Search

Metamorphic solar cells can have optimal bandgap combinations through the use of compositionally graded buffers, where the lattice constant is slowly varied over several microns of growth. Bragg reflectors consist of several microns of alternating layers with refractive index contrast and provide a useful internal reflection to multijunction solar cells with optically thin subcells. In this work, we implement distributed Bragg reflectors within the compositionally graded buffers of inverted metamorphic solar cells to add functionality to the buffer. The reflectance of this AlGaInAs “graded buffer Bragg reflector” is very similar to the reflectance of a similar AlGaAs Bragg reflector external to a buffer as well as the reflectance predicted by the transfer matrix model, indicating that the roughness of the buffer does not drastically reduce the reflection. Reflectance of 72%, 91%, and 98% is achieved in 2, 4, and 8 μ m buffers using AlGaInAs layers that alternate between 30% and 70% aluminum content. Using a 2 μ m graded buffer Bragg reflector, the 1.0-eV mismatched subcell of a GaAs/GaInAs tandem has a minor increase in threading dislocation density compared to a standard graded buffer and a small, 20 mV, loss in voltage. As the buffer is thickened, the voltage loss is recuperated and excellent subcell voltages are achieved, indicating that the Bragg reflector is not severely hindering dislocation glide. We demonstrate that the benefits of the graded buffer Bragg reflector for optically thin subcells and subcells containing quantum wells, and conclude that Bragg reflectors can effectively be implemented within graded buffers, adding functionality without adding cost.

The development of compositionally graded buffer layers (CGBs) with enhanced transparency would enable novel five and six junction solar cells, with efficiencies approaching 50% under high concentration. We demonstrate highly transparent grades between the GaAs and InP lattice constants on both A- and B-miscut GaAs substrates, employing Al x Ga y In1- x - y As and highly Se-doped Burstein–Moss (BM) shifted Ga x In 1- x P. Transparency to >810 and >890 nm wavelengths is demonstrated with BM-shifted Ga x In1- x P on B-miscut substrates and Al x Ga y In1- x - y As/Ga x In1- x P(Se) combined grades on A-miscut substrates, respectively. 0.74 eV GaInAs solar cells grown on these transparent CGBs exhibit ${{W}}_{{\rm{OC}}}= {\text{0.41}}\, {\text{V}}$ at ${\text{15}}\ {\text{mA/ cm}}^{2}$ , performance comparable with the state-of-the-art Ga x In1- x P grade employed in the four-junction-inverted metamorphic multijunction (IMM) cell. A ${\text{GaAs/ 0.74 eV}}$ GaInAs tandem cell was grown with a transparent BM-shifted Ga x In1- x P CGB to verify the CGB performance in a multijunction device structure. Quantum efficiency measurements indicate that the CGB is completely transparent to photons below the GaAs bandedge, validating its use in 4–6 junction IMM devices with a single-graded buffer. This tandem represents a highly efficient two-junction band gap combination, achieving 29.6% ± 1.2% efficiency under the AM1.5 global spectrum, demonstrating how the additional transparency enables new device structures.

To advance the state-of-the-art in III-V multijunction solar cells towards high concentration efficiencies approaching 50%, development of a high-quality ∼1.7 eV second junction solar cell is of key interest for integration in five or more junction devices. Quaternary GalnAsP solar cells grown lattice-matched on GaAs allows bandgap tunability in the range from 1.42 to 1.92 eV and offers an attractive Al-free alternative to conventional AlGaAs solar cells. In this work, we investigate the role of growth temperature towards understanding the optimal growth window for realizing high-quality GalnAsP alloys. We demonstrate bandgap tunability from 1.6 to 1.8 eV in GalnAsP alloys for compositions close to the miscibility gap, while still maintaining lattice-matched condition to GaAs. We perform an in-depth investigation to understand the impact of varying base thickness and doping concentration on the carrier collection and performance of these 1.7 eV GalnAsP solar cells. The photo-response of these cells is found to be very sensitive to p-type zinc dopant incorporation in the base layer. We demonstrate prototype 1.7 eV GalnAsP solar cell designs that leverage enhanced depletion width as an effective method to overcome this issue and boost long-wavelength carrier collection. Short-circuit current density (J sc ) measured in field-aided devices were as high as 17.25 m A/cm2. The best GalnAsP solar cell in this study achieved an efficiency of 17.2% with a J sc of 17 m A/cm2 and a fill-factor of 86.4%. The corresponding open-circuit voltage (V oc ) 1.7 eV measured on this cell represents the highest V oc reported for a 1.7 eV GalnAsP solar cell. These initial cell results are encouraging and highlight the potential of Al-free GalnAsP solar cells for integration in the next generation of III-V multijunction solar cells.

Dilute III-V alloys containing N or Bi share many features that are common, but some that are distinct. In GaP and GaAs, both the substituent species N and Bi behave as isoelectronic impurity traps and both lead to a giant bandgap bowing phenomenon. The isolated N and Bi impurities generate bound states in GaP but resonant states in GaAs. N impurity pairs have been observed as bound states in GaP and in GaAs whereas Bi impurity pairs have not been observed a bound states in GaP nor in GaAs. Low temperature photoluminescence studies on GaAs 1-x Bi x show undulations in the spectra but these are not associated with Bi-Bi pairs. Theoretical arguments for the differing behaviour of the N and Bi isolated impurities in GaAs as a function of pressure are provided.

Self-assembled InGaN quantum dots were grown in the Stranski-Krastanov mode by plasma-assisted molecular beam epitaxy. The average dot height, diameter and density are 3 nm, 30 nm and 7 x 1010 cm -2 , respectively. The dot density was found to decrease as the growth temperature increases. The cathodoluminescence emission peak of the InGaN/GaN multiple layer quantum dots (MQDs) was found to red shift 330 meV with respect to the emission peak of the uncapped single layer of InGaN QDs due to Quantum Confined Stark effect. Blue LEDs based on InGaN/GaN multiple quantum wells (MQWs) as well as green and red LEDs based on InGaN MQDs emitting at 440, 560 and 640 nm, respectively, were grown and fabricated. The electroluminescence peak positions of both the green and red InGaN MQD LEDs are shown to be more blue-shifted with increasing injection current than that of the blue InGaN/GaN MQW LEDs.

Multi-junction technology is rapidly advancing, which puts increasing demands on IV characterization resources. We report on a tool and procedure for fast turn-around of IV data under the reference conditions, but also under controlled variations from the reference conditions. This enhanced data set can improve further iterations of device optimization.

Initial tests are performed regarding the degradation of lattice-mismatched GaInAs solar cells. 1eV metamorphic GaInAs solar cells with 1-2×106 cm-2 threading dislocation density in the active region are irradiated with an 808 nm laser for 2 weeks time under a variety of temperature and illumination conditions. All devices show a small degradation in Voc that is logarithmic with time. The absolute loss in performance after 2 weeks illuminated at 1300 suns equivalent and 125°C is 7 mV Voc and 0.2% efficiency, showing these devices to be relatively stable. The dark current increases with time and is analyzed with a two-diode model. A GaAs control cell degrades at the same rate, suggesting that the observed degradation mechanism is not related to the additional dislocations in the GaInAs devices.

In this paper, we report the development of blue-green-red LEDs based on InGaN quantum dots (QDs) and quantum wells in the active region, and GaN QDs in the nucleation layer for dislocation filtering, by plasma assisted molecular beam epitaxy. Self-assembled InGaN QDs and GaN QDs were grown in the Stranski-Krastanov mode. For the GaN QDs grown at 770 °C, the height distribution of the dots shows a bimodal distribution, which can be attributed to the interaction of the GaN QDs with the threading dislocations. TEM and XRD studies indicate that GaN QDs in the nucleation region help threading dislocations to deviate and annihilate. The average dot height, diameter and density of the InGaN QDs were estimated to be 3 nm, 30 nm and 7×1010 cm−2, respectively. The cathodoluminescence emission peak of the InGaN/GaN multiple layer quantum dots (MQDs) was found to red shift 330 meV with respect to the emission peak of the uncapped single layer of InGaN QDs due to Quantum Confined Stark effect. Blue LEDs based on InGaN/GaN multiple quantum wells (MQWs) as well as green and red LEDs based on InGaN MQDs emitting at 440 nm, 560 nm and 640 nm with FWHM of 30 nm, 87 nm and 97 nm, respectively, were grown and fabricated. The electroluminescence peak positions of both the green and red InGaN MQD LEDs are shown to be more blue-shifted with increasing injection current than that of the blue InGaN/GaN MQW LEDs.

Much of the work on III-Nitride-based LEDs that has been published and applied commercially has been done using metal-organic chemical vapor deposition (MOCVD) as a method of film growth. We report on the growth and fabrication of visible light emitting diodes, by combining hydride vapor-phase epitaxy (HVPE) and rf plasma-assisted MBE (PAMBE) methods. Thick (∼7 μm to 10 μm) HVPE n+-GaN smooth and textured templates, were used as substrates for the growth of LED structures by rf-PAMBE. The active regions of the LED structures, which consist of InGaN/GaN MQWs, were grown using the pulsed nitrogen plasma technique leading to abrupt well and barrier interfaces as confirmed by x-ray diffraction (XRD) measurements. Using this method, we obtained InGaN/GaN MQWs whose room temperature photoluminescence (PL) spectra have a full width at half maximum (FWHM) of 12 nm (105 meV). Visible LEDs on smooth GaN templates emitting in the blue to green were produced with EL spectrum FWHM as narrow as 27 nm. On the other hand, white LEDs without the use of phosphor have been produced utilizing textured MQWs as the active region, a phenomenon we initially attribute to different incorporation of In on the different QW-planes. The growth and fabrication of these devices was preceded by detailed growth and doping studies of the various layers of the LED structure. These include detailed nucleation studies on (0001) sapphire substrates as well as identification of kinetic factors which lead to good crystalline-quality InGaN alloys and InGaN/GaN MQWs.

Transmission electron microscopy studies of GaAs1−xBix layers grown at low temperature by molecular beam epitaxy have revealed evidence of both atomic ordering and phase separation. In layers containing up to ∼10% Bi, the two variants of CuPtB-type atomic ordering on {111}B planes were observed and this is believed to be associated with the surface reconstruction present during growth. In a sample containing ∼13% Bi, no atomic ordering was observed but instead an anisotropic platelike structure was present that is believed to result from phase separation, possibly associated with the surface segregation of excess Bi during growth. Both of these effects are expected to have significant effects on the electrical and optical properties of the material.

Several benefits of in situ wafer curvature monitoring on simple structures with low misfit are discussed. The misfit of lattice-mismatched layers is measured during pseudomorphic growth, allowing for experiments that test relationships between misfit and growth conditions. As an example, Bi incorporation in GaAs is quantified by varying the substrate temperature throughout growth while using curvature measurements to continuously calculate the composition. Results agree well with x-ray diffraction measurements on individual GaAsBi samples, demonstrating the utility of this technique for the study of incorporation in mismatched systems. Once relaxation begins, the strain and dislocation energetics of low-misfit epilayers are determined from changes in wafer curvature. The authors perform several analyses on GaInAs epilayers with different misfits using an anisotropic thin film approximation. Substantial information on dislocation formation and motion is derived from the wafer curvature. Potential applications of this technique include the study of devices that utilize coherently strained layers and structures that intentionally use dislocations to relieve strain. Wafer curvature is a powerful method for comparing strain evolution in mismatched materials.

The authors report on the formation and evaluation of V-based Ohmic contacts to n-AlGaN films in the entire alloy composition. The films were produced by plasma assisted molecular beam epitaxy and doped n-type with Si. The conductivity of the films was determined to vary from 103to10−2(Ωcm)−1 as the AlN mole fraction increases from 0% to 100%. Ohmic contacts were formed by e-beam evaporation of V(15nm)∕Al(80nm)∕V(20nm)∕Au(100nm). These contacts were rapid thermal annealed in N2 for 30s at various temperatures. The optimum annealing temperature for this contact scheme to n-GaN is about 650°C and increases monotonically to about 1000°C for 95%–100% AlN mole fraction. The specific contact resistivity was found to be about 10−6Ωcm2 for all films up to 70% AlN mole fraction and then increases to 0.1–1Ωcm2 for films from 95%–100% AlN mole fraction. These results were accounted for by hypothesizing that vanadium, upon annealing, interacts with the nitride film and forms vanadium nitride, which is consistent with...

Acoustic pressures may generally be measured with much greater sensitivity, dynamic range, and frequency response than acoustic currents. Consequently, most measurements of acoustic impedance consist of comparison with standard impedances. The method reported here uses a semi‐infinite waveguide as the reference because its impedance is purely resistive, frequency independent and accurately known, independent of theories of the boundary layer. Waveguides are effectively infinite for pulses shorter than the echo return time, or if the attenuation due to wall losses (typically 80 dB) exceeds the dynamic range of the experiment. The measurement signal from a high output impedance source is calibrated to have Fourier components proportional to fn, where n may be 1 for convenience or chosen to improve the signal:noise ratio. The method has been used on diverse systems over the range 50 Hz to 13 kHz. When applied to systems with simple geometries, the technique yields results with a little higher wall losses tha...

In this paper, we report the growth and fabrication of non-polar A-plane AlGaN multiple quantum well based ultraviolet light emitting diodes (UV-LEDs). The LEDs were grown on R-plane sapphire substrates using molecular beam epitaxy (MBE). The Current-voltage characteristics of the fabricated devices demonstrated rectifying behavior with a series resistance of 38 ohms. An electro-luminescence emission at 338 nm was obtained.