GaAs solar cells grown on Si substrates for space use

GaAs single‐junction and InGaP/GaAs multi‐junction thin‐film solar cells fabricated on Si substrates have great potential for high‐efficiency, low‐cost, lightweight and large‐area space solar cells. Heteroepitaxy of GaAs thin films on Si substrates has been examined and high‐efficiency GaAs thin‐film solar cells with total‐area efficiencies of 18·3% at AM0 and 20·0% at AM 1·5 on Si substrates (GaAs‐on‐Si solar cells) have been fabricated. In addition, 1‐MeV electron irradiation damage to GaAs‐on‐Si cells has been studied. The GaAs‐on‐Si cells are found to show higher end‐of‐life efficiency than the conventional GaAs cells fabricated on GaAs substrates (GaAs‐ on‐GaAs cells) under high‐fluence 1‐MeV electron irradiation of more than 1 × 10¹⁵ cm⁻². The first space flight to make use of them has been carried out. Forty‐eight 2 × 2 cm GaAs‐on‐Si cells with an average AM0 total‐area efficiency of 16·9% have been evaluated in the Engineering Test Satellite No.6 (ETS‐VI). The GaAs‐on‐Si cells have been demonstrated to be more radiation‐resistant in space than GaAs‐on‐GaAs cells and 50, 100 and 200‐μm‐thick Si cells. These results show that the GaAs‐on‐Si single‐junction and InGaP/GaAs‐on‐Si multi‐junction cells have great potential for space applications. Copyright © 2001 John Wiley & Sons, Ltd.     

Activity and current status of R&D on space solar cells in Japan

Japan's Research and Development (R&D) activities on high‐performance III–V compound space solar cells are presented. Studies of new CuInGaSe₂ thin‐film terrestrial solar cells for space applications are also discussed. Performance and radiation characteristics of a newly developed InGaP/GaAs/Ge triple‐junction space solar cell, including radiation response, results of a flight demonstration test of InGaP/GaAs dual‐junction solar cells and CuInGaSe₂ thin‐film solar cells, and radiation response of three component sub‐cells are explained. This study confirms superior radiation tolerance of InGaP/GaAs dual‐junction cells and CuInGaSe₂ thin‐film cells by space flight experiments. Copyright © 2005 John Wiley & Sons, Ltd.     

Life‐cycle assessment of photovoltaic modules: Comparison of mc‐Si,InGaP and InGaP/mc‐Si solar modules

A higher conversion efficiency of photovoltaic modules does not automatically imply a lower environmental impact, when the life‐cycle of modules is taken into account. An environmental comparison is carried out between the production and use phase, except maintenance, of an indium–gallium–phosphide (InGaP) on multicrystalline silicon (mc‐Si) tandem module, a thin‐film InGaP cell module and a mc‐Si module. The evaluation of the InGaP systems was made for a very limited industrial production scale. Assuming a fourfold reuse of the GaAs substrates in the production of the thin‐film InGaP (half) modules, the environmental impacts of the tandem module and of the thin‐film InGaP module are estimated to be respectively 50 and 80% higher than the environmental impact of the mc‐Si module. The energy payback times of the tandem module, the thin‐film InGaP module and the mc‐Si module are estimated to be respectively 5.3, 6.3 and 3.5 years. There are several ways to improve the life‐cycle environmental performance of thin‐film InGaP cells, including improved materials efficiency in production and reuse of the GaAs wafer and higher energy efficiency of the metalorganic chemical vapour deposition process. Copyright © 2003 John Wiley & Sons, Ltd.     

Thin‐film GaAs epitaxial lift‐off solar cells for space applications

In the present work the space compatibility of thin‐film GaAs solar cells is studied. These cells are separated from their GaAs substrate by the epitaxial lift‐off (ELO) technique and mounted behind a CMG cover glass which at the same time serves as a stable carrier for the thin film cells. In the present initial stage of development these cells have an average efficiency of about 15·4% under AM0 illumination due to not yet optimized grid contacts and anti‐reflection coatings. Inspection after irradiation by 1 MeV electrons, thermal vacuum and thermal cycling experiments reveal that degradation of the cells is largely due to delamination and micro‐cracking. Based on these results, glass dehydration and adhesive degassing procedures are implemented in the ELO cell processing. As a consequence, even in this premature phase, newly produced cells show a radiation hardness comparable to or better than that of commercially available GaAs cells on Ge substrates and are virtually unaffected by severe thermal cycling. Copyright © 2005 John Wiley & Sons, Ltd.     

Fabrication and analysis of multijunction solar cells with a quantum dot (In)GaAs junction

InAs quantum dots (QDs) have been incorporated to bandgap engineer the (In)GaAs junction of (In)GaAs/Ge double‐junction solar cells and InGaP/(In)GaAs/Ge triple‐junction solar cells on 4‐in. wafers. One sun AM0 current–voltage measurement shows consistent performance across the wafer. Quantum efficiency analysis shows similar aforementioned bandgap performance of baseline and QD solar cells, whereas integrated sub‐band gap current of 10 InAs QD layers shows a gain of 0.20 mA/cm². Comparing QD double‐junction solar cells and QD triple‐junction solar cells to baseline structures shows that the (In)GaAs junction has a V_oc loss of 50 mV and the InGaP 70 mV. Transmission electron microscopy imaging does not reveal defective material and shows a buried QD density of 10¹¹ cm⁻², which is consistent with the density of QDs measured on the surface of a test structure. Although slightly lower in efficiency, the QD solar cells have uniform performance across 4‐in. wafers. Copyright © 2013 John Wiley & Sons, Ltd.     

GaAs/Ge tandem-cell space concentrator development

GaAs/Ge monolithic tandem two-junction concentrators are being developed by optimizing separate one-junction GaAs and Ge cells that simulate the GaAs top cell and Ge bottom cell of the tandem. Separation allows easier analysis of the tandem's top and bottom cells than if these two junctions were in series. The best GaAs top cell has an independently measured AM1.5D efficiency of 28.7% at 200 suns and 25°C (24.5% AM0 at 170 suns), a record for a monolithic cell without a prismatic cover. The Ge bottom cells have a GaAs optical filter (but no GaAs junction) to replicate the spectrum that the Ge cell sees when incorporated into a tandem. The best Ge-under-GaAs bottom cell efficiency is 4.6% AM0 at 103 suns. Evidence that the 900-1800-nm response seen from the Ge bottom cell is due to a p-n junction in the Ge and not a GaAs/Ge heterojunction is presented     

Metamorphic GayIn1−yP/Ga1−xInxAs tandem solar cells for space and for terrestrial concentrator applications at C > 1000 suns

The use of Ga_1−xIn_xAs instead of GaAs as a bottom solar cell in a Ga_yIn_1−yP/Ga_1−xIn_xAs tandem structure increases the flexibility of choosing the optimum bandgap combination of materials for a multijunction solar cell. Higher theoretical efficiencies are calculated and different cell concepts are suggested for space and terrestrial concentrator applications. Various Ga_yIn_1−yP/Ga_1−xIn_xAs material combinations have been investigated for the first time and efficiencies up to 24·1% (AM0) and 27·0% (AM1·5 direct) have been reached under one-sun conditions. An efficiency of 30·0–31·3% was measured for a Ga_0·35In_0·65P/Ga_0·83In_0·17As tandem concentrator cell with prismatic cover at 300 suns. The top and bottom cell layers of this structure are grown lattice-matched to each other, but a large mismatch is introduced at the interface to the GaAs substrate. This cell structure is well suited for the use in next-generation terrestrial concentrators working at high concentration ratios. For the first time a cell efficiency up to 29–30% has been measured at concentration levels up to 1300 suns. A small prototype concentrator with Fresnel lenses and four tandem solar cells working at C = 120 has been constructed, with an outdoor efficiency of 23%. Copyright © 2001 John Wiley & Sons, Ltd.     

100‐period, 1.23‐eV bandgap InGaAs/GaAsP quantum wells for high‐efficiency GaAs solar cells: toward current‐matched Ge‐based tandem cells

Major challenges for InGaAs/GaAsP multiple quantum well (MQW) solar cells include both the difficulty in designing suitable structures and, because of the strain‐balancing requirement, growing high‐quality crystals. The present paper proposes a comprehensive design principle for MQWs that overcomes the trade‐off between light absorption and carrier transport that is based, in particular, on a systematical investigation of GaAsP barrier effects on carrier dynamics that occur for various barrier widths and heights. The fundamental strategies related to structure optimization are as follows: (i) acknowledging that InGaAs wells should be thinner and deeper for a given bandgap to achieve both a higher absorption coefficient for 1e‐1hh transitions and a lower compressive strain accumulation; (ii) understanding that GaAs interlayers with thicknesses of just a few nanometers effectively extend the absorption edge without additional compressive strain and suppress lattice relaxation during growth; and (iii) understanding that GaAsP barriers should be thinner than 3 nm to facilitate tunneling transport and that their phosphorus content should be minimized while avoiding detrimental lattice relaxation. After structural optimization of 1.23‐eV bandgap quantum wells, a cell with 100‐period In_0.30GaAs(3.5 nm)/GaAs(2.7 nm)/GaAsP_0.40(3.0 nm) MQWs exhibited significantly improved performance, showing 16.2% AM 1.5 efficiency without an anti‐reflection coating, and a 70% internal quantum efficiency beyond the GaAs band edge. When compared with the GaAs control cell, the optimized cell showed an absolute enhancement in AM 1.5 efficiency, and 1.22 times higher efficiency with 38% current enhancement with an AM 1.5 cut‐off using a 665‐nm long‐pass filter, thus indicating the strong potential of MQW cells in Ge‐based 3‐J tandem devices. Copyright © 2013 John Wiley & Sons, Ltd.     

Wafer reuse for repeated growth of III–V solar cells

The epitaxial lift‐off (ELO) technique can be used to separate a III–V solar cell structure from its underlying GaAs or Ge substrate. ELO from 4‐inch Ge wafers is shown and 2‐inch GaAs wafer reuse after lift‐off is demonstrated without degradation in performance of the subsequent thin‐film GaAs solar cells that were retrieved from it. Since a basic wet chemical smoothing etch procedure appeared insufficient to remove all the surface contamination, wafer re‐preparation is done by a chemo‐mechanical polishing procedure. Copyright © 2010 John Wiley & Sons, Ltd.     

High-efficiency GaAs/Ge monolithic tandem solar cells

Tandem cells of GaAs grown by metalorganic chemical vapor deposition (MOCVD) on thin Ge to address both higher efficiency and reduced weight are discussed. GaAs/Ge monolithic tandem cells of 4-cm ² area have been produced with independently verified efficiencies up to 21.7% (AM0, one sun, 25°C, total area). Under AM1.5 global conditions, efficiencies are up to 24.3%. These are believed to be the highest one-sun efficiencies reported for GaAs/Ge cells, and the highest efficiency for a two-terminal monolithic tandem cell     

13·9%‐efficient CdTe polycrystalline thin‐film solar cells with an infrared transmission of ∼50%

To fabricate a high‐efficiency polycrystalline thin‐film tandem cell, the most critical work is to make a high‐efficiency top cell ( > 15%) with high bandgap (E_g = 1·5–1·8 eV) and high transmission (T > 70%) in the near‐infrared (NIR) wavelength region. The CdTe cell is one of the candidates for the top cell, because CdTe state‐of‐the‐art single‐junction devices with efficiencies of more than 16% are available, although its bandgap (1·48 eV) is slightly lower for a top cell in a current‐matched dual‐junction device. In this paper, we focus on the development of a: (1) thin, low‐bandgap Cu_xTe transparent back‐contact; and (2) modified CdTe device structure, including three novel materials: cadmium stannate transparent conducting oxide (TCO), ZnSnO_x buffer layer, and nanocrystalline CdS:O window layer developed at NREL, as well as the high‐quality CdTe film, to improve transmission in the NIR region while maintaining high device efficiency. We have achieved an NREL‐confirmed 13·9%‐efficient CdTe transparent solar cell with an infrared transmission of ∼50% and a CdTe/CIS polycrystalline mechanically stacked thin‐film tandem cell with an NREL‐confirmed efficiency of 15·3%. Copyright © 2005 John Wiley & Sons, Ltd.     

A thin‐film silicon solar cell and module

We have developed a new light‐trapping scheme for a thin‐film Si stacked module (Si HYBRID PULS module), where a (a‐Si:H/transparent interlayer/microcrystalline Si) thin‐film was integrated into a large‐area solar cell module. An initial aperture efficiency of 13·1% has been achieved for a 910 × 455 mm Si HYBRID PLUS module, which was independently confirmed by AIST. This is the first report of the independently confirmed efficiency of a large‐area thin‐film Si module with an interlayer. The 19% increase of short‐circuit current of this module was obtained by the introduction of a transparent interlayer that caused internal light‐trapping. A mini‐module was shown to exhibit a stabilized efficiency of 12%. Outdoor performance of a Si HYBRID (a‐Si:H / micro‐crystalline Si stacked) solar cell module has been investigated for over 4 years with two different kinds of module (top and bottom cell limited, respectively). The HYBRID modules limited by the top cell have exhibited a more efficient performance than the modules limited by the bottom cell, in natural sunlight at noon. Copyright © 2005 John Wiley & Sons, Ltd.     

Wafer bonded four‐junction GaInP/GaAs//GaInAsP/GaInAs concentrator solar cells with 44.7% efficiency

Triple‐junction solar cells from III–V compound semiconductors have thus far delivered the highest solar‐electric conversion efficiencies. Increasing the number of junctions generally offers the potential to reach even higher efficiencies, but material quality and the choice of bandgap energies turn out to be even more importance than the number of junctions. Several four‐junction solar cell architectures with optimum bandgap combination are found for lattice‐mismatched III–V semiconductors as high bandgap materials predominantly possess smaller lattice constant than low bandgap materials. Direct wafer bonding offers a new opportunity to combine such mismatched materials through a permanent, electrically conductive and optically transparent interface. In this work, a GaAs‐based top tandem solar cell structure was bonded to an InP‐based bottom tandem cell with a difference in lattice constant of 3.7%. The result is a GaInP/GaAs//GaInAsP/GaInAs four‐junction solar cell with a new record efficiency of 44.7% at 297‐times concentration of the AM1.5d (ASTM G173‐03) spectrum. This work demonstrates a successful pathway for reaching highest conversion efficiencies with III–V multi‐junction solar cells having four and in the future even more junctions. Copyright © 2014 John Wiley & Sons, Ltd.     

Life cycle assessment of thin‐film GaAs and GaInP/GaAs solar modules

This paper presents an environmental comparison based on life cycle assessment (LCA) of the production under average European circumstances and use in The Netherlands of modules based on two kinds of III–V solar cells in an early development stage: a thin‐film gallium arsenide (GaAs) cell and a thin‐film gallium‐indium phosphide/gallium arsenide (GaInP/GaAs) tandem cell. A more general comparison of these modules with the common multicrystalline silicon (multi‐Si) module is also included. The evaluation of the both III–V systems is made for a limited industrial production scale of 0·1 MW_p per year, compared to a scale of about 10 MW_p per year for the multi‐Si system. The here considered III–V cells allow for reuse of the GaAs wafers that are required for their production. The LCA indicates that the overall environmental impact of the production of the III–V modules is larger than the impact of the common multi‐Si module production; per category their scores have the same order of magnitude. For the III–V systems the metal‐organic vapour phase epitaxy (MOVPE) process is the main contributor to the primary energy consumption. The energy payback times of the thin‐film GaAs and GaInP/GaAs modules are 5·0 and 4·6 years, respectively. For the multi‐Si module an energy payback time of 4·2 years is found. The results for the III–V modules have an uncertainty up to approximately 40%. The highly comparable results for the III–V systems and the multi‐Si system indicate that from an environmental point of view there is a case for further development of both III–V systems. Copyright © 2006 John Wiley & Sons, Ltd.     

InGaP//GaAs//c‐Si 3‐junction solar cells employing spectrum‐splitting system

In this work, we practically demonstrated spectrum‐splitting approach for advances in efficiency of photovoltaic cells. Firstly, a‐Si:H//c‐Si 2‐junction configuration was designed, which exhibited 24.4% efficiency with the spectrum splitting at 620 nm. Then, we improved the top cell property by employing InGaP cells instead of the a‐Si:H, resulting in an achievement of efficiency about 28.8%. In addition, we constructed 3‐junction spectrum‐splitting system with two optical splitters, and GaAs solar cells as middle cell. This InGaP//GaAs//c‐Si architecture was found to deliver 30.9% conversion efficiency. Our splitting system includes convex lenses for light concentration about 10 suns, which provided concentrated efficiency exceeding 33.0%. These results suggest that our demonstration of 3‐junction spectrum‐splitting approach can be a promising candidate for highly efficient photovoltaic technologies. Copyright © 2016 John Wiley & Sons, Ltd.     

Properties of ZnO/CdS/CuInSe2 solar cells with improved performance

We report the growth and characterization of low‐bandgap record‐efficiency ZnO/CdS/CuInSe₂ thin‐film solar cells. The total area conversion efficiency for this cell is 14·5%. This result has been measured and confirmed at the National Renewable Energy Laboratory under standard reporting conditions (1000 W/m², 25°C, AM1·5 Global). The improved performance of the CuInSe₂ solar cell is primarily due to a high current density. Material and device characterization data are presented.. Published in 2004 by John Wiley & Sons, Ltd.     

Designing of 1 eV GaNAs/GaInAs superlattice subcell in current-matched four-junction solar cell

A reasonably-thick GaNAs/GaInAs superlattice could be an option as a roughly 1 eV subcell to achieve high-efficiency multi-junction solar cells on a lattice-matched Ge substrate. A detailed consideration of a high-efficiency design for a GaInP/GaAs/1 eV/Ge device is presented. Calculations have been done for this structure to obtain the confined energies of the electrons and holes by utilizing the Kronig-Penney model, as well as the absorption coefficient and thereby the external quantum efficiency. The effect of well layers, GaNAs or GaInAs, on the absorption and photocurrent density under the AM 1.5 condition is discussed in order to realize a requirement of current matching in the four-junction solar cells. The management of these considerations implies the feasibility of the GaNAs/GaInAs superlattice subcell design to improve the overall conversion efficiency of lattice matched GaInP/GaAs/1 eV/Ge cells.     

An Al x In1−x As/GaAs heterojunction ultra-thin film solar cell with 20% efficiency

An ultra-thin film photovoltaic cell, which incorporates an Al _x In_1?x As/GaAs heterojunction, is simulated using Adept 1D simulation tool, and it is with an energy conversion efficiency of 20.06% (under 1 sun, AM1.5G illumination) for 604 nm cell thickness (excluding the substrate thickness), and optimized layer thickness and doping concentration for each layer of the device. The device has an n-type AlAs window layer (highly doped), an n-type Al _x In_1?x As emitter layer and a p-type GaAs base layer. Germanium (Ge) substrate is used for the structure. The device parameters are optimized separately for each layer. Based on these optimizations, the ultra-thin film solar cell design is proposed after careful consideration of lattice mismatch between two adjacent layers of the device.     

Influence of double- and triple-layer antireflection coatings on the formation of photocurrents in multijunction III–V solar cells

The results of simulation by the transfer-matrix method of TiO₂/SiO₂ double-layer and TiO₂/Si₃N₄/SiO₂ triple-layer antireflection coatings for multijunction InGaP/GaAs/Ge heterostructure solar cells are presented. The TiO₂/SiO₂ double-layer antireflection coating is experimentally developed and optimized. The experimental spectral dependences of the external quantum yield of the InGaP/GaAs/Ge heterostructure solar cell and optical characteristics of antireflection coatings, obtained in the simulation, are used to determine the photogenerated current densities of each subcell in the InGaP/GaAs/Ge solar cell under AM1.5D irradiation conditions (1000 W/m²) and for the case of zero reflection loss. It is shown in the simulation that the optimized TiO₂/Si₃N₄/SiO₂ triple-layer antireflection coating provides a 2.3 mA/cm² gain in the photocurrent density for the Ge subcell under AM1.5D conditions in comparison with the TiO₂/SiO₂ double-layer antireflection coating under consideration. This thereby provides an increase in the fill factor of the current–voltage curve and in the output electric power of the multijunction solar cell.     

Evaluation of subcell power conversion efficiencies of radiation‐damaged triple‐junction solar cells using photoluminescence decays

We characterize the radiation‐induced damage of InGaP/GaAs/Ge solar cells for various proton irradiation energies and fluences using conventional current‐voltage (I‐V) measurements, external quantum efficiency, and a noncontact time‐resolved photoluminescence (PL) technique. From the I‐V curves, we obtain the conversion efficiency of the entire device. The external quantum efficiency showed that the short‐circuit current is only determined by the top InGaP subcell. To obtain accurate information about the point of maximum power, a new PL technique is introduced. The PL time decays of the InGaP and GaAs subcells are measured to obtain the characteristic decay time constants of carrier separation and recombination. We empirically verify that the time‐resolved PL method can be used to predict the electrical conversion efficiency of the subcells. We find that the limiting subcell at the point of maximum power is different from that for short‐circuit current. Radiation damage in unexpected regions of the device is revealed using this optical method.