Accurate nonequilibrium Green’s-functions simulations of record-performance 4.9 µm- and 8.3 µm-emitting QCLs, employing Photon-Induced Carrier Transport (PICT), require inclusion of graded interfaces when calculating Interface-Roughness (IFR) scattering. Matching threshold-current densities and V-I characteristics, all IFR parameters were extracted. The root-mean-square height and in-plane correlation length are found to be higher and lower, respectively, than when assuming abrupt interfaces. Abrupt-interfaces’ modeled 4.9 µm-emitting QCLs lack PICT action, which reduces the calculated maximum wall-plug efficiency, η(wp,max), value from 27% to 18.7%. Abrupt-interfaces’ modeled 8.3 µm-emitting QCLs have approximately 70% higher relative leakage-current density, which reduces the calculated η(wp,max) value from 17% to 11.7%.
Self-heating in mid-infrared QCLs leads to beam instabilities and facet related failures. Single-element 4.6 μm-emitting BH QCLs were fabricated, where a tapered region scales the output power and, ahead of the emitting aperture, a narrow section provides mode filtering for suppressing high-order spatial modes. Beam-stability measurements indicate a small degree of collimated-beam centroid motion (< 0.25 mrad) can be achieved at >1.5W QCW output powers. Comparisons between short-pulse current and QCW operation reveal the impact of thermal lensing on the beam properties, while full 3D modeling provides insights into influence of device geometry on mode selection.
Accurate simulation of V-I characteristics for mid-IR quantum cascade lasers (QCLs) with photon-induced carrier transport (PICT) is achieved by using the non-equilibrium Green’s function method coupled with the interface-roughness scattering formalism taking into account graded interfaces and axial correlation lengths. Analysis of 4.9 µm- and 8.3 µm-emitting, buried-heterostructure (BH) QCLs reveals that PICT action reduces the differential resistance by a factor of 2.5 and increases the maximum-current density by ~ 30 % compared to conventional BH QCLs, which explains their record-high, single-facet wall-plug efficiency values (i.e., 27 % and 17 %). Interface grading allows obtaining emission wavelengths close to experiment.
Scaling the coherent power of mid-infrared (IR)-emitting quantum cascade lasers (QCLs) to the multi-watt range remains an important objective for applications where the laser beam needs to travel through air to remote targets, such as freespace communication links. For such applications requiring long-range pointing accuracy, measurements of beam stability are also important. We present beam-quality measurement results of narrow-ridge (4-5 μm), 4.6 μm-emitting buriedheterostructure (BH) QCLs. A 40-stage, step-tapered active-region (STA) structure was grown by MOCVD, and ICP etching was used to make deep ridges. InP:Fe was preferentially regrown in the field regions by using an SiO2 mask for ridge etching and Hydride Vapor Phase Epitaxy (HVPE). The HVPE process is attractive for selective regrowth, since high growth rates (0.2-0.3 μm/min) can be utilized, and highly planar top surfaces can readily be obtained. HVPE regrowth has been previously employed for BH devices of MBE-grown QCL ridges, but beam-stability measurements were not reported. HR-coated, 7.5 mm-long devices were measured under QCW operation (100 μsec pulse width, 0.5%-10% duty cycle) – very good beam quality factors, M2 < 1.2, were observed for both 4 μm and 5 μm ridge widths, but the narrower ridge exhibited better pointing stability. Collimated 5 μm-wide BH devices displayed some small degree of centroid motion with increasing power (< 0.125 mrad). This corresponds to a targeting error of ~1.25 cm over a distance of 100 m. Significantly improved lateral-beam stability was observed for narrower ridge width, although at the expense of reduced output power.
Step-taper active-region (STA) quantum cascade lasers (QCLs) allow for both carrier-leakage suppression and fast, miniband-like lower-laser (ll) level depopulation. That has led to an internal-efficiency value of ~ 77 % for ~ 5 μmemitting devices, a record-high value for 4.5-6.0 μm-emitting QCLs. We have recently shown that that value can be basically accounted for when considering both LO-phonon- and interface-roughness (IFR)-triggered carrier leakage from the upper-laser (ul) level and injector states. The same carrier-leakage analysis is applied to MOCVD-grown ~ 8.0 μmemitting, STA-type QCLs, and we find that the internal efficiency reaches a high value of ~ 73.6 %, due to a record-high injection-efficiency value (86.7 %) and to an IFR-enhanced laser-transition efficiency value of ~ 85 %. In contrast, for a conventional MOCVD-grown, ~ 8.0 μm-emitting QCLs the internal efficiency is found to be only ~ 55 %, typical of values extracted from experimental data of mid-infrared-emitting conventional QCLs. The ul-level lifetime is found to be controlled by LO-phonon and alloy-disorder (AD) scattering for typical MOCVD-grown QCLs, just like for 4.5-5.0 μm-emitting QCLs. However, for typical MBE-grown QCLs the ul-level lifetime is controlled by LO-phonon, AD and IFR scattering. The ll-level lifetime is found to controlled by both LO-phonon and IFR scattering. Besides the high internal-efficiency value, the use of excited-state injection and a low voltage defect result in the STA QCL reaching a single-facet wall-plug efficiency value of 10.6 %, a record-high single-facet value for 8-11 μm-emitting QCLs grown by MOCVD and holding potential for CW operation.
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 (~7 μm-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 (~7 μm) 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 ~150 mW at 4.88 μm, with only ~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.
Step-taper active-region (STA) quantum cascade lasers (QCLs) allow for both carrier-leakage suppression and ultrafast, miniband-like carrier extraction. In turn, that has led to very high internal-efficiency ni values: ~ 77 % and 80-86 % from ~ 5.0 μm- and 8-9 μm-emitting QCLs, respectively. Based on extracted parameters that characterize the interfaceroughness (IFR) scattering, a study has been performed of the effects of elastic scattering, both IFR and alloy-disorder (AD) scattering, on 5.0 μm-emitting STA-QCLs. We find that the laser-transition efficiency ntr is enhanced by ~15 % (i.e., from 83 % to ~ 95 %) due to the much stronger effect of elastic scattering on the lower-laser-level lifetime than on the effective upper-laser-level lifetime. In turn, the injection efficiency: ninj = ni /ntr , reaches ~ 81 %; that is, the highest injection-efficiency value obtained to date from QCLs. Furthermore, we find that the projected upper limit for the pulsed wall-plug efficiency can reach values as high as 44.4 % for 4.6 μm-emitting devices; thus, raising the possibility of CW operation of 4.5-5.0 μm-emitting QCLs with wallplug-efficiency values as high as 40 %.
We are developing midwave infrared (mid-IR) quantum cascade lasers (QCLs) and interband cascade lasers (ICLs) bonded to silicon. The heterogeneous integration of mid-IR photonic devices with silicon promises to enable low-cost, compact sensing and detection capabilities that are compatible with existing silicon photonic and electronic technologies. The first Fabry-Perot QCLs on silicon were bonded to pre-patterned silicon-on-nitride-on-insulator (SONOI) substrates. Lateral tapers in the III-V mesas transferred the optical mode from the hybrid III-V/Si active region into the passive silicon waveguides, with feedback provided by reflections from both the III-V tapers and the polished passive silicon facets. Lasing was observed at 4.8 m with threshold current densities as low as 1.6 kA/cm2 when operated in pulsed mode at T = 20 ºC. The first mid-IR DFB lasers integrated on silicon employed gratings patterned into the silicon waveguides before bonding. Over 200 mW of pulsed power was generated at room temperature, and operated to 100 °C with T0 = 199 K. Threshold current densities were measured below 1 kA/cm2.The grating imposed considerable wavelength selectivity and 22 nm of thermal tuning, even though the emission was not spectrally pure. Ongoing research focuses on flip-chip bonding to improve heat sinking for continuous-wave operation, and arrayed waveguide gratings for beam combining. ICLs have also been bonded to silicon and the GaSb substrate has been chemically removed with an InAsSb etch-stop layer. Tapered ICL ridges designed for lasing in a hybrid III-V/Si mode have been processed above passive silicon waveguides patterned on SOI. A goal is to combine the power generated by arrays of QCLs and ICLs residing on the same chip into a single, high-quality output beam.
The modal characteristics of nonresonant five-element phase-locked arrays of 4.7-μm emitting quantum cascade lasers (QCLs) have been studied using spectrally resolved near- and far-field measurements and correlated with results of device simulation. Devices are fabricated by a two-step metal-organic chemical vapor deposition process and operate predominantly in an in-phase array mode near threshold, although become multimode at higher drive levels. The wide spectral bandwidth of the QCL’s core region is found to be a factor in promoting multispatial-mode operation at high drive levels above threshold. An optimized resonant-array design is identified to allow sole in-phase array-mode operation to high drive levels above threshold, and indicates that for phase-locked laser arrays full spatial coherence to high output powers does not require full temporal coherence.
We discuss novel approaches to improve the tuning bandwidth and power output of terahertz (THz) sources based on difference-frequency generation (DFG) in mid-infrared quantum cascade lasers (QCLs). Using a double Littrow external-cavity system, we experimentally demonstrate that both doubly-resonant terms and optical rectification terms in the expression for the intersubband optical nonlinearity contribute to THz generation in DFG-QCLs and report THz DFG-QCLs with the optimized optical rectification terms. We also demonstrate a hybrid DFG-QCL device on silicon that enables significant improvement on THz out-coupling efficiency and results in more than 5 times higher THz output power compared to that of a reference device on its native semi-insulating InP substrate. Finally, we report for the first time the THz emission linewidth of a free-running continuous-wave THz DFG-QCL.
Implementation of the step-taper active-region (STA) design to 8-9 μm-emitting quantum cascade lasers (QCLs) has resulted in both high T0 and T1 values: 220 K and 665 K, and short lower-level lifetimes: 0.12 ps. In turn, the internal differential efficiency ηid, which is the product of the injection efficiency and the differential laser-transition efficiency, reaches values as high as 86 % for both 8.4 μm- and 8.8 μm-emitting QCLs. Such ηid values are 30-50% higher than those obtained from conventional QCLs emitting in the 7-11 μm wavelength range. Achieving both carrier-leakage suppression and miniband-like carrier extraction in mid-infrared (IR) QCLs leads to ηid values close to the fundamental limit of ~ 90 %. In turn, the currently employed fundamental wallplug-efficiency limits over the mid-IR wavelength range have to be increased by ~ 34 % (e.g., the wallplug-efficiency limit at λ= 4.6 μm increases from 29 % to 39 %). Preliminary results from STA-type 4.8-5.0 μm-emitting QCLs include 1.5 W CW operation, and 77 % internal differential efficiency; that is, 30-50% higher than the ηid values obtained from conventional 4.0-6.5μm-emitting QCLs.
Silicon integration of mid-infrared (MIR) photonic devices promises to enable low-cost, compact sensing and detection capabilities that are compatible with existing silicon photonic and silicon electronic technologies. Heterogeneous integration by bonding III-V wafers to silicon waveguides has been employed previously to build integrated diode lasers for wavelengths from 1310 to 2010 nm. Recently, Fabry-Perot Quantum Cascade Lasers integrated on silicon provided a 4800 nm light source for MIR silicon photonic applications. Distributed feedback (DFB) lasers are appealing for many high-sensitivity chemical spectroscopic sensing applications that require a single frequency, narrow-linewidth MIR source. While heterogeneously integrated 1550 nm DFB lasers have been demonstrated by introducing a shallow surface grating on a silicon waveguide within the active region, no mid-infrared DFB laser on silicon had previously been reported. Here we demonstrate quantum cascade DFB lasers heterogeneously integrated with silicon-on-nitride-oninsulator (SONOI) waveguides. These lasers emit over 200 mW of pulsed power at room temperature and operate up to 100 °C. Although the output is not single mode, the DFB grating nonetheless imposes wavelength selectivity with 22 nm of thermal tuning.
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) quantum-cascade lasers (QCLs) are demonstrated with high-power, single-lobe surface emission. 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 mode by causing resonant coupling of the guided optical mode to the antisymmetric surface-plasmon modes of the
2nd-order metal/semiconductor grating. In turn, the antisymmetric longitudinal modes are strongly absorbed by the metal in the grating, causing the symmetric longitudinal mode to be favored to
lase, which produces a single lobe beam over a grating duty-cycle range of 36-41 %. Simulations
indicate that the symmetric mode is always favored to lase, independent of the random phase of
residual reflections from the device’s cleaved ends. Peak pulsed output powers of ~ 0.4 W were measured with single-lobe, single-mode operation near 4.75 μm.
Remarkable progress made in quantum cascade lasers (QCLs) has led them to find an increasing number of applications in remote sensing, chemical sensing, and free space communications, in addition to potential space applications. However, little has been reported on reliability and failure modes of QCLs although it is crucial to understand failure modes and underlying degradation mechanisms in developing QCLs that meet lifetime requirements for space missions. Focused ion beam (FIB) techniques have been employed to investigate failure modes in various types of laser diodes. Our group has also used FIB to study failure modes in single-mode and multi-mode InGaAs-AlGaAs strained QW lasers, but few groups have used this technique to investigate failure modes in QCLs. In our study, we report on destructive physical analysis (DPA) of degraded InGaAs-InAlAs QCLs using FIB and high-resolution TEM techniques. The active region of QCLs that we studied consisted of two-23 stage layers of InGaAs-InAlAs separated by a 0.5 μm thick InP spacer layer for 8.4μm QCLs and 30-stage layers of lattice-matched InGaAs-InAlAs heterostructure for 4.7μm QCLs. The MOVPE-grown laser structures were fabricated into deep-etched ridge waveguide QCLs. L-I-V-spectral characteristics were measured at RT under pulsed operation. Our 8.4μm QCLs with as-cleaved and HR-coated facets showed a laser threshold of 1.7 A and a threshold voltage of 13 V at RT, whereas our 4.7μm QCLs without facet coating showed threshold currents of 320 - 400 mA and threshold voltages of 13 - 13.5V. Failures were generated via short-term tests of QCLs. FIB systems were used to study the damage area on the front facet and also to prepare TEM cross sections at different locations along the waveguide for defect and chemical analyses using a HR-TEM. In contrast to the COMD damaged area showing as a blister on the front facet of QW lasers, the damaged area of QCLs was significantly extended into the InP substrate due to a much less absorption of lasing photons in QCLs. Our detailed destructive physical analysis results are reported including defect, structural, and chemical analysis results from degraded QCLs.
Three-dimensional above-threshold analyses of high-index-contrast (HC) photonic-crystal (PC) quantum-cascade-laser arrays (QCLA) structures, for operation at watt-range CW powers in a single spatial mode, have been performed. Threeelement HC-PC structures are formed by alternating active- antiguided and passive-guided regions along with respective metal-electrode spatial profiling. The 3-D numerical code takes into account absorption and edge-radiation losses. Rigrod’s approximation is used for the gain. The specific feature of QCLA is that only the transverse component of the magnetic field sees the gain. Results of above-threshold laser modeling in various approximate versions of laser-cavity description are compared with the results of linear, full-vectorial modeling by using the COMSOL package. Additionally, modal gains for several higher-order optical modes, on a ‘frozen gain background’ produced by the fundamental-mode, are computed by the Arnoldi algorithm. The gain spatial-hole burning effect results in growth of the competing modes’ gain with drive current. Approaching the lasing threshold for a competing higher-order mode sets a limit on the single-mode operation range. The modal structure and stability are studied over a wide range in the variation of the inter-element widths. Numerical analyses predict that the proper choice of construction parameters ensures stable single-mode operation at high drive levels above threshold. The output power from a single- mode operated QCLA at a wavelength of 4.7 μm is predicted to be available at multi-watt levels, although this power may be restricted by thermal effects.
On-chip resonant leaky-wave coupling of quantum cascade lasers (QCLs) emitting at 8.36 μm has been realized by selective regrowth of interelement layers in curved trenches, defined by dry and wet etching. The fabricated structure provides large index steps (Δn = 0.10) between antiguided-array element and interelement regions. In-phase-mode operation to 5.5 W front-facet emitted power in a near-diffraction-limited far-field beam pattern, with 4.5 W in the main lobe, is demonstrated. A refined fabrication process has been developed to produce phased-locked antiguided arrays of QCLs with planar geometry. The main fabrication steps in this process include non-selective regrowth of Fe:InP in interelement trenches, defined by inductive-coupled plasma (ICP) etching, a chemical polishing (CP) step to planarize the surface, non-selective regrowth of interelement layers, ICP selective etching of interelement layers, and non-selective regrowth of InP cladding layer followed by another CP step to form the element regions. This new process results in planar InGaAs/InP interelement regions, which allows for significantly improved control over the array geometry and the dimensions of element and interelement regions. Such a planar process is highly desirable to realize shorter emitting wavelength (4.6 μm) arrays, where fabrication tolerance for single-mode operation are tighter compared to 8 μm-emitting devices.
Strained-layer superlattice (SL) structures have been grown by metalorganic vapor phase epitaxy
(MOVPE) on metamorphic buffer layers (MBLs) for application in intersubband-transition devices,
such as quantum cascade lasers. Using the MBL as an adjustable lattice-parameter platform, we have
designed relatively-low-strain quantum-cascade-laser structures that will emit in the 3.0-3.5 μm
wavelength range while suppressing carrier leakage from the upper laser level. Thick (10-12 μm)
compositionally-graded, hydride-vapor-phase-epitaxy (HVPE)-grown MBL structures are employed.
To improve the planarity of the MBL surface, we employ chemical mechanical polishing (CMP)
followed by wet chemical etching prior to the growth of the SL/device structures. We find that the
wet-chemical etching step is crucial to remove residual damage introduced during CMP. 20-period
InxGa1-xAs (wells)/AlyIn1-yAs (barriers) SLs grown on the MBLs are characterized by x-ray
diffraction (XRD). Intersubband electroluminescence emission is observed in the 3.5 μm wavelength
range from devices employing such SL structures.
Bulk InGaAs layers with a 1eV band-gap grown on GaAs substrates are attractive for high efficiency multi-junction
solar cells. However, a large amount of lattice mismatch between bulk InGaAs layer and GaAs substrate
necessitates development of novel metamorphic buffer layers (MBL). A number of research groups have reported
various MBLs for applications including HBTs, HEMTs, lasers, and solar cells. In this study, we report carrier
dynamics and defects in MOVPE-grown bulk InGaAs layers (Eg = ~ 1.0 - 1.1 eV at 300K) with two different types
of MBLs including InGaAs and InGaPSb. We also report the effect of chemical-mechanical polishing (CMP)
process on carrier lifetimes and the properties of the films subsequently grown on top of the MBL. We employed
time-resolved photoluminescence (TR-PL) techniques to study carrier dynamics in InxGa1-xAs samples with and
without the CMP process and a high resolution TEM to study defects in various structures.
A new deep-well (DW) quantum-cascade laser (QCL) design: Tapered Active-Region (TA), for which the barrier layers
in each active region are tapered such that their conduction band edges increase in energy from the injection barrier to
the exit barrier, causes a significant increase in the energy difference between the upper laser level and the next higher
energy level, E54; thus, resulting in further carrier-leakage suppression compared to DW QCLs. High E54 values (80 -100
meV) are primarily obtained because the energy separation between the first excited states of a pair of coupled QWs
(CQWs) is larger when the CQWs are asymmetric than when they are symmetric. Then, we reach an optimized TA-QCL
design (λ= 4.7 μm) for which E54 values as high as 99 meV are obtained, while insuring good carrier depopulation of the
lower laser level (i.e., τ3 = 0.2 ps) via the double-phonon-resonance scheme. In addition, the upper-laser-level lifetime
increases by ~ 15 % compared to that for conventional QCLs. As a result, the relative carrier leakage decreases to values
≤ 1% and the room-temperature (RT) threshold-current density decreases by ~ 25 % compared to that for conventional
QCLs. Then, we estimate that single-facet, continuous-wave (CW) RT wallplug-efficiency values as high as 27 % are
possible. Preliminary results from TA QCLs include T0 and T1 values as high as 231 K and 797 K, respectively, over the
20-60 °C heatsink-temperature range.
Dilute nitride materials with a 1eV band-gap lattice matched to GaAs substrates are attractive for high-efficiency
multi-junction solar cells. Carrier lifetime measurements are crucial in optimizing material growth and p-i-n field-aided
carrier-extraction-device design. One research group has reported carrier lifetimes of MBE-grown bulk
InGaNAsSb materials, but there has been no report of carrier lifetime measurements from bulk InGaNAsSb grown
by MOVPE. In this study, we report the growth of bulk InGaNAsSb by MOVPE and the first carrier lifetime
measurement from MOVPE-grown bulk InGaNAsSb materials with Eg= 1.0 - 1.2eV at 300K. We studied carrier
dynamics in MOVPE-grown bulk dilute nitride materials nominally lattice matched to GaAs (100) substrates: 1μm
thick In0.035GaN0.025As (Eg= 1.0eV at 300K) and ~0.2μm thick In(0.05-0.07)GaN(0.01-0.02)AsSb(0.02-0.06) layers (Eg= 1.2eV
at 300K). Both structures are fully strained. The incorporation of N in InGaNAs leads to degradation in
photoluminescence efficiency, but prior studies indicate the addition of Sb in MBE-grown InGaNAsSb improved the
PL efficiency. Two-step post-growth thermal annealing processes were optimized to obtain maximum PL
efficiencies that yielded a typical blue shift of 50 and 30meV for InGaNAs and InGaNAsSb, respectively. We
employed a streak camera to measure carrier lifetimes from both as-grown and thermally annealed samples. Carrier
lifetimes of <30psec were obtained from the InGaNAs samples, whereas carrier lifetimes of up to ~150psec were
obtained from the InGaNAsSb samples. We discuss possible reasons for short carrier lifetimes measured from
MOVPE-grown InGaNAs(Sb) materials.
We employ an alternate approach to Stranski-Krastanow (SK) QD formation involving the use of nanopatterning with
diblock copolymers combined with selective MOCVD growth, enabling QD formation over large surface areas intended
for device application. This approach allows for increased control over the QD size and distribution and elimination of
the problematic wetting layer associated with SK QDs. Cross-sectional TEM studies of the nanopatterned QD active
regions confirm the absence of a wetting layer, and AFM/SEM measurements indicate high QD densities are achieved
(>6x1010 cm-2). Furthermore, the process is applicable to large surface coverage, showing promise for implementation
into long wavelength (λ = 1.3-1.5μm) sources employing either lattice-matched or strained QDs. Preliminary device
results demonstrate LT (up to 170K) InP-based laser operation from devices employing patterned lattice-matched InxGa1-
xAs QD (~ 20 nm dia.) active regions. The formation of high density compressively strained InAs QDs on InP substrates
is also demonstrated using the nanopatterning and selective growth process.
The equations for the threshold-current density Jth, differential quantum efficiency ηd and maximum wallplug efficiency
ηwp,max for quantum-cascade lasers (QCLs) have been modified for electron leakage and backfilling. We used a thermalexcitation
model of "hot" injected electrons from the upper laser state to upper active-region energy states to calculate
leakage currents. Then the calculated characteristic temperature T0 for Jth was found to agree well with experiment for
both conventional and deep-well QCLs. The characteristic temperature T1 for ηd was deduced to be due to both electron
leakage and an increase in the waveguide-loss coefficient. For conventional mid-infrared QCLs ηwp,max is found to be
strongly temperature dependent which explains experimental data. By using a new concept: tapered active-region (TA),
deep-well QCLs have been optimized for virtual suppression of the electron-leakage currents. In turn, at room
temperature, for continuous-wave (CW)-operating, 4.5-5.0 μm-emitting TA QCLs we estimate the threshold current to
decrease by ~ 25 %, the active-region temperature rise at the ηwp,max point to decrease by ~ 30 %, and the single-ended,
ηwp,max value to become at least 22 %. Preliminary results from TA QCLs include T1 values as high as 454 K, over the
20-60 oC heatsink-temperature range.
The optical spectral gain characteristics and overall radiative efficiency of MOCVD grown InGaAs quantum dot lasers
have been evaluated. Single-pass, multi-segmented amplified spontaneous emission measurements are used to obtain the
gain, absorption, and spontaneous emission spectra in real units. Integration of the calibrated spontaneous emission
spectra then allows for determining the overall radiative efficiency, which gives important insights into the role which
nonradiative recombination plays in the active region under study. We use single pass, multi-segmented edge-emitting in
which electrically isolated segments allow to vary the length of a pumped region. In this study we used 8 section devices
(the size of a segment is 50x300 μm) with only the first 5 segments used for varying the pump length. The remaining
unpumped segments and scribed back facet minimize round trip feedback. Measured gain spectra for different pump
currents allow for extraction of the peak gain vs. current density, which is fitted to a logarithmic dependence and directly
compared to conventional cavity length analysis, (CLA). The extracted spontaneous emission spectrum is calibrated and
integrated over all frequencies and modes to obtain total spontaneous radiation current density and radiative efficiency,
ηr. We find ηr values of approximately 17% at RT for 5 stack QD active regions. By contrast, high performance InGaAs
QW lasers exhibit ηr ~50% at RT.
We present the heterogeneous integration of a 3.8 μm thick InGaAs/GaAs edge emitting laser that was metal-metal
bonded to SiO2/Si and end-fire coupled into a 2.8 μm thick tapered SU8 polymer waveguide integrated on the same
substrate. The system was driven in pulsed mode and the waveguide output was captured on an IR imaging array to
characterize the mode. The waveguide output was also coupled into a multimode fiber, and into an optical head and
spectrum analyzer, indicating lasing at ~997 nm and a threshold current density of 250 A/cm2.
The integration of thin film edge emitting lasers onto silicon enables the realization of planar photonic structures for
interconnection and for miniaturized optical systems that can be integrated in their entirety at the chip scale. These thin
film emitters are compound semiconductor lasers that are optimized for operation without the growth substrate.
Removal of the laser growth substrate, coupled with bonding to the silicon host substrate, enable the integration of high
quality edge emitting lasers with silicon. This paper explores the challenges, approaches, fabrication processes, and
progress in the integration of thin film edge emitting lasers integrated onto silicon.
In this work we present the characteristics of a novel type of quantum-cascade (QC) laser: the deep-well (DW) QC
device, which, unlike conventional QC lasers, contains a superlattice of quantum wells and barriers of different
composition, respectively. The fabrication of DW-QC devices is made possible by the use of metal-organic chemical
vapor deposition (MOCVD), a crystal growth technique which allows one to easily vary the composition of wells and
barriers within QC structures, thus providing significantly increased flexibility in optimizing the device design. We have
designed such varying-composition QC structures to have deep quantum wells in and tall barriers in and around the
active region. DW- QC laser structures have fabricated into 19 μm-wide ridges and 3 mm-long chips. Threshold-current
densities as low as 1.5 kA/cm2 are obtained at room temperature in the 4.6-4.8 μm wavelength region. In conventional
QC lasers emitting in the 4.5-5.5μm range there is substantial thermionic carrier leakage from the upper laser level to the
continuum, as evidenced by a significant decrease in the slope efficiency above 250 K, which is understandable given
the relatively small (i.e., ~ 200 meV) energy differential, δE, between the upper lasing level and the top of the exit
barrier. For the DW design carrier leakage is suppressed due to deep active wells and tall barriers, such that δE reaches
values in excess of 400 meV. Preliminary results include a threshold-current characteristic temperature, T0, value of 218
K over the temperature range: 250- 340 K.
Intersubband Quantum-Box (IQB) lasers; that is, devices consisting of 2-D arrays of ministacks (i.e., 2-4 stages) intersubband QB emitters are proposed, as an alternative to 30-stage quantum-cascade (QC) devices, as sources for efficient room-temperature (RT) emission in the mid-infrared (4-6 µm) wavelength range. Preliminary results include: 1) the design of devices for operation with 50% wallplug efficiency at RT; 2) realization of a novel type of QC device: the deep-well (DW) QC laser, that has demonstrated at =4.7µm low temperature sensitivity of the threshold current, a clear indication of suppressed carrier leakage; 3) the formation of 2-D arrays at nanopoles by employing nanopatterning and dry etching; 4) the formation of 40nm-diameter, one-stage IQB structures on 100nm centers by preferential regrowth via metal-organic vapor phase epitaxy (MOVPE).
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.