K. A. Bertness

29.5%‐efficient GaInP/GaAs tandem solar cells

We report on multijunction GaInP/GaAs photovoltaic cells with efficiencies of 29.5% at 1‐sun concentration and air mass (AM) 1.5 global and 25.7% 1‐sun, AM0. These values represent the highest efficiencies achieved by any solar cell under these illumination conditions. Three key areas in this technology are identified and discussed; the grid design, front surface passivation of the top cell, and bottom surface passivation of both cells. Aspects of cell design related to its operation under concentration are also discussed.

Band‐gap narrowing in ordered Ga0.47In0.53As

We report the first observation of band‐gap energy reduction in Ga0.47In0.53As deposited on (100) InP by atmospheric pressure organometallic vapor phase epitaxy due to CuPt‐type ordering. A reduction of more than 65 meV in the band‐gap energy is observed for lattice‐matched samples that show strong CuPt‐like ordering by transmission electron microscopy. By comparison samples that show no CuPt‐like ordering diffraction signatures, do not have reduced band‐gap energies. Studies of the influence of growth parameters on the band‐gap energy indicate a U‐shaped dependence on the growth temperature with a minimum around 550 °C and decreasing band‐gap energies with increasing growth rate (at a constant V/III ratio) over the range 0.5–4 μm/h.

NEAR-FIELD OPTICAL CHARACTERIZATION OF THE PHOTOLUMINESCENCE FROM PARTIALLY ORDERED (GAIN)P

The two photoluminescence bands typically observed in partially ordered (GaIn)P are studied with a spatial resolution of 270 nm by use of scanning near‐field optical microscopy at low temperature. Local luminescence spectra show a strong spatial variation of the low energetic emission in lineshape and peak position whereas the same attributes of the high energetic emission differ only slightly for all investigated areas of the sample. The intensities of both photoluminescence bands are anticorrelated in space on a 0.5–1.5 μm scale. TEM investigations show that the structure of the sample is inhomogeneous on the same length scale.

Effect of faceting on the band gap of ordered GaInP

It has been shown that under certain growth conditions the pseudobinary semiconductor alloy GaInP shows cation site ordering into the Cu‐Pt structure, and that this ordering results in a lowering of the band gap Eg from that of the disordered alloy. The Eg lowering is known to depend on growth conditions, including the orientation of the substrate. We study the dependence of Eg on epilayer thickness for GaInP grown by metal‐organic vapor‐phase epitaxy. For epilayers grown on singular (100) substrates under growth conditions conventionally used to produce ordered material, Eg decreases dramatically with increasing epilayer thickness: Eg for a 10‐μm‐thick epilayer is ∼40 meV lower than for a 1‐μm‐thick epilayer. This dependence of Eg on thickness can be understood in terms of the recently observed faceting of the GaInP growth surface.

GaInP/GaAs monolithic tandem concentrator cells

This paper discusses design considerations for the GaInP/GaAs monolithic tandem concentrator cell. The prototype device achieves a peak efficiency of 30.2% in a range of 140-180 suns, making this the first two-terminal device to demonstrate a verified efficiency exceeding 30%. At 425 suns the efficiency is still above 29%. We focus on the issues of grid design, top-cell thickness, and antireflectance coat. We also examine ways in which these aspects of the device may be modified to provide further performance improvements for future devices.

Low‐band‐gap Ga0.5In0.5P grown on (511)B GaAs substrates

The band gap and microstructure of Ga0.5In0.5P have been shown to vary with deposition conditions. However, growth on (511)B GaAs substrates has been reported to give Ga0.5In0.5P with band gaps close to that of disordered material. It is shown here, that with appropriate selection of the growth parameters, Ga0.5In0.5P can be grown with low band gap and significant ordering on even the (511)B substrates, implying that surface steps play an important role in the ordering process. For the lattice‐matched composition, a band gap of 1.83 eV was obtained using low growth temperature (575 °C), low growth rate (0.55 μm/h), and high phosphine pressure (5 Torr).

High-efficiency GaInP/GaAs tandem solar cells for space and terrestrial applications

GaInP/GaAs tandem solar cells are proving to be a highly adaptable, high efficiency photovoltaic technology. GaInP/GaAs tandem cells have now exceeded all confirmed two-terminal efficiencies for cells of any material combination at one-sun air-mass 1.5 global (AM1.5G) with an efficiency /spl eta/=29.5%, at 160-suns AM1.5 direct (AM 1.5D) with /spl eta/=30.2%, at one-sun AMO with /spl eta/=25.7%, and at AMO with /spl eta/=19.6% after 1-MeV, 10/sup 15/-cm/sup -2/ electron irradiation. These high efficiencies reduce the balance-of-system costs for a photovoltaic power system relative to a system using less efficient cells. This balance-of-system leverage has made this technology competitive in the near term for space and concentrator terrestrial applications even through the GaInP/GaAs cells are relatively expensive. A starting point for these high efficiencies is the high quality of the single-crystal epitaxial material. Recent advances are attributed to improvements in interface passivation layers and cell designs.

High-efficiency GaInP/GaAs tandem solar cells

GaInP/GaAs tandem solar cells have achieved new record efficiencies, specifically 25.7% under air-mass 0 (AMO) illumination, 29.5% under AM 1.5 global (AM1.5G) illumination, and 30.2% at 140-180x concentration under AM 1.5 direct (AM1.5D) illumination. These values are the highest two-terminal efficiencies achieved by any solar cell under these illumination conditions. The monolithic, series-connected design of the tandem cells allows them to be substituted for silicon or gallium arsenide cells in photovoltaic panel systems with minimal design changes. The advantages of using GaInP/GaAs tandem solar cells in space and terrestrial applications are discussed primarily in terms of the reduction in balance-of-system costs that accrues when using a higher efficiency cell. The new efficiency values represent a significant improvement over previous efficiencies for this materials system, and we identify grid design, back interface passivation, and top interface passivation as the three key factors leading to this improvement. In producing the high-efficiency cells, we have addressed nondestructive diagnostics and materials growth reproducibility as well as peak cell performance.<<ETX>>

19.6% electron-irradiated GaInP/GaAs cells

Record air mass zero (AMO) efficiencies are reported for two-terminal, two-junction 0.25-cm/sup 2/ Ga/sub 0.5/In/sub 0.5/P/GaAs devices irradiated by 10/sup 15/ cm/sup -2/ 1 MeV electrons. Devices optimized for end-of-life (EOL) had beginning-of-life (BOL) and EOL efficiencies of 23% and 19.6%, respectively. A range of device structures gave EOL efficiencies greater than 18%. The design of the device for optimal radiation hardness is shown to depend on low bottom-cell base doping and optimally thin top cells, as these lead to current-matched EOL devices. Surprisingly, the damage coefficients for Ga/sub 0.5/In/sub 0.5/P are significantly larger than those previously measured and depend on the bottom-cell base doping and top-cell thickness.

Homogeneous and inhomogeneous linewidths of excitons in partially ordered Ga0.52In0.48P

We report picosecond four‐wave mixing experiments on Ga0.52In0.48P grown by organometallic vapor phase epitaxy on GaAs substrates. The spectral behavior of the homogeneous linewidth in the range of the inhomogeneously broadened band gap excitonic resonance is found to be different for a more disordered as compared to a partially ordered structure. Whereas the former shows the normal alloy behavior, the behavior of the partially ordered sample supports the assumption that its structure consists of ordered domains with varying degrees of order. This means, in particular, that the main origin of the inhomogeneous broadening is different for the disordered and ordered case. In addition, a polarization dependence of the four‐wave mixing signal is only observed for the more disordered sample.