SPIE Startup Challenge 2015 Founding Partner - JENOPTIK Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:

SPIE Photonics West 2017 | Register Today

SPIE Defense + Commercial Sensing 2017 | Call for Papers

Get Down (loaded) - SPIE Journals OPEN ACCESS


Print PageEmail PageView PDF

Lasers & Sources

Toward 3 micron wavelength semiconductor disk lasers

High-performance semiconductor disk lasers for the wavelength range between 1.9 and 2.8μm open up new opportunities in gas sensing, communications, and materials processing.
1 September 2010, SPIE Newsroom. DOI: 10.1117/2.1201007.002987

There are numerous industrial, medical, and security-related applications for which high-brightness, flexible laser sources emitting at IR wavelengths in the 2–3μm range are highly desirable. They include long-range sensing—such as light detection and ranging (LIDAR) for clear-air turbulence or gas detection—long-range free-space optical communications, as well as medical diagnostics and therapy. Medical applications make use of the distinct absorption features of human tissue (which are mostly determined by water absorption) at 2 and 2.9μm, making these wavelengths ideal for high-precision laser scalpels. This has driven the development of a new category of semi-conductor laser, long-wavelength optically pumped semi-conductor disk lasers (OPSDLs), also known as vertical external-cavity surface-emitting lasers (VECSELs).1

In the 1μm wavelength range, with laser structures based on the gallium arsenide (GaAs) material system, OPSDLs with output powers of up to several tens of Watts have been realized.2 Most recently, these lasers have entered the commercial market with great success, either with their fundamental laser emission around 1μm or frequency doubled to cover the visible range.2,3 At wavelengths longer than 1μm, however, the output power and efficiency of OPSDLs generally degrade, and often significantly so.

We have developed a mature OPSDL technology based on the group III-antimonide material system to take advantage of the 2μm wavelength range. In demonstrating the first 2.8μm room-temperature continuous-wave (CW) OPSDL, we have further expanded the wavelength range of this class of semiconductor laser. This was possible through the combined effect of improved semiconductor structure design, and better molecular-beam-epitaxy (MBE) fabrication steps. The MBE growth of this long-wavelength structure is very challenging because of its overall layer thickness of nearly 12μm and the special material composition needed for the long-wavelength quantum wells in the active region. Moreover, we have shown the versatility of these OPSDLs by demonstrating high-power, tunable, and narrow-linewidth 1.9–2.8μm lasers.

To reach high output powers, we had to ensure efficient heat removal from the active region. Therefore, we bonded small pieces of the grown heterostructure to a silicon carbide (SiC) or diamond transparent heat spreader using the liquid capillary bonding technique. We then mounted the bonded samples into a special submount (see Figure 1). This mounted semiconductor structure is the basis for all our subsequent laser-resonator and module development.

Figure 1. Optically pumped semiconductor disk laser (OPSDL) structure with transparent heat spreader, mounted inside a submount.

By optimizing heterostructure design, MBE growth, and bonding technology, the semiconductor OPSDL structures set new international standards for high-power, high-efficiency semiconductor disk lasers above 2μm wavelength.4 We achieved an output power of 3W in CW operation at 2.0μm emission wavelength for a heat-sink temperature of 20°C, and up to 6W when we thermoelectrically cooled the sample to −15°C (see Figure 2).5 In pulsed operation (200ns pulse length), we measured over 21W of on-time output power at room temperature. The optical quantum efficiency of the devices reaches very high values of 45% at room temperature and 55% at −15°C heat-sink temperature. We can achieve further power scaling of these 2.Xμm OPSDLs by distributing the heat load over more than one chip,6 i.e., using a double-chip cavity. This way, we attained more than 8W output power for a cavity using two 2.0μm gain elements at a heat-sink temperature of −5°C.7

Figure 2. Continuous-wave (CW) output power versus absorbed pump power of a 2.0μm OPSDL pumped at 980nm for different heat-sink temperatures.

The external resonator of OPSDLs is flexible, and so we were able to develop different resonator systems using the same basic chip for different applications. One development strand was a tunable single-mode OPSDL for the 2μm wavelength range (see Figure 3). Using a birefringent filter inside a V-shaped laser cavity allowed the laser to operate in single longitudinal mode (see Figure 4).8 We could tune the emission wavelength of the laser within a window of 120nm around the central lasing wavelength of 1.98μm by simply rotating the birefringent filter. For a single-mode laser, the output power was still very high, with a maximum value above 500mW at the central frequency, dropping to around 100mW at the edges of the tuning range (see Figure 5). We deduced the linewidth of the laser, measured with a special etalon, also known as a scanning Fabry-Perot Interferometer (FPI), to be less than 2.3MHz. Since this is the resolution limit of the FPI used in the present experiment, the true linewidth of the disk laser could be even smaller. We performed all measurements on a simple cavity setup with no active stabilization or feedback. The birefringent filter was the only wavelength-selective element.

Figure 3. Schematic of an OPSDL in a V-shaped cavity with a birefringent filter for wavelength selection and stabilization. VECSEL: Vertical external-cavity surface-emitting laser.

Figure 4. Single-mode spectra of the tunable laser setup for one specific emission wavelength of 2.00μm. a.u.: Arbitrary units.

Figure 5. Output power versus emission wavelength of the tunable OPSDL (black line) and the free-running laser without birefringent filter (red star).

To extend the wavelength range, we fabricated an OPSDL based on gallium antimonide (GaSb) that emits at 2.8μm.7Critical issues for the MBE growth of this semiconductor structure were the total layer thickness of nearly 12μm and the high indium content needed for the long-wavelength quantum wells. This required optimization of the growth conditions, together with a comprehensive optical analysis of the grown wafers, to ensure a close match between the design and the actual structure. We achieved up to 120mW output power in CW operation at a submount temperature of 20°C. Under pulsed excitation, we obtained more than 500mW peak output power (see Figure 6). These results for a 2.8μm emitting GaSb-based disk laser demonstrate the potential for the wavelength range of OPSDLs to be significantly expanded toward 3μm and beyond, while still obtaining acceptable device performance at room temperature.

Figure 6. Output-power characteristics of a 2.8μm OPSDL in CW and pulsed operation at 20°C. The inset shows the emission spectrum in pulsed mode, revealing multimode emission filtered by the etalon modes of the silicon carbide intracavity heat spreader.

We followed up the initial laboratory setup experiments by developing compact and rugged laser modules, such as a prototype hermetically sealed laser module, including pump laser, monitor photodiode, and red pilot laser (see Figure 7). All 2.Xμm OPSDL laser modules have proved their robustness and reliability in tests, with no re-adjustment required after either shipment or prolonged operation.

Figure 7. Hermetically sealed OPSDL laser system, including pump laser, monitor photodiode, and red pilot laser.

Semiconductor disk lasers have the advantage of the wavelength flexibility provided by the semiconductor gain medium. This contrasts with classical solid-state lasers covering this wavelength range (such as holmium- or erbium-doped lasers). We can readily realize a specific emission wavelength to optimize the laser for a specific application. Furthermore, we can substantially tune each laser chip around its central emission wavelength (see Figure 5). Compared with diode lasers, which also cover this wavelength range, the beam quality at high output powers of the OPSDL is far superior: it is capable of delivering diffraction-limited output with beam-quality parameter (M2) values in the range of 1–3, depending on whether the module is optimized for beam quality or output power. Thus, these flexible laser sources in the 1.9–2.8μm range are ideal both for seeding solid-state or fiber lasers and amplifiers, for example, and also for direct high-brightness applications like medical treatment or materials processing. We see no reason why these long-wavelength OPSDLs should not follow the commercial success of their 1μm counterparts. Our future research will include further optimizing the semiconductor structure and thermal management to improve the output power and wavelength coverage. We will also develop new laser-resonator configurations, including optical elements, to increase the functionality of application-specific laser modules.

This research was funded by the Sixth Framework Programme of the European Union within the project VERTIGO (Versatile Two Micron Light Source), number 034692.

Benno Rösener, Marcel Rattunde, Christian Manz, Joachim Wagner
Fraunhofer Institute for Applied Solid State Physics
Freiburg, Germany

Benno Rösener joined the Institute in 2006, where he is now a senior researcher. His current work includes research on semiconductor disk lasers as well as development of compact laser modules.

Marcel Rattunde is a research team leader for IR semiconductor lasers. His current work includes 2.Xμm diode lasers, laser modules, and semiconductor disk lasers. He was the coordinator of the European research project VERTIGO on mid-IR semiconductor disk lasers.

Christian Manz is an engineer working on molecular-beam epitaxy of III-arsenium antimonide (AsSb) semiconductor heterostructures. His work includes the growth of (aluminum gallium indium)(AsSb) VECSELs and quantum cascade lasers as well as optimization of the respective growth conditions.

Joachim Wagner is deputy director of the Institute and head of the Opto-electronic Modules Department. He is also professor at the Physics Department of the University of Freiburg. His current research interest is in III–V-semiconductor heterostructures and their application in optoelectronic devices both for the IR and visible/UV spectral ranges.

John-Mark Hopkins, David Burns
Institute of Photonics
Glasgow, UK

John-Mark Hopkins joined the Institute in 2001 and is now a senior researcher. His research interests currently include doped-dielectric and optically pumped semiconductor disk lasers in the mid-IR regime.

David Burns is an associate director and team leader for solid-state laser development. His research interests include optically pumped semiconductor lasers, high-power mode-locked solid-state lasers, stabilization techniques, and adaptive optics.

Karsten Scholle
LISA laser products OHG
Freiburg, Germany