR (for details see ) is placed instead of the 200 nm AlAs layer. The whole structure of the MECSEL was designed to depict a cavity for a wavelength of 665 nm, which leads to a designed thickness of 587 nm.
B. Membrane Process and Bonding
The following process was applied to etch the sample and finally bond the semiconductor membrane to the diamond heat spreaders: A 200 nm AlAs layer separates the active region from the substrate (see Fig. 2) and introduces the selectivity for the etching process. The sample, a roughly piece cleaved out of the wafer, is fixed to a stripe of a silicon wafer, which acts as sample carrier with the epitaxial side down, using Crystalbond. An ammonium hydroxide solution (, 1: 3) is used to completely dissolve the substrate. The ammonium process stops selectively at the AlAs layer. The AlAs layer is then removed by dipping the sample into hydrofluoric acid (, 1: 9) for just five seconds. The result is an approximately 600 nm thick semiconductor membrane (see Fig. 3) consisting of the active region of a VECSEL only (see Fig. 2) and still sticking to the silicon sample carrier. The sample carrier is then cleaved several times to get rid of the sample’s edge parts, which were attacked during the foregone process steps. Then the sample is dipped into acetone for about three hours, which completely dissolves the Crystalbond, leaving a fully released quantum membrane floating in acetone. The acetone is replaced by clean isopropanol, and a small piece of the membrane () is transferred to one of the diamond heat spreaders. After that, the other diamond heat spreader is placed onto the membrane when there is still some isopropanol left. Then the whole package is mechanically squeezed in a brass mount which is designed to also hold the sandwiched active region during laser operation. This leads to a mechanically enforced bonding between the membrane and the heat spreaders (see Fig. 4).
C. Quality Inspection
A close look at the microscope photograph in Fig. 4 reveals two important details. First of all, no damage aside from some cracks in the upper right corner and a small damaged area in the center right is visible. Furthermore, the surface of the membrane seems to be totally flat. This represents the most important precondition to realizing a good bonding. Secondly, the color gradient, which is induced by the Fabry–Perot effect between the two diamond heat spreaders and surrounds the whole membrane, illustrates the homogeneous thickness since the same color appears at all positions close to the membrane. This supports the impression of a good bonding when looking at the microscope membrane photograph itself. After this quality check, the diamond-squeezed gain membrane is immediately ready to act as a gain element in a laser resonator.
3. CHARACTERIZATION OF THE MECSEL
A. Laser Setup
The laser experiments were performed in a linear concentric resonator (Fig. 5). The resonator mirrors could be adjusted in all degrees of freedom and the sample holder could be shifted in the sample’s plane and also tilted in two axes. A 532 nm, 5 W Finesse was used as pump laser, irradiating the sample under an angle of 15° to its normal. The two diamond heat spreaders enclosing the wet chemically released semiconductor gain membrane were antireflection-coated for the laser emission wavelength on the outer facets. For the power measurements, an output coupler with a reflectivity of (radius of curvature ) and a highly reflective mirror ( for 500–760 nm, ) were chosen. The cavity length was adjusted to roughly 148.5 mm, which leads to a mode diameter of approximately 76 μm at the beam waist of the resonator where the membrane was placed. In terms of mode matching, the pump spot diameter was adjusted by adapting the distance of the pump lens to 103 mm, corresponding to a pump spot diameter of approximately 80 μm in the short axis.
B. Power Measurements
The system was operated at a heatsink temperature of 10°C. Figure 6 shows the output power of the MECSEL as a function of incident pump power. A linear input-to-output behavior with a slope of 22.3% can be clearly seen. The maximum output power of 595 mW was reached at an incident pump power of 3.67 W with the laser threshold at 1.0 W. The corresponding VECSEL comprising a DBR (see Subsection 2.A) was tested under the same conditions with respect to pump spot size, mode diameter, outcoupling mirror, and heatsink temperature. The measurement is also plotted in Fig. 6. The best slope efficiency achieved here with the green pumped VECSEL was 18.8%, with a threshold pump power of 0.8 W. While the threshold of the MECSEL is slightly higher, the slope efficiency of 22.3% significantly exceeds any slope efficiency published before  with green pumped conventional VECSELs in the AlGaInP material system at the elevated heatsink temperature of . Furthermore, the maximum output power of only 570 mW for the VECSEL was achieved at 4.6 W of incident pump power where the thermal rollover already led to a saturation of the output power while, for the MECSEL, a maximum of 595 mW was reached at 3.7 W incident pump power before an output power breakdown was visible. The not-prefigured breakdown of the MECSEL indicates sudden loss of bonding between the diamond heat spreaders and the semiconductor membrane.
C. Pumping Scheme
Another major advantage of the MECSEL over the conventional VECSEL concerns the pumping process. Usually, the DBR has to be designed in such a way that the emitted laser light as well as the unabsorbed pump light are reflected. If the DBR cannot be fabricated in such a manner due to epitaxial restrictions (material parameters, strain, etc.), the residual pump light will be absorbed in the DBR. Together with the quantum defect, which is roughly 20% in the AlGaInP material system with a barrier pumping scheme at 532 nm, unwanted heat is produced inside but also close to the active region in the DBR. This further restricts the performance of the conventional VECSEL. In the MECSEL, the pump light that is not directly absorbed is just transmitted without disturbing the gain system. The characterization of the absorption efficiency can be directly performed inside the cavity and exactly determined by subtracting the transmitted and reflected light from the incident pump power (Fig. 7). Accordingly, the absorption efficiency , which is part of the differential efficiency of the laser, can be exactly calculated . Therefore, the internal parameters of the MECSEL (efficiencies, etc.) can be determined more accurately. These values are also valid for conventional VECSELs if the different refractive index transition between DBR and active region, instead of diamond and active region, is taken into account. In this way, the MECSEL itself can be used as characterization setup for isolated semiconductor gain regions, separated from the influence of the surrounding semiconductor material.
D. Wavelength Tuning
To obtain information about the spectral range of the gain delivered by the membrane, a wavelength-tuning measurement is of interest. In particular, the reduced active region subcavity enhancement  of the MECSEL (two times diamond/active region interface instead of the combination of diamond/active region and active region/DBR) possibly supports a larger spectral width of amplification. Therefore, a set of spectra was taken (see Fig. 8) at a heatsink temperature of 3°C and absorbed pump power of . Broadband highly reflective mirrors ( for 640–700 nm) in a 50–150 mm cavity (see Fig. 5) were used. For wavelength tuning, a 1 mm birefringent filter was adjusted at Brewster’s angle inside the cavity and rotated around the normal axis of its surface to perform the spectral shift. The measurements reveal a tuning range of nearly 24 nm (649.9–673.6 nm), which is the highest value achieved in this spectral range by semiconductor lasers to date. The conventional VECSEL showed a tuning range of (656–678 nm) under the same conditions but at a heatsink temperature of . The tuning measurements revealed that the spectral range of the MECSEL was about 5 nm blueshifted compared to the conventional VECSEL, which cannot be completely explained by the 7°C lower heatsink temperature. According to earlier work , there is the possibility that strain, which sums up when stacking the QW packages during growth and therefore causes a redshift of several nanometers, relaxes again when the gain membrane is released from its substrate (see Subsections 3.E and 3.F).
E. Spectra of the Free-Running Lasers
Figure 9(a) shows a typical spectrum of the free-running (no birefringent filter or etalon in the cavity) MECSEL recorded during the power measurement. First of all, the width and the shape of the spectrum are very conspicuous here. The spectral width of laser emission of the corresponding free-running VECSEL is below 2 nm and can be seen in Fig. 9(b). For the MECSEL, we observed simultaneous laser emission in a range of more than 6.5 nm. This can be explained by the reduction of the sub-cavity (DBR and semiconductor–heat spreader interface) effect , which is indicated by the typical cavity dip occurring in reflectivity measurements of conventional VECSELs . In the MECSEL, this narrowing and frequency preselective part is missing and the gain bandwidth can show more of its potential. A closer inspection of the spectrum reveals further details; one is the Fabry–Perot oscillation, which is visible in both MECSEL and VECSEL spectra [Figs. 9(a) and 9(b)] with a spacing of and is impressed onto the whole emission spectrum due to the two approximately 550 μm thick IC diamond heat spreaders. Although the diamonds used for the MECSEL are antireflection-coated on one side each and in contact with the semiconductor membrane on the other, their impact is still large enough to show the diamond-introduced Fabry–Perot oscillation. Another point is the beat note that can be identified in the MECSEL’s spectrum [Fig. 9(a)]. It originates from slightly different thicknesses of the two diamond heat spreaders. One is measured to be and the other is in thickness. The Fabry–Perot effect and therefore the beat note can be avoided by the use of wedged heat spreaders or a tilt of the whole gain package, preventing an overlap of internal reflections.
F. Photoluminescence Characterization
In order to determine possible reasons for the blueshift described in Subsections 3.D and 3.E and visible in the spectra in Fig. 9, photoluminescence (PL) measurements have been performed (see Fig. 10) comparing the unprocessed semiconductor membrane (as grown, with substrate), the corresponding VECSEL structure, and the fully processed and diamond-squeezed MECSEL. The PL measurements from the side are chosen to bypass effects of the DBR and the sub-cavity of the VECSEL. The spectra obtained for surface and edge PL of the unprocessed semiconductor membrane sample coincide with each other. The edge PL measurement of the VECSEL shows only a spectral difference of less than 2 nm, which can be due to fabrication tolerances. However, the surface PL spectrum of the fully processed and diamond-squeezed MECSEL shows a 4 nm blueshift compared to the unprocessed semiconductor membrane sample. The comparison of the laser spectra of the free-running systems [see Figs. 9(a) and 9(b)] even shows an offset of 11 nm and supports the assumption that strain effects are responsible for the spectral shift as well, and possibly have an even stronger impact on the spectral position of the gain. Further investigations are necessary here to fully resolve the effects connected to strain relaxation after processing and external stress  applied by the diamond heat spreaders.
G. Beam Quality
Figure 11 shows the beam profile, recorded with a CMOS camera at a distance of 20 cm behind the highly reflective mirror. The carefully adjusted resonator delivers a fundamental Gaussian mode with a beam quality factor of ( and , including a device accuracy of , see Fig. 12), measured at 160 mW output power with a Coherent ModeMaster.
With this work we have realized for the first time, to our knowledge, a novel laser system: the heat spreader sandwiched semiconductor MECSEL. We have demonstrated the superior properties of the MECSEL, including near-room-temperature operation with improved slope efficiency compared to the conventional VECSEL system as well as the highest output power achieved at this heatsink temperature to date with a barrier-pumped AlGaInP material system. Nevertheless, there are processing issues which currently prevent this MECSEL from showing an even better performance. It is well known that the quality of the thermal contact between heat spreader and semiconductor is essential for the performance of VECSELs. The slightly higher threshold of the MECSEL is either caused by the doubled number of passes through the diamond per roundtrip or a first hint that the bonding was not optimized. Additionally, the breakdown of emission of the membrane laser occurred slightly earlier compared to the thermal rollover of the VECSEL, although less heat per incident power is introduced into the semiconductor structure since the residual pump light is transmitted and does not heat up the DBR. Local debonding introduced by thermal expansion of the membrane is a possible explanation. To fully exploit the potential of this new semiconductor laser concept, the bonding process needs to be further investigated. High-quality bonding processes  could lead to important improvements here.
This concept also overcomes the former limitations imposed by the need to grow a distributed Bragg reflector, thereby expanding the choice of possible materials and compositions and, by this, the accessible wavelength range. The Gallium–Nitride material system, where demonstrated VECSELs  nowadays deliver relatively low performance, could benefit greatly from this new laser design and make blue and green  laser emission possible. The MECSEL approach could also enable AlGaInP-based optically pumped semiconductor lasers in the orange spectral range , which is currently not possible due to absorption of the emitted orange light in the DBR required by VECSEL approaches. Lasers based on the InP material system  for emission wavelengths around 1.5 μm could also greatly benefit from this technology as the thermal conductivity, especially that of the DBR, is significantly low . The necessary high-quality selective etching processes are also available in other material systems [28,29] and are scalable to the size of whole wafers . Also, large-scale bonding processes  can be applied. As a further benefit, the gain package could be sealed to prevent oxidation of the semiconductor membrane. Furthermore, there is the possibility of in-well and multipass pumping [31,32] in a transmission configuration, where the pump light is recycled and folded several times through the active region, which would push the pump efficiency of such devices to new frontiers. In addition, all this can be directly adapted to the classical solid-state thin-disk laser concept [33,34], increasing its performance due to the optimized thermal management. This is also under current investigation .
Deutsche Forschungsgemeinschaft (DFG) (Br 3606/4-1, Mi 900/24-1).
The authors would like to thank Thomas Schwarzbäck for giving the pioneering idea for this project, Sergej Vollmer for technical support with the MOVPE, Jelde Elling for SEM pictures, and Annebärbel Fuoss for performing the HF process step.
1. O. G. Okhotnikov, ed., Semiconductor Disk Lasers: Physics and Technology (Wiley-VCH, 2010).
2. B. W. Tilma, M. Mangold, C. A. Zaugg, S. M. Link, D. Waldburger, A. Klenner, A. S. Mayer, E. Gini, M. Golling, and U. Keller, “Recent advances in ultrafast semiconductor disk lasers,” Light Sci. Appl. 4, e310 (2015). [CrossRef]
3. B. Heinen, T. L. Wang, M. Sparenberg, A. Weber, B. Kunert, J. Hader, S. W. Koch, J. V. Moloney, M. Koch, and W. Stolz, “106 W continuous-wave output power from vertical-external-cavity surface-emitting laser,” Electron. Lett. 48, 516–517 (2012). [CrossRef]
4. 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, 2228–2236 (2009). [CrossRef]
5. Ł. Piskorski, R. P. Sarzała, and W. Nakwaski, “Enhanced single-fundamental lp01 mode operation of 650-nm GaAs-based GaInP/AlGaInP quantum-well VCSELs,” Appl. Phys. A 98, 651–657 (2010). [CrossRef]
6. M. A. Afromowitz, “Thermal conductivity of ga1-xalx GaAlAs alloys,” J. Appl. Phys. 44, 1292–1294 (1973). [CrossRef]
7. 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, 148–155 (2005). [CrossRef]
8. P. Holl, M. Rattunde, S. Adler, S. Kaspar, W. Bronner, A. Bächle, R. Aidam, and J. Wagner, “Recent advances in power scaling of GaSb-based semiconductor disk lasers,” IEEE J. Sel. Top. Quantum Electron. 21, 324–335 (2015). [CrossRef]
9. M. Kuznetsov, F. Hakimi, R. Sprague, and A. Mooradian, “High-power (>0.5-W CW) diode-pumped vertical-external-cavity surface-emitting semiconductor lasers with circular tem00 beams,” IEEE Photon. Technol. Lett. 9, 1063–1065 (1997). [CrossRef]
10. E. Gerster, I. Ecker, S. Lorch, C. Hahn, S. Menzel, and P. Unger, “Orange-emitting frequency-doubled GaAsSb/GaAs semiconductor disk laser,” J. Appl. Phys. 94, 7397–7401 (2003). [CrossRef]
11. A. Rantamäki, E. Saarinen, J. Lyytikäinen, J. Heikkinen, J. Kontio, K. Lahtonen, M. Valden, and O. Okhotnikov, “Thermal management in long-wavelength flip-chip semiconductor disk lasers,” IEEE J. Sel. Top. Quantum Electron. 21, 336–342 (2015). [CrossRef]
12. A. Rantamäki, E. J. Saarinen, J. Lyytikäinen, K. Lahtonen, M. Valden, and O. G. Okhotnikov, “High power semiconductor disk laser with a semiconductor-dielectric-metal compound mirror,” Appl. Phys. Lett. 104, 101110 (2014). [CrossRef]
13. K. Gbele, A. Laurain, J. Hader, W. Stolz, and J. V. Moloney, “Design and fabrication of hybrid metal semiconductor mirror for high-power VECSELs,” IEEE Photon. Technol. Lett. 28, 732–735 (2016). [CrossRef]
14. Z. Yang, A. R. Albrecht, J. G. Cederberg, and M. Sheik-Bahae, “Optically pumped DBR-free semiconductor disk lasers,” Opt. Express 23, 33164–33169 (2015). [CrossRef]
15. V. Iakovlev, J. Walczak, M. Gębski, A. K. Sokół, 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 47, 065104 (2014). [CrossRef]
16. H. Kahle, C. M. N. Mateo, U. Brauch, R. Bek, T. Schwarzbäck, M. Jetter, T. Graf, and P. Michler, “Gain chip design, power scaling and intra-cavity frequency doubling with LBO of optically pumped red-emitting AlGaInP-VECSELs,” Proc. SPIE 9734, 97340T (2016).
17. T. Schwarzbäck, M. Eichfelder, W.-M. Schulz, R. Roßbach, M. Jetter, and P. Michler, “Short wavelength red-emitting AlGaInP-VECSEL exceeds 1.2 W continuous-wave output power,” Appl. Phys. B 102, 789–794 (2011). [CrossRef]
18. T. Schwarzbäck, R. Bek, F. Hargart, C. A. Kessler, H. Kahle, E. Koroknay, M. Jetter, and P. Michler, “High-power InP quantum dot based semiconductor disk laser exceeding 1.3 W,” Appl. Phys. Lett. 102, 092101 (2013). [CrossRef]
19. 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 Photon. Rev. 3, 407–434 (2009). [CrossRef]
20. T. Schwarzbäck, H. Kahle, M. Jetter, and P. Michler, “Strain compensation techniques for red AlGaInP-VECSELs: performance comparison of epitaxial designs,” J. Cryst. Growth 370, 208–211 (2013). [CrossRef]
21. M. Müller, N. Linder, C. Karnutsch, W. Schmid, K. P. Streubel, J. Luft, S.-S. Beyertt, A. Giesen, and G. H. Doehler, “Optically pumped semiconductor thin-disk laser with external cavity operating at 660 nm,” Proc. SPIE 4649, 265–271 (2002). [CrossRef]
22. O. P. Kowalski, J. W. Cockburn, D. J. Mowbray, M. S. Skolnick, R. Teissier, and M. Hopkinson, “GaInP–AlGaInP band offsets determined from hydrostatic pressure measurements,” Appl. Phys. Lett. 66, 619–621 (1995). [CrossRef]
23. 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, 10512–10526 (2016). [CrossRef]
24. T. Wunderer, J. E. Northrup, Z. Yang, M. Teepe, N. M. Johnson, P. Rotella, and M. Wraback, “In-well pumped blue GaN-based vertical-external-cavity surface-emitting lasers,” Jpn. J. Appl. Phys. 52, 08JG11 (2013). [CrossRef]
25. B. Galler, M. Sabathil, A. Laubsch, T. Meyer, L. Hoeppel, G. Kraeuter, H. Lugauer, M. Strassburg, M. Peter, A. Biebersdorf, U. Steegmueller, and B. Hahn, “Green high-power light sources using InGaN multi-quantum-well structures for full conversion,” Phys. Status Solidi C 8, 2369–2371 (2011). [CrossRef]
26. L. Toikkanen, A. Härkönen, J. Lyytikäinen, T. Leinonen, A. Laakso, A. Tukiainen, J. Viheriälä, M. Bister, and M. Guina, “Optically pumped edge-emitting GaAs-based laser with direct orange emission,” IEEE Photon. Technol. Lett. 26, 384–386 (2014). [CrossRef]
27. H. Lindberg, M. Strassner, J. Bengtsson, and A. Larsson, “InP-based optically pumped VECSEL operating CW at 1550 nm,” IEEE Photon. Technol. Lett. 16, 362–364 (2004). [CrossRef]
28. J. M. M. Santos, B. E. Jones, P. J. Schlosser, S. Watson, J. Herrnsdorf, B. Guilhabert, J. J. D. McKendry, J. D. Jesus, T. A. Garcia, M. C. Tamargo, A. E. Kelly, J. E. Hastie, N. Laurand, and M. D. Dawson, “Hybrid GaN LED with capillary-bonded II-VI MQW color-converting membrane for visible light communications,” Semicond. Sci. Technol. 30, 035012 (2015). [CrossRef]
29. E. Kantola, T. Leinonen, J.-P. Penttinen, V.-M. Korpijärvi, and M. Guina, “615 nm GaInNAs VECSEL with output power above 10 W,” Opt. Express 23, 20280–20287 (2015). [CrossRef]
30. C.-W. Cheng, K.-T. Shiu, N. Li, S.-J. Han, L. Shi, and D. K. Sadana, “Epitaxial lift-off process for gallium arsenide substrate reuse and flexible electronics,” Nat. Commun. 4, 1577 (2013). [CrossRef]
31. C. M. N. Mateo, U. Brauch, T. Schwarzbäck, H. Kahle, M. Jetter, M. A. Ahmed, P. Michler, and T. Graf, “Enhanced efficiency of AlGaInP disk laser by in-well pumping,” Opt. Express 23, 2472–2486 (2015). [CrossRef]
32. C. M. N. Mateo, U. Brauch, H. Kahle, T. Schwarzbäck, M. Jetter, M. A. Ahmed, P. Michler, and T. Graf, “2.5 W continuous wave output at 665 nm from a multipass and quantum-well-pumped AlGaInP vertical-external-cavity surface-emitting laser,” Opt. Lett. 41, 1245–1248 (2016). [CrossRef]
33. A. Giesen and J. Speiser, “Fifteen years of work on thin-disk lasers: results and scaling laws,” IEEE J. Sel. Top. Quantum Electron. 13, 598–609 (2007). [CrossRef]
34. P. Millar, R. B. Birch, A. J. Kemp, and D. Burns, “Synthetic diamond for intracavity thermal management in compact solid-state lasers,” IEEE J. Quantum Electron. 44, 709–717 (2008). [CrossRef]
35. Ultrafast High-Average Power Ti:Sapphire Thin-Disk Oscillator and Amplifiers, 2013, http://www.tisa-td.eu/.