Vertical External Cavity Surface Emitting Lasers E-Book
162,99 €
Mehr erfahren.
- Herausgeber: Wiley-VCH GmbH
- Kategorie: Wissenschaft und neue Technologien
- Sprache: Englisch
Vertical External Cavity Surface Emitting Lasers
Provides comprehensive coverage of the advancement of vertical-external-cavity surface-emitting lasers
Vertical-external-cavity surface-emitting lasers (VECSELs) emit coherent light from the infrared to the visible spectral range with high power output. Recent years have seen new device developments – such as the mode-locked integrated (MIXSEL) and the membrane external-cavity surface emitting laser (MECSEL) – expand the application of VECSELs to include laser cooling, spectroscopy, telecommunications, biophotonics, and laser-based displays and projectors.
In Vertical External Cavity Surface Emitting Lasers: VECSEL Technology and Applications, leading international research groups provide a comprehensive, fully up-to-date account of all fundamental and technological aspects of vertical external cavity surface emitting lasers. This unique book reviews the physics and technology of optically-pumped disk lasers and discusses the latest developments of VECSEL devices in different wavelength ranges. Topics include OP-VECSEL physics, continuous wave (CW) lasers, frequency doubling, carrier dynamics in SESAMs, and characterization of nonlinear lensing in VECSEL gain samples. This authoritative volume:
- Summarizes new concepts of DBR-free and MECSEL lasers for the first time
- Covers the mode-locking concept and its application
- Provides an overview of the emerging concept of self-mode locking
- Describes the development of next-generation OPS laser products
Vertical External Cavity Surface Emitting Lasers: VECSEL Technology and Applications is an invaluable resource for laser specialists, semiconductor physicists, optical industry professionals, spectroscopists, telecommunications engineers and industrial physicists.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 792
Veröffentlichungsjahr: 2021
Ähnliche
Table of Contents
Cover
Title Page
Copyright
Preface
Part I: Continuous wave VECSEL
1 History of Optically Pumped Semiconductor Lasers – VECSELs
1.1 Introduction
1.2 OPS‐VECSELs: Concept and History
1.3 Micracor
1.4 OPSL Development at Micracor: First Steps
1.5 OPS Development at Micracor: Pushing Forward
1.6 OPS Development at Micracor: Final Chapter
1.7 VECSELs beyond Micracor
References
2 VECSELs in the Wavelength Range 1.18–1.55 μm
2.1 Introduction
2.2 Overview of GaAs‐based Gain Mirror Technologies for Long‐wavelength Infrared VECSELs
2.3 Overview of InP‐based Gain Mirror Technologies for Long‐wavelength Infrared VECSELs
2.4 Conclusion
References
3 Single‐frequency and High Power Operation of 2–3 Micron VECSEL
3.1 Introduction
3.2 Semiconductor Lasers for the MIR Range
3.3 III‐Sb Material System
3.4 2–3 μm VECSEL Design
3.5 Mounting Technologies
3.6 Single‐frequency Operation (SFO) of 2–3 μm VECSEL
3.7 Conclusion
References
4 Highly Coherent Single‐Frequency Tunable VeCSELs: Concept, Technology, and Physical Study
4.1 Introduction: Lasers for Applications
4.2 The “Ideal” Laser
4.3 Toward Single‐Mode Operation
4.4 Toward High Coherence
4.5 The VeCSEL in the State of the Art
4.6 Highly Coherent, Tunable VeCSEL Design
4.7 Limits and Solutions
4.8 Highly Coherent, Tunable VeCSEL: Main Characteristics
4.9 Ultrahigh‐Purity Single‐mode Operation
4.10 Spatial Coherence
4.11 Time Domain Coherence and Noise
4.12 Conclusion
Acknowledgements
References
Notes
5 Terahertz Metasurface Quantum Cascade VECSELs
5.1 Introduction
5.2 Metasurface Design
5.3 QC‐VECSEL Model
5.4 THz QC‐VECSEL Performance: Power, Efficiency, and Beam Quality
5.5 Polarization Control in QC‐VECSELs
5.6 Conclusion
References
6 DBR‐free Optically Pumped Semiconductor Disk Lasers
6.1 Introduction
6.2 DBR‐free Semiconductor Disk Lasers
6.3 Device Fabrication
6.4 DBR‐free SDL Implementation
6.5 Novel Concepts
6.6 Conclusions
References
7 Optically Pumped Red‐Emitting AlGaInP‐VECSELs and the MECSEL Concept
7.1 Introduction
7.2 Direct Red‐Emitting AlGaInP‐VECSELs and Second‐Harmonic Generation
7.3 The Membrane External‐Cavity Surface‐Emitting Laser (MECSEL)
7.4 Conclusions
References
Part II: Mode‐Locked VECSEL
8 Recent Advances in Mode‐Locked Vertical‐External‐Cavity Surface‐Emitting Lasers
8.1 Introduction
8.2 Ultrafast Pulse Formation in a Surface‐Emitting Semiconductor Laser
8.3 Performance of Passively Mode‐Locked Semiconductor Lasers
8.4 Applications
8.5 Summary and Outlook
References
9 Ultrafast Nonequilibrium Carrier Dynamics in Semiconductor Laser Mode‐Locking
9.1 Introduction
9.2 Background Theory
9.3 Domain Setup/Modeling
9.4 Numerical Results
9.5 Outlook
References
Note
10 Mode‐Locked AlGaInP VECSEL for the Red and UV Spectral Range
10.1 Introduction
10.2 Epitaxial Layer Design of AlGaInP‐SESAM Structures
10.3 Temporal Response of AlGaInP SESAMs
10.4 Cavity Designs
10.5 Characterization Methods
10.6 Mode‐Locking Results
10.7 Second Harmonic Generation into the UV Spectral Range
10.8 Summary and Outlook
References
11 Colliding Pulse Mode‐locked VECSEL
11.1 Introduction
11.2 Principle of Colliding Pulse Modelocking
11.3 Requirements for Stable Colliding Pulse Modelocking
11.4 Design of an Ultrafast CPM VECSEL
11.5 Modelocking Results
11.6 Pulse Interactions in the Saturable Absorber
11.7 Summary and Outlook
Acknowledgments
References
12 Self‐Mode‐Locked Semiconductor Disk Lasers
12.1 Introduction
12.2 Mode‐Locking Techniques for Optically Pumped SDLs at a Glance
12.3 History of Saturable‐Absorber‐Free Pulsed VECSELs
12.4 Overview on SESAM‐Free Mode‐Locking Achievements
12.5 Investigations into the Mechanisms and Outlook
Acknowledgments
References
Index
End User License Agreement
List of Tables
Chapter 5
Table 5.1 Values of
and
extracted via simulation for three metasurface des...
Chapter 7
Table 7.1 From the data plotted in Figure 7.6 with the fit Equation 7.9 deriv...
Chapter 8
Table 8.1 Overview of femtosecond mode‐locking results in InGaAs‐based VECSE...
Table 8.2 Overview of fiber‐amplified ML‐VECSEL results.
Chapter 10
Table 10.1 Overview of reported mode‐locked AlGaInP VECSELs
Chapter 12
Table 12.1 Highlights of SESAM mode‐locking with QW‐SDLs
Table 12.2 Developments concerning SML VECSELs; reported results (extrema in ...
List of Illustrations
Chapter 1
Figure 1.1 (a) Semiconductor laser with radiating mirrors.(b) Optically pumped...
Figure 1.2 Aram Mooradian and the Micracor logo.
Figure 1.3 Multi‐bounce OPS laser Micracor promised in 1993 to make ultimately...
Figure 1.4 Band diagram of the proposed optically pumped surface‐emitting semi...
Figure 1.5 Single‐bounce OPS laser configuration in the April 1994 SBIR Phase ...
Figure 1.6 Excellent beam profile of the pulsed optically pumped 900 nm InGaAs...
Figure 1.7 Two figures from the Micracor second quarterly report to the sponso...
Figure 1.8 (a) Pulsed laser power up to 110 °C with weak temperature dependenc...
Figure 1.9 Photoluminescence image of the strained multi‐quantum‐well InGaAs/G...
Figure 1.10 (a) OPS CW output power in TEM01 and TEM00 (Gaussian beam) modes, ...
Figure 1.11 OPS CW output power in TEM31, TEM00 (Gaussian beam) modes, and pow...
Figure 1.12 OPS power coupled into single mode fiber, fourth generation wafer ...
Figure 1.13 OPS power in TEM11 mode, TEM00 fundamental circular Gaussian mode,...
Figure 1.14 Evolution of Micracor OPS laser power over time since the start of...
Chapter 2
Figure 2.1 The output characteristic of the 1185 nm VECSEL. The maximum output...
Figure 2.2 A representation of the bandgap and lattice constants for dilute ni...
Figure 2.3 An illustrative example of a type‐II “W”‐type QW. The electrons are...
Figure 2.4 Main approaches for integration of gain and reflector regions. (a) ...
Figure 2.5 The number of DBR pairs required to attain a reflectivity of >99.8%...
Figure 2.6 The output characteristics of a flip‐chip VECSEL with a mirror sect...
Figure 2.7 A generic illustration of wafer bonding.
Figure 2.8 Generic summary of direct wafer‐bonding methods.
Figure 2.9 Some of the initial high‐power demonstrations of 1.3–1.6 μm wafer‐f...
Figure 2.10 The output characteristics of a wafer‐fused flip‐chip VECSEL gener...
Figure 2.11 The gain mirror structure used for the first flip‐chip VECSEL at ∼...
Figure 2.12 A summary of output power levels achieved using flip‐chip VECSELs ...
Figure 2.13 Wafer cleaning and activation via sputtering to generate dangling ...
Figure 2.14 Generic summary of the intermediate layers that have been used in ...
Figure 2.15 Wafer bonding of long‐wavelength VECSELs via intermediate SiO
2
lay...
Figure 2.16 Wafer bonding of long‐wavelength VECSELs via intermediate MPTMS mo...
Figure 2.17 The output characteristic of an MPTMS‐bonded VECSEL emitting at 1....
Figure 2.18 The cavity configuration of a pulsed VECSEL emitting at 1.36 μm. T...
Chapter 3
Figure 3.1 Overview of various semiconductor‐based laser sources for the MIR r...
Figure 3.2 Relative position of the conduction and heavy hole band and the fir...
Figure 3.3 (a) Band edge profile of 2.8 μm GaInAsSb QWs embedded between diffe...
Figure 3.4 (a) Standard VECSEL design for a barrier‐pumped active region. With...
Figure 3.5 (a) In‐well pumped VECSEL design for an increased quantum efficienc...
Figure 3.6 (a) LQD barrier pumping, using GaSb‐barriers and 1470 nm pump diode...
Figure 3.7 (a) Output power characteristics of a 2.5 μm LQD‐VECSEL compared to...
Figure 3.8 Three heat removal approaches and their dominant heat removal pathw...
Figure 3.9 Simulated temperature increase relative to the heatsink temperature...
Figure 3.10 Emission spectra of an ICH sample for different heatsink temperatu...
Figure 3.11 Output power vs. emission wavelength of VECSEL chips with ICH made...
Figure 3.12 (a) Output power characteristics of a thin device for different pu...
Figure 3.13 Comparison of output power characteristics for two heat management...
Figure 3.14 Noise sources and their typical time scales adversely affecting th...
Figure 3.15 If the pump spot is not a perfect top‐hat profile, the effective a...
Figure 3.16 Typical V‐shaped cavity setup for a tunable VECSEL (a) and the cor...
Figure 3.17 (a) FTIR measurement of the emission spectra, revealing an SMSR of...
Figure 3.18 Coarse wavelength tuning by rotating the BRF. A 2.0 μm standard 98...
Figure 3.19 (a) Measurement setup for a heterodyne beat note. Two identical VE...
Figure 3.20 Long‐term stability of a free‐running VECSEL module at 2.0 μm: the...
Figure 3.21 Setups for active wavelength stabilization: (a) Side‐of‐fringe set...
Figure 3.22 VECSEL linewidth obtained by beat‐note spectra vs. the integration...
Figure 3.23 (a) Emission spectra of a VECSEL with wedged and AR‐coated heatspr...
Figure 3.24 Schematic of the μC‐VECSEL setup. An HR coating deposited on the t...
Figure 3.25 (a) Output power of the μC‐VECSEL vs. absorbed pump power at room ...
Figure 3.26 (a) Wavelength tuning with a temperature change of 120 °C. The ins...
Figure 3.27 (a) Short cavity setup for single‐frequency 2.3 μm VECSEL and a la...
Figure 3.28 Measured CH
4
absorption spectra with the laser setup shown in Figu...
Figure 3.29 Schematic setup of the long V‐shaped cavity used for wavelength tu...
Figure 3.30 (a) Emission wavelength measured with a 20 MHz resolution wavemete...
Figure 3.31 (a) Emission wavelength vs. tilt angle of the etalon. (b) Detailed...
Figure 3.32 Emission wavelength in a 60 hour time span for the free running si...
Chapter 4
Figure 4.1 Basic design of a VeCSEL. The longitudinal (
) axis and the transve...
Figure 4.2 (a) Transverse intensity distribution of some possible transverse m...
Figure 4.3 (a) Various degrees of time coherence, depending on multimode opera...
Figure 4.4 Relation of dispersion (Eq. 4.1) for optical modes. (a) Longitudina...
Figure 4.5 Modes competition in a laser [18, 19]. (a) Frequency difference ind...
Figure 4.6 Spatial Hole Burning in lasers. (a) Gain saturation as a function o...
Figure 4.7 (a) Ability of a cavity to filter a noisy seeding as a function of ...
Figure 4.8 State of the art of continuous‐wave lasers and VeCSELs. (a) Classif...
Figure 4.9 (a) Main structure of a VeCSEL. (b) Reflectivity of a DBR. (c) Ener...
Figure 4.10 Micro‐cavity‐induced thermal rollover.
Figure 4.11 Thermal Management. (a) Usual techniques. (b, c) Impact for a pump...
Figure 4.12 Overview of the VeCSEL characteristics. (a–c) Display the output p...
Figure 4.13 (a) Setup for RF measurement of TMSR. (b) Example of result for a
Figure 4.14 Spatial coherence of the VeCSEL beam. (a) Propagation of a nonidea...
Figure 4.15 (a) Macroscopic intensity noise related to microscopic events. (b)...
Figure 4.16 (a) Example of noise spectrum, exhibiting both RIN and spectral de...
Figure 4.17 (a) Typical frequency‐noise setup using a Fabry–Perot interferomet...
Figure 4.18 (a) Theoretical frequency noise spectrum computed from the model d...
Chapter 5
Figure 5.1 (a) Schematic figure of metasurface QC‐VECSEL configuration. (b) Cl...
Figure 5.2 Scanning electron microscope images at various magnification of a Q...
Figure 5.3 (a) Simulated reflectance spectra for four metasurfaces with ridge ...
Figure 5.4 (a) Schematic of model VECSEL cavity. (b) Schematic three‐level sys...
Figure 5.5 (a) Diagram of square metasurface with biased circular area of radi...
Figure 5.6 Calculated output power per period
vs. output coupler reflectance...
Figure 5.7 (a) Maximum power per period
change with
for different external...
Figure 5.8 (a) Measured lasing spectra for some demonstrated QC‐VECSELs based ...
Figure 5.9 (a) Measured 77 K pulsed
–
characteristics for QC‐VECSELs built w...
Figure 5.10 (a) Simulated reflectance and reflection phase from a uniform meta...
Figure 5.11 (a) Pulsed
–
–
curves for the
mm focusing metasurface QC‐VECSE...
Figure 5.12 (a) Measured 1D beam pattern cuts in
and
directions for a QC‐V...
Figure 5.13 (a) Measured 77 K pulsed
–
–
for an intra‐cryostat cavity QC‐VEC...
Figure 5.14 Measured 77 K cw
–
–
for an intra‐cryostat cavity QC‐VECSEL base...
Figure 5.15 (a) SEM image of polarimetric metasurface covering an area of 2 × ...
Figure 5.16 (a) Measured total power through the polarizer versus the polarize...
Chapter 6
Figure 6.1 Comparison of sample, heatspreader, and heatsink geometries for (a)...
Figure 6.2 Schematic diagram of a typical linear DBR‐free SDL cavity.
Figure 6.3 Maximum active region temperature rise with incident pump power for...
Figure 6.4 Position dependent (
z
) integrated gain for a periodic MQW (
N
= 12) ...
Figure 6.5 Integrated gain factors for (a) VECSEL and (b) DBR‐free geometry in...
Figure 6.6 Nomarski micrographs of bonding examples on single‐crystalline chem...
Figure 6.7 Lifetime characterization of an active region before (a) and after ...
Figure 6.8 (a) Laser performance comparison between gain structures with singl...
Figure 6.9 Laser performance for a gain chip bonded onto a single‐crystalline ...
Figure 6.10 Active region uniformity check at different pump powers. With 5.8 ...
Figure 6.11 Laser tuning spectra of DBR‐free SDLs with active regions designed...
Figure 6.12 RF spectra for DBR‐free geometry with near‐symmetric (a) and perfe...
Figure 6.13 Wafer‐scale manufacturing process for a dual‐SiC‐heatspreader (SiC...
Figure 6.14 Schematic of the TIR‐based monolithic ring DBR‐free SDLs. (a) Equi...
Figure 6.15 (a) Schematic of the total internal‐reflection‐based V‐shaped cavi...
Figure 6.16 The GEMM structure as an active mirror in a VECSEL.
Chapter 7
Figure 7.1 Close‐up view of a red‐emitting AlGaInP‐VECSEL in a V‐shaped cavity...
Figure 7.2 Epitaxy map of the AlGaInP material system.
Figure 7.3 (a–c) Schematic of the gallium to indium composition influencing th...
Figure 7.4 Index of refraction (black curve) and electric field intensity (red...
Figure 7.5 (a) Laser spectra of the different epitaxial designs. (b) Power tra...
Figure 7.6 Differential efficiency
plotted versus outcoupler reflectivity
...
Figure 7.7 Gain schematics, gain (arbitrary units) is plotted versus wavelengt...
Figure 7.8 (a) Optical output power versus (top) incident and (bottom) absorbe...
Figure 7.9 (a) Mirror and laser beam schematic of a linear VECSEL with multi‐p...
Figure 7.10 (a) Laser performance with 0.5% of output coupling and a heat sink...
Figure 7.11 (a) Frequency‐doubled UV output power (330 nm) plotted over pump p...
Figure 7.12 Photograph of the operating semiconductor membrane external‐cavity...
Figure 7.13 (a) Electric field intensity and index of refraction in the RPG st...
Figure 7.14 SEM picture of the gain membrane, taken from a free‐standing piece...
Figure 7.15 (a) Microscope picture of the semiconductor gain membrane squeezed...
Figure 7.16 Schematic drawing of the MECSEL setup: A linear resonator with a b...
Figure 7.17 High dynamic range photograph of the gain membrane holder, also sh...
Figure 7.18 (a) Output power plotted over incident pump power of the MECSEL an...
Figure 7.19 (a) A typical beam profile of the MECSEL demonstrating a
mode ()...
Figure 7.20 Beam propagation plot (beam radii versus distance) of the Coherent...
Figure 7.21 (a) A typical spectrum of the free‐running MECSEL (upper graph) at...
Chapter 8
Figure 8.1 Coherent optical pulse train. Left: time‐varying E‐field. The carri...
Figure 8.2 Schematic comparison of (a) ML‐VECSEL and (b) MIXSEL.
Figure 8.3 Layer design of a broadband gain chip with plot of
as a function ...
Figure 8.4 Dispersion and reflectivity calculations for the structure of Figur...
Figure 8.5 Repetition rate tuning ranges (GHz) for ML‐VECSELs (dark gray [74, ...
Figure 8.6 Pulse durations of selected mode‐locked semiconductor disk lasers o...
Chapter 9
Figure 9.1 A schematic of a linear VECSEL cavity with a reflecting mirror on t...
Figure 9.2 A figure showing the effect of adding constant dispersion of (a) ev...
Figure 9.3 A comparison of the reflection (a), and three orders of dispersion ...
Figure 9.4 A schematic representation of four cavities: (a) V‐cavity, (b) Z‐ca...
Figure 9.5 The solid (dashed) line in (a) shows an example where the gain is v...
Figure 9.6 (a) The calculated characteristic carrier scattering rate from a QW...
Figure 9.7 A schematic of the material layers in a linear VECSEL cavity with d...
Figure 9.8 (a) A schematic of an MQW structure with 10 QWs placed inside a sin...
Figure 9.9 An example that shows the emergence of a pulse train in the cavity ...
Figure 9.10 A diagram showing the development from noise to a mode‐locked puls...
Figure 9.11 Inversion of the lowest electron and hole subband (left) and linea...
Figure 9.12 (a, b) Time evolution of the inversion in a QW that is excited wit...
Figure 9.13 Time‐dependent (negative) index change of mode‐locked VECSELs [28]...
Figure 9.14 Time evolution of the sum of electron and hole occupations in a SE...
Figure 9.15 Saturation fluence as function of the excitation wavelength for th...
Figure 9.16 Time evolution of the index change in a SESAM (left) and GSAM (rig...
Figure 9.17 The initial loss and gain in the cavity (a) and the GDD in the gai...
Figure 9.18 Here we see the pulse formation process from spontaneous emission ...
Figure 9.19 The refractive index change as computed with the Kramers–Kronig tr...
Figure 9.20 An overview of the three families of mode‐locked pulses: the singl...
Figure 9.21 An overview of the dynamics for a single stable pulse (a, c, e) an...
Figure 9.22 Here we see snapshots of the inversion during the buildup of the
p
...
Figure 9.23 Overview of the mode‐locked pulse FWHM (a) and output fluence (b) ...
Figure 9.24 (a) A snapshot of the time‐dependent inversion in the QWs that res...
Figure 9.25 Here we see the mode‐locked pulse peak intensity (a) and FWHM (b) ...
Figure 9.26 (a) The spectral modulation in Eq. (9.5) that comes from placing t...
Chapter 10
Figure 10.1 (a) Near‐resonant SESAM design with two GaInP QWs near the semicon...
Figure 10.2 (a) Near‐antiresonant SESAM design with one InP QD layer. The fiel...
Figure 10.3 (a) Pump‐probe result for a SESAM with two QWs close to the surfac...
Figure 10.4 Cavity configurations used for mode locking of AlGaInP VECSELs wit...
Figure 10.5 High output power results from a QW mode‐locked AlGaInP VECSEL. (a...
Figure 10.6 Results from a QW mode‐locked AlGaInP VECSEL with femtosecond puls...
Figure 10.7 Autocorrelation histogram recorded with a HBT type setup using APD...
Figure 10.8 (a) Power transfer curve of the QD VECSEL with QD absorber structu...
Figure 10.9 Close‐up view of the V‐shaped mode locking setup with intracavity ...
Figure 10.10 Optical spectrum of the pulsed laser showing the fundamental and ...
Chapter 11
Figure 11.1 Cavity geometries and circulation of the pulses in a colliding pul...
Figure 11.2 Illustration of the gain recovery dynamic after the passage of two...
Figure 11.3 Example of polarization arrangements for two noncollinear beams co...
Figure 11.4 Schematic layout of a colliding pulse mode‐locked VECSEL in a fold...
Figure 11.5 Calculation of the cavity mode width on the SESAM and VECSEL struc...
Figure 11.6 Beam width evolution along the cavity and illustration of the gain...
Figure 11.7 Left: Schematic layout of a VECSEL structure with nonuniform QW di...
Figure 11.8 Reflectivity spectra of the VECSEL structure with and without the ...
Figure 11.9 Influence of the incidence angle on the GDD spectrum of a typical ...
Figure 11.10 Influence of the type of coating on the gain and GDD spectrum. We...
Figure 11.11 Schematic layout of a SESAM structure incorporating an InGaAs QW ...
Figure 11.12 Measured group delay dispersion spectrum of the SESAM and VECSEL ...
Figure 11.13 Schematic layout of the experimental CPM VECSEL device.
Figure 11.14 Output power from one output beam versus incident pump power at d...
Figure 11.15 Measurement setup of the noncollinear sum frequency generation cr...
Figure 11.16 Measure of the noncollinear cross correlation of the two output b...
Figure 11.17 Measure of the microwave spectrum. In single pulse operation (top...
Figure 11.18 Measured noncollinear SHG autocorrelation of the single pulse ope...
Figure 11.19 Measured optical spectrum of the output beam consisting of a sing...
Figure 11.20 Microwave spectrum of the laser output, with a RBW of 100 kHz and...
Figure 11.21 Left: Measurement of the noncollinear autocorrelation in multipul...
Figure 11.22 Field intensity distribution of two coherent Gaussian beams colli...
Figure 11.23 (a) Microscopically calculated QW absorption and corresponding SE...
Figure 11.24 Experimentally measured and simulated reflectivity change in the ...
Figure 11.25 Carrier density in the SESAM QW generated by the colliding of two...
Figure 11.26 Normalized carrier density distribution in the SESAM QW along the...
Figure 11.27 Simulated carrier density in the SESAM at the center of the beam ...
Figure 11.28 Simulated absorption losses from a single pulse of 90 mW and a du...
Figure 11.29 Simulated pulse intensity before and after reflection on the SESA...
Figure 11.30 Absorption losses from the SESAM as a function of the intracavity...
Figure 11.31 Absorption losses from the SESAM as a function of the intracavity...
Figure 11.32 Recorded intensities of the two counter‐propagating output beams:...
Chapter 12
Figure 12.1 (a) Overview on the annual publication output concerning ML VECSEL...
Figure 12.2 Different mode‐locking techniques for VECSELs. (a) SESAM ML device...
Figure 12.3 Pulse characterization as shown for the first spontaneously mode‐l...
Figure 12.4 Schematic representation of the Kerr lensing effect for high‐inten...
Figure 12.5 Schematic representation of an SML‐VECSEL's cavity configuration (...
Figure 12.6 Schematic representation of (a) hard‐ and (b) soft‐aperture Kerr l...
Figure 12.7 Photo of the actual setup used for external SHG, using the origina...
Figure 12.8 (a)–(c) Number of pulses in the cavity (sketch) and corresponding ...
Figure 12.9 Number of pulses in the cavity (orange balls, left axis) and the a...
Figure 12.10 (a) Schematic drawing of the optically pumped SML QD‐VECSEL setup...
Figure 12.11 (a) Linear‐cavity geometry as used for an SML QD VECSEL. (b) Z‐ca...
Figure 12.12 Photo of the SML AlGaInP VECSEL with characterization of the puls...
Figure 12.13 Red SML VECSEL results. Source: Adapted with permission. [51] Cop...
Figure 12.14 Pulse durations of mode‐locked optically pumped VECSELs.
Figure 12.15 (a) Schematics of the laser cavity used for SESAM‐free mode‐locki...
Figure 12.16 Wavelength and angle‐dependent nonlinear absorption (a) and nonli...
Guide
Cover
Table of Contents
Title Page
Copyright
Preface
Begin Reading
Index
WILEY END USER LICENSE AGREEMENT
Pages
iii
iv
xiii
xiv
1
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
109
110
111
112
113
114
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
387
388
389
390
391
392
393
394
Vertical External Cavity Surface Emitting Lasers
VECSEL Technology and Applications
Edited byMichael JetterPeter Michler
Editors
Dr. Michael JetterUniversity of StuttgartInstitut für HalbleiteroptikAllmandring 370569 StuttgartGermany
Prof. Dr. Peter MichlerUniversity of StuttgartInstitut für HalbleiteroptikAllmandring 370569 StuttgartGermany
Cover: © fotojog/Getty Images
All books published by WILEY‐VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.
Library of Congress Card No.: applied for
British Library Cataloguing‐in‐Publication Data A catalogue record for this book is available from the British Library.
Bibliographic information published bythe Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.
© 2022 WILEY‐VCH GmbH, Boschstr. 12, 69469 Weinheim, Germany
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.
Print ISBN: 978‐3‐527‐41362‐1ePDF ISBN: 978‐3‐527‐80794‐9ePub ISBN: 978‐3‐527‐80797‐0oBook ISBN: 978‐3‐527‐80795‐6
Preface
The first reports in 1997 of vertical external‐cavity surface‐emitting laser (VECSEL), also known as optically pumped semiconductor disk laser (SDL), attracted a lot of attention in the research and laser community. This relatively new type of laser combines the versatility in wavelength from the semiconductor‐active region with superior emission properties such as multi‐watt output powers in cylindrically symmetric TEM00 mode and nearly diffraction‐limited beam quality. The extended cavities allow an efficient intracavity frequency conversion and enable, by incorporating semiconductor saturable absorber mirrors (SESAM), short‐pulse mode locking operation.
Since the first summarizing book of the achievements in this field in 2010 from Oleg G. Okhotnikov, a tremendous progress in demonstrating the expected VECSEL properties took place. The direct emitting wavelength of these lasers expands meanwhile from the blue to the mid‐infrared range. With frequency doubling, even the UV spectral range is reached with high optical powers. SESAM mode‐locked lasers demonstrated pulse widths below 100 fs and pulse repetition rates over 100 GHz. However, not only the predicted features were confirmed but also new varieties of VECSELs were invented in the meantime. The most prominent ones are the mode‐locked integrated external‐cavity surface‐emitting laser (MIXSEL), combining the SESAM and gain mirror in one semiconductor stack, and the membrane external‐cavity surface‐emitting laser (MECSEL), basically a heat spreader–semiconductor sandwich in an external cavity, enabling wavelength ranges which were not reachable due to semiconductor growth restrictions. Next to these nice efforts and lively research, these devices as well took the step to commercialization. Several companies offer nowadays laser equipment including SDL either in cw or mode‐locked operation.
It is our pleasure and honor to present with this book some of the most recent developments in VECSEL research. It is written by internationally renowned experts who actively advance this field of laser research. The book is structured into two sections, a continuous wave and a pulsed laser part. It includes the development of the VECSELs, recent advances, technology aspects, and some applications of VECSEL. It can be useful for engineers and scientists working in this field as well as for graduate students interested in the technology of these laser devices.
We like to thank all the authors for their valuable contributions to this book and their patience. Furthermore, we like to thank the editorial staff of Wiley‐VCH GmbH for keeping this book on track for publication.
Stuttgart, March 2021
Michael Jetter
Peter Michler
Part IContinuous wave VECSEL
1History of Optically Pumped Semiconductor Lasers – VECSELs
Mark E. Kuznetsov
Axsun/Excelitas Technologies, Billerica, MA, USA
1.1 Introduction
Optically pumped semiconductor lasers (OPSLs) are an old concept that originated in the early days of semiconductor lasers in the 1960s, and that remained a scientific curiosity until the mid‐1990s, when its potential capabilities for high power with excellent beam quality were first fully demonstrated, spurring subsequent rapid development of the science and technology of these versatile lasers. Distinguishing features of OPSLs today are light emission normal to the plane of the semiconductor chip, laser cavity external to the chip to stabilize fundamental transverse laser mode and to enable insertion of intracavity functional optical elements, and the use of optical pumping for efficient and high output power operation. A wide range of applications is enabled by additional remarkable properties of this laser family, such as wavelength operation from ultraviolet (UV) to mid‐infrared (IR) and even terahertz range, and passively modelocked operation with output pulses shorter than 100 fs. These lasers are also widely known by two other names, descriptive of their geometry: vertical external‐cavity surface‐emitting lasers (VECSELs) and semiconductor disk lasers (SDLs). Alternatively, VECSELs can also be electrically pumped, but achievable laser output powers are then typically much lower than for the optically pumped version.
OPSL development in the 1990s was spearheaded by Aram Mooradian in Micracor, a small start‐up company that spun out technology from Aram's group in the MIT Lincoln Laboratory. I worked with Aram in Micracor to carry out this development. In 2011, the first annual VECSEL conference was held at SPIE Photonics West, with the VECSELs‐XI conference scheduled for 2022. The first book about these lasers, Semiconductor Disk Lasers: Physics and Technology[1], was published in 2010; it was edited by Oleg Okhotnikov from the Tampere University of Technology in Finland and described the then state of the art in chapters contributed by researchers from around the world. Since the publication of this book, science, technology, and applications of VECSELs have made a significant step forward, and hence the present book to bring VECSELs overview up to date. This chapter describes the history of OPSLs, the people that took part in their development, and it's also a personal story of the OPSL development by our team at Micracor. Sadly, both Aram Mooradian and Oleg Okhotnikov, who have contributed so much to the early development of these lasers, have passed away since the publication of the first book. This historical chapter is dedicated to their memory.
1.2 OPS‐VECSELs: Concept and History
The first laser invented in 1960 was a flashlamp‐pumped solid‐state ruby laser. Other laser gain media and pumping schemes soon followed, and in 1962 a semiconductor diode laser pumped by current injection in a semiconductor p‐n junction was demonstrated [2–6]. Semiconductors offered the possibility of operating at different wavelengths, depending on the material composition – already in 1962, together with the near‐IR operation of binary GaAs lasers, ternary alloy GaAsP semiconductor diode laser in the visible was also demonstrated [5]. Electron‐hole pairs for laser excitation in semiconductors can be created by various means. Current injection pumping of diode lasers, although requiring more complex device fabrication, is appealing for its simplicity of use and the possibility of direct laser output modulation by current modulation. Yet other schemes for semiconductor laser pumping were also investigated, including optical and electron beam pumping. Semiconductor diode lasers have seen tremendous development from the 1960s to the 1990s, driven primarily by applications in optical fiber communication, CD and DVD optical disk readout, and pumping of solid‐state and fiber lasers and amplifiers. One challenge had remained, however. While diode lasers could produce very large, from watts to hundreds of watts, powers, this power was produced with poor beam quality: in highly transverse multimode, high aspect ratio output beams, or from large arrays of lasers. Single transverse mode output, and especially circular Gaussian output beam, could be produced at only much smaller power levels of at most a few hundred milliwatts. Optical pumping had remained an essentially experimental tool to demonstrate lasing capability of the semiconductor gain medium or laser structure, on the way to the ultimately useful diode‐current‐injection electrical pumping.
Why would semiconductor lasers with high multiwatt output power and beam quality be useful and important? The alternative high‐power laser technologies, e.g. solid state, gas, and atomic vapor lasers, rely on discrete atomic transitions and thus are restricted to discrete unique operating wavelengths. Semiconductor lasers, via material composition and bandgap‐engineered quantum‐confined structures, can produce a very wide range of operating wavelengths, from UV to mid‐IR, both directly and using nonlinear optical, including intracavity, conversion. This allows, by design, an essentially continuous coverage of this spectrum and even dual wavelength laser operation. High‐power good beam quality semiconductor lasers can offer unique operating parameters not accessible by other types of lasers. If you add to this femtosecond pulse capability with high peak powers, potentially compact size and low fabrication cost, this makes such semiconductor lasers useful, and sometimes possibly unique, for a wide range of applications.
Early semiconductor lasers were edge emitting and emitted light in the plane of the wafer, so that enough gain could be accumulated over the length of the device. Such edge‐emitting geometry, both gain and index guided, limits transverse profile beam quality for high powers. Vertical cavity surface‐emitting lasers (VCSELs) were first described by Kenichi Iga at the Tokyo Institute of Technology in 1979 [7, 8] and further developed to efficient low threshold operation by Jack Jewell at Bell Laboratories, Holmdel, in 1989 [9–11]. VCSELs surface‐emitting geometry, with light emitted normal to the plane of the wafer, because of low gain of the short gain region, requires very high reflectivity semiconductor or dielectric multilayer mirrors. Such vertical cavity geometry allows single transverse mode circular Gaussian beam output, but typically only for milliwatt level powers, limited by the difficulty of heat dissipation from the small mode area of a few microns in diameter.
Optical pumping of semiconductor lasers has a long history; it has been used for various purposes, such as characterization of novel semiconductor laser materials and structures, generation of higher output powers, or short pulse generation. As early as 1965, an OPSL has been demonstrated by Robert Phelan and Robert Rediker at MIT Lincoln Laboratory [12], where an edge‐emitting InSb laser was pumped by an edge‐emitting GaAs diode laser. Remarkably, both concepts, optical pumping of semiconductors and the use of diode lasers as pumps, are already in use this early in the history of lasers. In 1966, Nikolay Basov's group at the Lebedev Physical Institute in Moscow introduced the concept of a “radiating mirror” [13], Figure 1.1a: a thin semiconductor gain layer placed on top of a mirror and a heatsink, with an external output coupling mirror completing the laser cavity. Both optical and electron beam pumping were envisaged and demonstrated as the possible excitation sources. Large lateral extent of the gain medium, greater than its thickness, would ensure effective heat removal and thus the possibility of high output power. This is essentially the concept of a “disk laser” geometry, which would prove so effective many years later in both solid‐state [14, 15] and semiconductor [1] laser configurations. Basov's 1966 “radiating mirror” concept, Figure 1.1a, is remarkably similar to the 1996 Micracor OPSL configuration, Figure 1.1b. In his paper, Basov reported operation of a “radiating mirror” laser with optical pumping of CdSe using two‐photon absorption of a Q‐switched Nd‐doped glass pump laser. So as a concept, SDL had been already introduced and demonstrated in 1966; however, its full potential was yet to be explored and developed.
In the late 1960s, Nick Holonyak's group at the University of Illinois, Urbana, reported several studies of optically pumped CdS, GaAs, and GaAsP semiconductor thin platelet lasers [16], some using GaAsP diode laser pumping, and considered both edge‐ and surface‐emitting laser geometry. Transparent sapphire heatsinking windows had been used to remove heat and help improve power performance of the devices, foreshadowing the future use of such transparent heatspreaders. Optical pumping here is mainly used to explore lasing in different semiconductor materials. In a 1973 publication from Aram Mooradian's MIT Lincoln Lab group, Stephen Chinn demonstrated pulsed operation of optically pumped edge‐emitting bulk GaAs semiconductor lasers [17], with the goal of efficient high power generation. Later in 1981, Julian Stone, Jay Wiesenfeld, Andrew Dentai, and coworkers from the Bell Laboratories Crawford Hill Lab, used surface‐emitting thin‐film ultrashort cavity InGaAsP lasers [18] to generate gain‐switched picosecond pulses in the 0.83–1.59 mm wavelength range using dye laser pumping. Wavelength versatility of semiconductor materials is explored here, with short pulse generation as the primary goal. Using an external optical cavity for pulse repetition rate and transverse mode control, optically pumped mode locking was demonstrated with a CdS thin platelet laser by Charles Roxlo and Michael Salour at MIT in 1981 [19]. In 1991, Mooradian's group observed high peak power in an external‐cavity GaAs platelet laser pumped by a Ti:sapphire laser [20]. Note that the above OPSLs largely used bulk semiconductor material without internal heterostructures. As a result, the performance of these lasers was limited.
Figure 1.1 (a) Semiconductor laser with radiating mirrors.
Source: Reprinted, with permission, from Basov et al. [13], © 1966 IEEE.
(b) Optically pumped semiconductor vertical external‐cavity surface‐emitting laser (OPS‐VECSEL), Micracor, 1996.
When semiconductor heterostructures and quantum wells had been developed, they also had been used for OPSL experiments, e.g. in Holonyak's group [21]. When the first efficient VCSEL semiconductor lasers were demonstrated by Jack Jewell in 1989, early experiments used optical pumping of bulk GaAs [9] and later InGaAs and GaAs quantum well laser structures [10, 11]. On‐wafer high reflectivity semiconductor multilayer distributed Bragg reflector (DBR) mirrors were used for these vertical cavity lasers. At this point, the full power of the semiconductor bandstructure engineering was beginning to be applied to OPSLs. But again, optical pumping was used only as an initial characterization of laser structures on the way to eventual diode current‐injection pumping [11].
In the 1990s, a series of papers explored optical pumping of semiconductor lasers with the goal of obtaining high power with good beam quality, as well as short pulse generation. A wide variety of laser configurations were explored, both edge and surface emitting, with a range of pump laser options: diode, solid‐state, and gas lasers. Using diode laser pumping, low‐power 10 mW continuous wave (CW) operation was demonstrated with GaAs VCSEL by Steve Brueck's group at the University of New Mexico [22]; in external cavity, however, such lasers emitted only 20 μW [23, 24]. External resonator was also used with an electrically pumped VCSEL by a group from the University of California, Berkeley, in an attempt to increase its single‐transverse‐mode output power; however, only 2.4 mW was achieved in CW operation [25]. A diode‐laser‐pumped surface‐emitting optical amplifier was demonstrated at 1.5 μm using InGaAs–InGaAlAs multi‐quantum‐well structures by Shojiro Kawakami's group at Tohoku University [26]. Using 77 K low temperature operation and a Nd:YAG pump laser, 190 mW continuous output power was obtained from an external‐cavity InGaAs–InP surface‐emitting laser by Wenbin Jiang in the Yoshihisa Yamamoto's group at the NTT Basic Research Laboratories [27]. The same group used a similar configuration to demonstrate an external‐cavity GaAs VCSEL at 77 K with CW output power of 700 mW using a 1.8 W krypton–ion pump laser [28]. Modelocked femtosecond pulse operation was demonstrated with a periodic gain vertical cavity laser in an external cavity by Wenbin Jiang in the John Bowers group at the University of California, Santa Barbara [29]; with synchronous pumping by a modelocked Ti:sapphire laser, 6 mW average output power was obtained. Specially designed edge‐emitting InGaAs–GaAs laser structures were used with diode laser pumping to generate as much as 4 W average power by Han Le at MIT Lincoln Laboratory [30, 31]; however, the beams were strongly elongated with aspect ratios between 10 and 50 to 1. To summarize the state of the art by the mid‐1990s, semiconductor lasers could emit watt‐class power only with poor beam quality from edge‐emitting structures; vertical cavity lasers operated with good beams but only with milliwatt class output, external‐cavity operation of surface‐emitting lasers hadn't produced particularly high powers, and optical pumping remained just a tool for scientific exploration of novel laser configurations.
1.3 Micracor
Micracor (1992–1997) was a small company started by Aram Mooradian, see Figure 1.2, in Acton, a suburb of Boston, to commercialize several technologies from the Quantum Electronics Group at MIT Lincoln Laboratory, where Aram was the group leader for more than 20 years. Micracor's core technologies and nascent products were diode‐pumped solid‐state microchip lasers, work led by Kevin Wall, and tunable external cavity diode lasers, work led by Ken German. OPSLs were a concept explored previously in Aram's group at MIT Lincoln Lab. Micracor made a quixotic attempt to take this, at the time vague, concept, demonstrate it, and develop and commercialize such devices. The key initial target application was 980 nm pumping of Er‐doped fiber amplifiers. Micracor's core technologies weren't very successful commercially, while OPSL development after three and a half years showed tremendous promise. But investors eventually lost patience, and at the end of 1996 the company was shut down. Luis Spinelli from Coherent Inc. recognized the potential of OPSL technology and drove the purchase of Micracor's assets by Coherent. At this point our group at Micracor published two papers on OPS‐VECSELs [32, 33], where the term VECSEL was first introduced. The results of Micracor's more than three‐year effort on OPSLs were finally made public. These publications triggered subsequent development and exploration of VECSELs by the scientific community around the world. VECSEL technology was successfully commercialized by Coherent for applications as diverse as entertainment, forensics, life sciences, and medical. If it weren't for OPSLs, Micracor would be forgotten today. A recent Google search on Micracor yielded a puzzled response – “Did you mean: microcar?”
Figure 1.2 Aram Mooradian and the Micracor logo.
How did I come to Micracor and what was my role there in the development of OPSLs? After Micracor was founded, in 1993 Aram got a Small Business Innovation Research (SBIR) grant from the US Department of Defense to develop high‐power OPSLs. However, there was nobody at the company who could actually carry out this work. Phase I money was being spent, but no progress was made. I was hired in August of 1993 to carry the development of what would become OPSLs. Micracor was funded by Rothschild Ventures, and I remember visiting their offices in New York, with Rothschild family portraits on the walls. I came to Micracor after graduate school at MIT with Hermann Haus and Erich Ippen and seven years at Crawford Hill Lab of Bell Laboratories in the department of Ivan Kaminow. Looking back, I was well prepared and had the right background and experience to embark on the risky and challenging development of OPSLs. At Bell Labs, I had worked on electronically tunable quantum well diode lasers, I had both theoretical and experimental experience with semiconductor lasers, and my work has involved extensive modeling and design of edge‐emitting semiconductor laser structures, semiconductor device processing, laser fabrication and characterization, as well as device performance analysis. I had also been exposed to earlier work on OPSLs: at MIT, Michael Salour's lab was next door and I had attended multiple talks from the group; at Bell Labs, Julian Stone, Jay Wiesenfeld, and Jack Jewell were colleagues whose work I closely followed. Micracor was a small company of about 20 people working on a variety of projects; to develop OPSLs, I had to rely on myself to get the job done – no grand team to attack the problem; we had limited resources, equipment, and money. Compare this to Novalux, founded by Aram Mooradian several years after closing of Micracor, where $193 million (!) was spent developing electrically pumped VECSELs. Several people at Micracor played crucial roles in the OPSL development: Bob Sutherland (thin semiconductor wafer polishing), Bob Sprague (AR coatings), and Farhad Hakimi (pushing powers higher at the later stages of development). Aram Mooradian was the visionary who initiated the program, guided the program along, and with whom we discussed all aspects and nuances of the work to overcome a continuous string of challenges. Aram had the physical intuition to see OPS laser operation, even when he couldn't tell exactly how to get there. I felt we were “father and mother” team with Aram, nursing and raising our “baby” OPSLs.
Why do I think we succeeded at Micracor? Various groups had worked on OPSLs for several decades, but a common thread was that when they got results good enough for a publication, they were satisfied and stopped at the publication. We were a small company and had to push OPS concept to the commercial performance level to get funding for our work and for the company, both from government program sponsors and from investors. We were fighting for survival as a company, and just a publication was not good enough. And we had persevered, overcoming challenge, after challenge, after challenge, finding a path forward at every step. First, we got some initial miniscule amount of light from the laser. At that point, we had something to work with, and we just kept optimizing and improving, and we never stopped, until the company went out of business. In this process, physical understanding was critical, and fabrication ability was critical. Our work went through multiple iteration cycles: physical modeling and design – device fabrication – characterization of the materials and devices – analysis of the device characteristics – and then finding ways to improve in the next iteration cycle.
Another reason we succeeded at Micracor is the tremendous progress that had been made in the semiconductor technology in the preceding years and that was now available to us. Semiconductor lasers had progressed from simple edge‐emitting homojunction GaAs devices grown by liquid‐phase epitaxy (LPE) in the 1960s, to the MBE (molecular‐beam epitaxy) and MOCVD/MOVPE (metal‐organic chemical vapor deposition/vapor‐phase epitaxy) grown semiconductor structures with bandgap engineering to manipulate their electronic and optical properties, to strain‐engineered and strain‐compensated quantum‐confined structures, to vertical cavity laser structures with grown multilayer semiconductor DBR mirrors. Such semiconductor materials and structures were now also understood well enough to be grown commercially in companies such as Epitaxial Products International (EPI) PLC (now IQE PLC). These developments enabled diode VCSELs in the late 1980s; we applied these technologies to the optically pumped vertical external‐cavity configuration in the early 1990s. Another key enabler was the availability of new multiwatt multitransverse‐mode semiconductor diode pump lasers. Such pump lasers had been developed in the 1980s for pumping solid‐state lasers, e.g. 808 nm pumps for Nd:YAG lasers.
What were the resources and facilities available to us at Micracor? We had an electron beam evaporator for optical dielectric coatings and a thermal evaporator for chip metallization; we had a polishing facility for thinning the wafers; we also had a spectrophotometer for optical characterization of wafers and coatings. However, we did not have an optical spectrum analyzer (OSA) – commercial OSA was just too expensive. So I made a homemade OSA, converting a grating monochromator to a rudimentary calibrated OSA by rigging a fiber input and a detector output to the input and output slits, driving the grating stepper motor from a computer D/A output and reading the detector output into computer A/D. I periodically calibrated this OSA using a He‐Ne laser. To my surprise, coming from Bell Labs, I found that it is possible to do interesting science in a small company environment and with limited resources, of course given the right circumstances.
1.4 OPSL Development at Micracor: First Steps
In August of 1993 when I came to Micracor, it had an SBIR program running, Phase I, on the topic of “High Power Multi‐Segmented Semiconductor Lasers.” Figure 1.3 shows the device that was promised ultimately to the sponsors, a high‐power multiple‐bounce optically pumped surface‐emitting semiconductor laser. The project was funded by the Department of Defense, U.S. Army Space and Strategic Defense Command, colloquially known then as the “Star Wars” initiative. With not much to show for the program accomplishments at this point, I had two months to finish the project, write the final report, and convince program managers to give us more money for Phase II of the project.
I wrote and submitted the final report for Phase I at the end of September 1993: we had seen no laser light but had done enough experiments to project high hopes for the future; now we just had to sell these hopes to the program managers. We had two semiconductor samples, epigrown DBR stack plus quantum‐well gain region on top, to work with, grown and available through Aram's numerous connections: an 850 nm AlGaAs/GaAs wafer from Art D'Asaro at Bell Labs and a 900 nm InGaAs/GaAs wafer grown by Stephen Hersee at the University of New Mexico. Both wafers had major design shortcomings. What had been accomplished in Phase I? I had learned to align pump optics and laser cavity using Si cameras, I imaged and characterized pump and sample spontaneous emission beams, I had demonstrated 1.2 W pump power into ∼100 μm diameter spot from a 785 nm 3 W/500 μm‐wide stripe diode laser, and I observed strong amplified spontaneous emission (ASE) at 900 nm into the laser mode – but no lasing. I had sketched out a basic OPS laser model, e.g. see Figure 1.4, and determined a set of parameters to make a “single‐bounce” OPS surface‐emitting laser. Proposed designs had a semiconductor wafer structure with a DBR mirror and a gain region with ∼10 quantum wells in a resonant periodic gain structure [34], with two specific wafer designs proposed, an external spherical mirror with ∼100 mm radius of curvature and a reflectivity R > 97%, and a stable fundamental transverse mode laser cavity with ∼1 W threshold pump power for a 100 μm diameter mode. And I had a basic pump optics design with cylindrical lenses. Based on all this, while lacking an actual laser demonstration, we had claimed in the Phase I final report that we had demonstrated the “feasibility” of our optically pumped semiconductor OPS laser approach.
Figure 1.3 Multi‐bounce OPS laser Micracor promised in 1993 to make ultimately in its SBIR program “High Power Multi‐Segmented Semiconductor Lasers.”
Figure 1.4 Band diagram of the proposed optically pumped surface‐emitting semiconductor laser structure in the September 1993 final report for SBIR Phase I. Quantum wells are placed in a resonant periodic gain structure [34].
By the time I wrote Phase II SBIR project proposal in April 1994, we had a first major milestone accomplishment – we had seen first laser light at 900 nm with antireflection (AR) coated samples. Only a tiny amount of pulsed light was observed, but we already had something to work with. I wrote a very optimistic proposal, trying to overcome rudimentary results of the first phase, and I described in detail our approach to getting high power operation. Figure 1.5 shows our proposed initial single‐bounce OPS laser configuration. We got funding for Phase II.
When we started Phase II of the project in June 1994, we were already characterizing low power pulsed laser light. Our 900 nm InGaAs/GaAs samples had four quantum wells; the chips were AR coated, thinned to 100 μm, metallized, and soldered onto Cu heatsink. With R > 99% output mirror, and the laser driven with 100 μs pump pulses at 1 kHz, we saw very low power laser light but with an excellent beam profile, see Figure 1.6; chip heating turned the laser off for longer pulses. Thermal impedance of the chip was high; thermal path from active region to heatsink included DBR region and a 100 μm thick residual substrate. But the important thing was that we already had some light to work with, and we started to characterize and optimize our laser.
With money now available to continue the project, in the summer of 1994 we started by designing and growing a new optimized wafer structure based on our prior analysis and experiments. We targeted operating wavelength of 980 nm for application to pumping Er‐doped fiber amplifiers; wafer structures included 10 strained quantum wells in the mature InGaAs/GaAs material system. The structures were designed to be pumped at 808 nm and even included pump‐reflecting mirror in the DBR for more efficient double pass pump absorption. The wafers were grown successfully by MOVPE at EPI PLC in Cardiff, United Kingdom. These were fairly complex designs with many DBR and quantum well semiconductor layers; it's impressive that a commercial company at the time could already grow successfully such sophisticated structures. By then, we had developed techniques to characterize the grown OPS wafers with spectrophotometer wafer reflectance spectra, as well as surface and edge photoluminescence; laser power and spectral behavior were also extensively characterized, as well as their temperature dependence. We continued detailed modeling to explain the observed wafer and laser characteristics, both optical and thermal, and had designed several different pump optics configurations. In our first Phase II quarterly report in September 1994, we described our accomplishments: the new wafers lased only in the pulsed mode; using R = 97% output reflecting mirror, threshold was ∼400 mW and maximum pulsed output power was only 30 mW, limited by heating.
Figure 1.5 Single‐bounce OPS laser configuration in the April 1994 SBIR Phase II proposal.
Figure 1.6 Excellent beam profile of the pulsed optically pumped 900 nm InGaAs/GaAs laser, June 1994.
Figure 1.7 Two figures from the Micracor second quarterly report to the sponsor, January 1995: (a) laser and pump configuration, (b) pulsed and CW laser spectra. (c) Figure from the Phase II final report, July 1996: thermal offset design for OPS vertical cavity laser.
So a year after the program started, we had learned a lot, but got only 30 mW pulsed – pretty far away from the promised watts of CW power. This was time to get philosophical in the face of challenges, to collect thoughts and to see how to proceed forward. At the time, I was reminded of the quote from Harold Edgerton, MIT professor of the flash photography fame, “We worked and worked, but didn't get anywhere. That's how you know you're doing research.” Well, we definitely were doing research. I was also thinking of my work then as a friendly wrestling match with Nature; Nature is a tough opponent, but she is not malicious. I was learning to speak the language of Nature, learning to listen carefully and understand what she says, and trying to speak back to her in her own language; and if I were successful, she would listen and understand me, and do what I asked.
1.5 OPS Development at Micracor: Pushing Forward
In the next quarter we pushed forward, and here are the important advances from the second quarterly report in January of 1995, see Figure 1.7. Laser chips were thinned to 20–30 μm, to reduce thermal impedance between the quantum well active layer and heatsink, and soldered to Cu submounts – we got 130 mW pulsed output power and, crossing an important barrier, we had ∼1 mW CW! (at 0 °C). We had concluded that thermal effects limited our CW output power, and we focused on several approaches to improve further laser power performance. (i) We lowered thermal impedance of on‐chip mirror and substrate in order to decrease detrimental temperature rise of the active region. (ii) We made laser threshold less sensitive to temperature. To this end, we improved quantum well design with stronger carrier confinement to prevent carrier escape from the wells at higher temperatures. And finally, (iii) we placed the lasing mode resonance on the long wavelength side of the material gain peak at room temperature. This would compensate the larger temperature shift of the gain spectrum compared to the smaller temperature shift of the resonant wavelength and would align material gain spectral peak and lasing mode wavelength at the increased operating temperature of the laser active region. The key to laser power performance was addressing these thermal issues.
Figure 1.8 (a) Pulsed laser power up to 110 °C with weak temperature dependence, third quarterly report, March 1995. (b) 160 mW CW output power, program extension proposal, May 1995.
Our understanding and improvement steps proved to be correct. By the time I wrote our Phase II third quarterly report in March of 1995, we had made significant progress. Our second iteration redesigned 980 nm wafer structure was grown – the measured thermal impedance improved by a factor of two to 45 °C/W, in part due to elimination of the pump‐reflecting mirror; improved quantum well (QW) carrier confinement resulted in weaker dependence of threshold power on temperature, see Figure 1.8a; CW lasing was now observed at up to 30 °C; and we measured 57 mW CW output power at −5 °C. By May 1995 we had improved performance even further – the chip was now only 17 μm thick and soldered onto diamond heatspreader, which was in turn soldered to a Cu heatsink; CW output power was now 160 mW! See Figure 1.8b. And, based on these results, we got infusion of additional money from the sponsor in the program extension. We could keep going with OPS laser development.
In August of 1995, I had a chance for the first time to attend the Topical Meeting on Semiconductor Lasers in Keystone, Colorado – attending conferences was a luxury not always available in
