Grating-coupled, surface-emitting (GCSE) quantum-cascade lasers (QCLs) offer a pathway towards realizing watt-range, surface-emitted output powers in the mid-infrared spectral region with high beam quality. Previously we have reported wide-ridge GCSE QCLs which employed metal/semiconductor, 2nd-order distributed feedback (DFB) gratings with distributed Bragg reflector (DBR) terminations. We report here on the lasing characteristics of narrow-ridge (similar to 7 mu m-wide) GCSE devices, which employ the STA-RE-type active-region design, for obtaining single-spatial-mode both laterally and longitudinally. The QCL structure was grown using Metalorganic Chemical Vapor Deposition (MOCVD) and the grating was defined using a combination of e-beam lithography patterning and wet-chemical etching, and the ridge (similar to 7 mu m) was dry-etched. The total length of the DFB + DBR regions is 5.1 mm, and was electrically isolated in the DBR regions by employing AlOx. Due to resonant coupling of the guided light to the antisymmetric surface-plasmon modes of the 2nd-order grating, the antisymmetric (A) modes are strongly absorbed; thus, allowing for the symmetric (S) mode to be favored to lase. Initial devices have demonstrated maximum pulse output power from the surface of similar to 150 mW at 4.88 mu m, with only similar to 10% power emitted from the edge facets. An anti-reflective (AR) coating of a quarter-wavelength Y2O3 layer was applied on the emission window, drastically improving the far-field beam pattern, that resulting in a central, near-diffraction-limited single-lobe beam pattern. COMSOL simulations were performed to further optimize the SE-base design for high CW performance. Parameter sweeps of cladding-layer thickness, grating height, and grating duty cycle were performed, which identified design tradeoffs for the various structural parameters.

When conventionally calculating carrier leakage for state-of-the-art quantum cascade lasers (QCLs), that is, LO-phonon-assisted leakage from the upper laser level via electron thermal excitation to high-energy active-region (AR) states, followed by relaxation to low-energy AR states, ∼18%-wide gaps were recently found between calculated and experimentally measured internal efficiency values. We incorporate elastic scattering [i.e., interface-roughness (IFR) and alloy-disorder scattering] into the carrier-leakage process and consider carrier leakage from key injector states as well. In addition, the expressions for LO-phonon and IFR-triggered carrier-leakage currents take into account the large percentage of thermally excited electrons that return back to initial states via both inelastic and elastic scattering. As a result, we find that the gaps between theoretical and experimental internal efficiency values are essentially bridged. Another finding is that, for the investigated state-of-the-art structures, IFR scattering causes the total carrier leakage to reach values as much as an order of magnitude higher than conventional inelastic scattering-only leakage. The developed formalism opens the way to significantly increase the internal efficiency (i.e., to more than 80%) via IFR-scattering engineering, such that maximum wall-plug efficiencies close to projected fundamental, both-facets values (e.g., 42% at λ = 4.6 μm) can be achieved. By employing this formalism, we reached a 4.6 μm-emitting-QCL preliminary design for suppressing IFR-triggered carrier leakage, which provides an internal efficiency of 86% as well as a projected single-facet wall-plug efficiency value of 36% at a heatsink temperature of 300 K.

Quantum cascade lasers (QCLs) employ intersubband transitions between conduction-band (CB) energy states in multi-quantum-well (MQW) structures, carrier tunneling between adjacent MQW stages, and coherent-light emission from multiple (30-50) stages. Unlike interband-transition semiconductor lasers, QCLs are not affected by Auger recombination in the mid-infrared wavelength range (λ= 3-20 μm); thus, allowing them to readily reach lasing at room temperature (RT). However, using multiple stages leads to high voltages (≥ 10 V) that significantly decrease the wallplug efficiency, η wp . Thus, although pulsed RT operation was obtained in 1996, it took until 2002 to achieve CW RT operation and until 2008 to obtain 1 W CW power.

Investigation of catastrophic degradation in high-power, buried-heterostructure quantum cascade lasers, using focused ion beam and high-resolution TEM techniques, has revealed dislocations generated, as a result of degradation, mostly in areas away from the active region. © 2019 The Author(s)

Thermal profiles of a single-element quantum cascade laser (QCL) and a five-element QCL array at different bias currents, under quasi-continuous-wave (QCW) conditions, are obtained using the charge-coupled device (CCD)-based thermoreflectance imaging technique. Peak temperature changes of 55 K and 105 K are measured on the single-element QCL (operating at 1.2 A) and the central element of the QCL array (operating at 4.2 A), respectively. The average facet temperature of the single QCL device shows a linear relationship with the dissipated power, indicating an effective thermal resistance of Rth = 3.0 ± 0.2 K/W (7%) for the device. The thermal transient behavior of the single QCL device, in response to a 35 μs-wide heating pulse, is also measured. From the transient curve, an effective thermal time constant of τth = 9.5 ± 0.4 μs (4%) is obtained. Experimental results are compared to the results obtained from heat-transfer models for both the single-element and array devices. Thermal profiles show a thermal lensing effect at the facet of the single-element QCL. In the array device, a more pronounced heating is observed at the center of the device while the temperature gradually decreases away from the central element.

A numerical model is developed for a phase-locked array of quantum cascade lasers. The population density is derived from rate equations. The temperature distribution for stationary generation is produced by ohmic heating. It is shown that the results of above-threshold operation modeling by the semi-vectorial beam propagation method are in good agreement with the modal analysis provided by the vectorial-COMSOL solver supplemented with the Rigrod’s model estimations. The wall-plug efficiency and the limits of the single-mode lasing are found. Thermal lensing is shown to be the main reason limiting the single-spatial-mode power under CW-operating conditions.

A three-element phase-locked array of ∼5.3 μm-emitting quantum cascade lasers fabricated by a two-step MOCVD process is demonstrated which operates in a diffraction-limited beam to 5.3 W peak pulsed power. To compensate for overetching at the array edges, the fabrication process was modified such that it resulted in array elements of equal width. Then the array operates in a pure in-phase array mode, in agreement with simulations based on the actual device dimensions, which indicate that resonant leaky-wave coupling occurs for this device's geometry. Measured uniform near-field intensity profiles confirm resonant coupling. Simulations indicate that chirping of the element widths would allow for modification of the near-field intensity profile to mitigate the effects of thermal lensing in continuous-wave operation.

Thermal management efforts in nanoscale devices must consider both the thermal properties of the constituent materials and the interfaces connecting them. It is currently unclear whether alloy/alloy semiconductor superlattices such as InAlAs/InGaAs have lower thermal conductivities than their constituent alloys. We report measurements of the crossplane thermal resistivity of InAlAs/InGaAs superlattices at room temperature, showing that the superlattice resistivities are larger by a factor of 1.2-1.6 than that of the constituent bulk materials, depending on the strain state and composition. We show that the additional resistance present in these superlattices can be tuned by a factor of 2.5 by altering the lattice mismatch and thereby the phonon-mode mismatch at the interfaces, a principle that is commonly assumed for superlattices but has not been experimentally verified without adding new elements to the layers. We find that the additional resistance in superlattices does not increase significantly when the layer thickness is decreased from 4 to 2 nm. We also report measurements of 250-1000 nm thick films of undoped InGaAs and InAlAs lattice-matched to InP substrates, for there is no published thermal conductivity value for the latter, and we find it to be 2.24 ± 0.09 at 22 °C, which is ∼2.7 times smaller than the widely used estimates.

Carrier-leakage suppression and fast carrier extraction allow quantum cascade lasers to reach internal efficiencies close to fundamental limits (∼ 90 %). Then, CW wallplug efficiencies > 40 % and powers > 10 W become possible.

High-CW-power (i.e., watt-range), mid-infrared (IR) (λ= 3-15 μm) quantum cascade lasers (QCLs) are needed for a wide range of applications, from remote sensing to infrared countermeasures. Many of these applications require single-spatial-mode operation with beam stability to multi-watt-range output powers. Scaling the CW single-mode output power requires optimization of the QCL active region, as well as device architectures allowing for scaling the lateral width. Single-element, edge-emitting QCLs operating in the 4.5-5.0 μm wavelength region generally requires a relatively narrow element width (~ 5 microns) to maintain stable, single-spatial-mode CW operation up to the 1.5-2.0 watt-range output power levels. Larger width devices operate to high CW output powers (~ 5W), but suffer from multi-mode operation as well as beam instabilities with drive current [1].

Mid-infrared quantum cascade lasers (QCLs) are a growing industry and are being introduced into the marketplace primarily for low-output-power operation. High-power (> 1 W) continuous wave (CW) QCLs are expected to become commercially viable as they become more efficient and the necessary thermal dissipation requirements are achieved. However, there is a relative lack of knowledge regarding the degradation and failure mechanisms of QCLs under high power CW operation [1,2]. QCLs are expected to have different degradation and failure modes than diode lasers, because nonradiative recombination at the facets is not an issue. To push towards wider commercial adoption, lifetesting and failure analyses of high-power QCLs are performed. Previously reported QCL lifetests were carried out at relatively low output powers (~200 mW) and revealed activation energies as high as 1.2 eV, with the primary failure mechanism being reported to be oxidation of the front facet [3]. Here, we report on initial constant-power lifetest studies of QCLs emitting at λ ~ 5.0 μm and operating at 5 times the output power previously reported (i.e. at 1W CW). To mitigate the failure mechanism previously observed and improve device output, both facets have coatings: a high-reflectivity (HR) back-facet coating and a 14% low-reflectivity (LR) front-facet coating. The devices are mounted epi-side-down on copper with indium and tested under constant-power operation in a controlled environment.

We demonstrate the conversion of lattice-matched InGaAs/InAlAs quantum-cascade-laser (QCL) active-region material into an effective current-blocking layer via proton implantation. A 35-period active region of an 8.4 μm-emitting QCL structure was implanted with a dose of 5 × 1014 cm−2 protons at 450 keV to produce a vacancy concentration of ∼1019 cm−3. At room temperature, the sheet resistance, extracted from the Hall measurements, increases by a factor of ∼240 with respect to that of an unimplanted material. Over the 160–320 K temperature range, the activation energy of the implanted-material Hall sheet-carrier density is 270 meV. The significant increase in room-temperature sheet resistance indicates that upon implantation deep carrier traps have been formed in the InAlAs layers of the superlattice. Fabricated mesas show effective current blocking, at voltages ≥10 V, up to at least 350 K. Thus, the implanted InGaAs/InAlAs superlattices are highly resistive to at least 350 K heat sink temperature. Such implanted material should prove useful for effective current confinement in 8–15 μm-emitting InP-based single-emitter QCL structures as well as in resonant leaky-wave coupled phase-locked arrays of QCLs.

Narrow beam (3.2 × D.L. FWHM) is demonstrated up to 5.85 W pulsed output power from a five-element phase-locked array of 4.7 μm-emitting quantum cascade lasers. Devices are fabricated by a two-step MOCVD process and operate predominately in an in-phase array mode, in agreement with design simulation studies.

Five-element phase-locked arrays of 4.7 μm-emitting quantum cascade lasers are demonstrated which operate either in a near-diffraction-limited (D.L.) beam to 3.6 W peak pulsed power or in a narrow beam (<; 3.5 × D.L.) up to 6.1 W peak pulsed power. Devices are fabricated by a two-step MOCVD process and operate predominantly in an in-phase array mode, in agreement with design-simulation studies. Analysis based on the actual device dimensions indicate that near-resonant leaky-wave coupling occurs. Scaling to a larger number of array elements and optimization for resonant operation is expected to lead to further increases in output power while maintaining high beam quality.

Phase-locking, via leaky-wave coupling, of five 4.7 µm-emitting quantum cascade lasers is demonstrated for coherent-power scaling. Non-resonant devices fabricated by two-step MOCVD operate in a mixture of in-phase and out-of-phase modes to 3.85 W peak pulsed output power. Design analysis shows pure in-phase-mode operation under resonant-coupling occurs for optimized devices.

By stepwise tapering both the barrier heights and quantum-well depths in the active regions of 8.7 μm- and 8.4 μm-emitting quantum cascade lasers (QCLs) virtually complete carrier-leakage suppression is achieved, as evidenced by high values for both the threshold-current characteristic temperature coefficient T0 (283 K and 242 K) and the slope-efficiency characteristic temperature coefficient T1 (561 K and 279 K), over the 20–60 °C heatsink-temperature range, for low- and high-doped devices, respectively. Such high values are obtained while the threshold-current density is kept relatively low for 35-period, low- and high-doped devices: 1.58 kA/cm2 and 1.88 kA/cm2, respectively. In addition, due to resonant extraction from the lower laser level, high differential-transition-efficiency values (89-90%) are obtained. In turn, the slope-efficiency for 3 mm-long, 35-period high-reflectivity (HR)-coated devices are: 1.15-1.23 W/A; that is, 30- 40 % higher than for same-geometry and similar-doping conventional 8-9 μm-emitting QCLs. As a result of both efficient carrier-leakage suppression as well as fast and efficient carrier extraction, the values for the internal differential efficiency are found to be ≈ 86%, by comparison to typical values in the 58-67 % range for conventional QCLs emitting in the 7-11 μm wavelength range.

Grating-coupled surface-emitting (GCSE) lasers generally operate with a double-lobed far-field beam pattern along the cavity-length direction, which is a result of lasing being favored in the antisymmetric grating mode. We experimentally demonstrate a GCSE quantum-cascade laser design allowing high-power, nearly single-lobed surface emission parallel to the longitudinal cavity. A 2nd-order Au-semiconductor distributed-feedback (DFB)/distributed-Bragg-reflector (DBR) grating is used for feedback and out-coupling. The DFB and DBR grating regions are 2.55 mm- and 1.28 mm-long, respectively, for a total grating length of 5.1 mm. The lasers are designed to operate in a symmetric (longitudinal) grating mode by causing resonant coupling of the guided optical mode to the antisymmetric surface-plasmon modes of the 2nd-order metal/semiconductor grating. Then, the antisymmetric modes are strongly absorbed by the metal in the grating, causing the symmetric mode to be favored to lase, which, in turn, produces a single-lobed beam over a range of grating duty-cycle values of 36%–41%. Simulations indicate that the symmetric mode is always favored to lase, independent of the random phase of reflections from the device's cleaved ends. Peak pulsed output powers of ∼0.4 W were measured with nearly single-lobe beam-pattern (in the longitudinal direction), single-spatial-mode operation near 4.75 μm wavelength. Far-field measurements confirm a diffraction-limited beam pattern, in agreement with simulations, for a source-to-detector separation of 2 m.

By stepwise tapering, both the barrier heights and quantum-well depths in the active regions of 8.7–8.8 μm-emitting quantum-cascade-laser (QCL) structures, virtually complete carrier-leakage suppression is achieved. Such step-taper active-region-type QCLs possess, for 3 mm-long devices with high-reflectivity-coated back facets, threshold-current characteristic temperature coefficients, T0, as high as 283 K and slope-efficiency characteristic temperature coefficients, T1, as high as 561 K, over the 20–60 °C heatsink-temperature range. These high T0 and T1 values reflect at least a factor of four reduction in carrier-leakage current compared to conventional 8–9 μm-emitting QCLs. Room temperature, pulsed, threshold-current densities are 1.58 kA/cm2; values comparable to those for 35-period conventional QCLs of similar injector-region doping level. Superlinear behavior of the light-current curves is shown to be the result of the onset of resonant extraction from the lower laser level at a drive level of ∼1.3× threshold. Maximum room-temperature slope efficiencies are 1.23 W/A; that is, slope efficiency per period values of 35 mW/A, which are 37%–40% higher than for same-geometry conventional 8–9 μm-emitting QCLs. Since the waveguide-loss coefficients are very similar, we estimate that the internal differential efficiency is at least 30% higher than in conventional QCLs. Such high internal differential efficiency values reflect the combined effect of nearly complete carrier-leakage suppression and high differential efficiency of the laser transition (∼90%), due to resonant extraction from the lower laser level.